I’ve seen hell and it’s featuring Dante from the Devil May Cry™ series.
I’ve been playing Persona games since like 15 years ago, and was always acutely aware of the Shin Megami Tensei series, but I was never actually tempted to pick up any of ’em. But I recently got Shin Megami Tensei III Nocturne HD Remaster (presumably named that because visually it’d look right at home on a 720p TV) as a Christmas present, so I figured I’d finally tackle that challenge. And now that I’ve beaten it, I don’t think I would recommend it to others.
I didn’t exactly hate the story, but it was kind of hard to care about the story or the characters. The plot feels like you’re just getting trashed around the place at random, like you were being carried away by waves. Your friends don’t really even feel like actual characters but rather walking plot devices, and their characters are so weak that the game even leaves naming them up to you. I guess you can more accurately rename Isamu to Buttface McDouche thanks to this weird feature, but I have no idea why it exists. Really the only characters I liked were the Manikins and the demons. To be fair, most of the characters are either Manikins or demons. They’re just not as important as the somewhat-named humans are.
The dungeons are mostly an exercise in frustration. It’s not really enough that there’s a demon trying to kill you every five-to-ten steps, but they need to have annoying puzzle mechanics too. And every time that I thought that “That dungeon sure sucked, glad to be moving on”, there was a worse one yet to come, with the final dungeon obviously being the worst one of all. ATLUS must’ve been taking kickbacks from walkthrough book publishers back in 2003 as many of the dungeons require you to navigate a specific path in a maze using a skill called “guessing”. The dungeons are not really that interesting visually either, so they don’t really have even that going for them.
Demon mechanics are awful too. Talking a demon into joining your party is just RNG with the added benefit of having to give them money and items before the coin toss. Demon fusing half-forcing you to wait for the right phase of the moon is such an annoying mechanic, since it only forces you to run around in a circle for a minute until you get the right phase. And then you get a fusion accident, forcing you to reload your save and run around again. Speaking from experience. Also, I don’t remember ever having anything good come out of a demon wanting to change its skill. At the end of the game, I instantly rejected any skill change attempts, since it was always trading a good skill for something utterly pointless. Every single one of my demons was very disappointed by this.
The difficulty curve is a massive rollercoaster. Matador wiped the floor with my face, and then the next half a dozen bosses are much easier. The antepenultimate boss was a 40-minute desperate struggle, but the final boss just laid there and allowed me to twist its glowing nipples. And if the computer has had enough of you, any random demon you encounter can just cast Dragon Eye and focus + critical hit you until the light in your demon eyes goes dark. I definitely had several moments where I almost entirely dropped the game (hello, Trumpeter, you miserable git), but was somehow able to complete the game through sheer stubbornness (and strategies posted online).
I don’t think even ATLUS thinks you should play Nocturne in this day and age, because they sure as hell didn’t put much effort into it. The game is hard-locked to 30 frames per second (unless you fix it via mods) and has audio recorded straight from someone’s old MP3 player (unless you fix that via mods too). Persona 4 Golden was released with 120 FPS support and working audio a year prior, so screwing things up this bad must surely be on purpose. The abysmal audio quality in particular is such a shame since the music is one of the singular highlights of the game. On Steam Deck the game also weirdly lacks camera control via the right analogue stick until you go back into the operating system menu and back. Not exactly a big deal, but still some jank on top of the other jank.
It’s not a completely appalling experience, and I don’t exactly regret having played through Nocturne, but I’d struggle to recommend it to anyone else. The gameplay is just so very dated at this point and what’s on offer just isn’t that good from a gameplay or technical perspective. Maybe if you really enjoy early 2000s game design and are masochistic, with very little care for audiovisual quality. If that’s not you, it’s mostly a badly remastered slog.
Also, I lied — the game only features Dante from the Devil May Cry™ series if you buy the DLC. That’ll be 10 euro please. Capcom needs the cash.
This has been a weird year for me, with the biggest thing being that I suddenly found myself without a job. This was a bit of a new experience for me, since I hadn’t been unemployed ever since I graduated. Had to fill all kinds of government paperwork about it. Not exactly the happiest of circumstances to find yourself in, but I came into it from a much better starting position than many of my fellow layabouts, so I’ll refrain from crying about it too much.
Now I’m back in the workforce again, still making computers do things for us. Those things are just a bit different, and I’m doing them in a different environment, but I didn’t suddenly become a bricklayer or anything. I’m still even a remote employee, so I didn’t even have to go into the office like many of my peers did. The biggest difference though? I am now in a company where my coworkers, or at least the most vocal of them, have taken the AI pill – hard. And I really don’t care for it.
I’m a proponent of the idea that you get better at things by doing them. That’s why I have a goal of playing my guitar for at least 15 minutes every day, no matter what. I’m currently ~1700 days into this experiment, so that should have given me 425+ hours to improve my playing. According to Malcolm Gladwell, that’s over 4% of mastery. It’s also part of why I started a blog, since even if nobody were to read anything that I wrote here, I’d at least give myself a chance to improve my writing. Not that I write here that often, but I feel like it has still helped.
This is one of the reasons why I’m skeptical of "agentic coding": I don’t think I will become a better developer by letting an AI model write code for me. And I’m almost certain that I won’t become a better developer by doing it like some of my coworkers are doing it. Some of them do not seem to do any kind of validation on the code beyond the fact that the unit tests (which the AI wrote) pass, the integration tests (which the AI wrote) work, and that the documentation (which the AI wrote) is accurate (as evaluated by the AI).
Supposedly the easiest way of catching students using AI to cheat on their homework is simply by asking them questions about their work. The chances are that they will not be able to actually tell you what is inside whatever it is that they’ve just turned in. I can’t vouch for the veracity of this method, but I have noticed that the people who are using agents to code don’t really seem to know what’s inside the commits they’re shipping out. Presumably there are teams where AI is used in a responsible manner and people make sure to understand what they are shipping, but at least where I work, velocity seems to come before reason or understanding.
I’ve seen the LinkedIn post of the tech executive praising their AI-native chief vibist, who is pushing out 250,000+ lines of code every month. That comes out to around 12,000 lines of code generated each working day. Obviously I can’t write 12,000 lines of code in a single day, but I don’t think I can review 12,000 lines of code in a day either. Is the price of our glorious AI-generated future that we just don’t look at the code that we ship?
It really boggles my mind how some people can put so much trust in these language models. I don’t even trust my own code this much; I always go through my own code again before I submit it to be reviewed by another person. Either I’m the odd one here for double-checking my work, or LLMs have blinded developers into thinking they’re near-infallible 10x coders and reviewing their changes serves only to unnecessarily slow down your business. But from my experience working with a codebase that has a significant amount of AI-generated code in it, I think mistrust is the right call.
The general quality of our AI-generated code is quite poor. Claude does very little in terms of adhering to DRY, it’s writing CRUD tests against completely barren databases, and if a test doesn’t pass, it might solve the issue not by changing the core business logic, but rather by changing the test until it passes. And I can't even really get into the truly important specifics, like for example what Claude is doing with access control. We probably wouldn't pass a security audit even if the audit was done by Claude itself.
It feels like every day I discover a new problem that can be described as "developer prompted the ticket, didn’t check it, didn’t run it". There’s only so much code that I’ve had time to look at, or run, so I don’t even know how far the rot has spread. Most of these are in a greenfield project that has yet to land on the monitors of end-users, so the customer experience thankfully hasn’t been impacted. And most importantly, I have never had to interact with a customer, so even if they were interacting with all of that vibe, they couldn’t associate my face with it. If I had to meet with the clients, I'd prefer to hide my face with a paper bag or something.
Unfortunately these terrors are often committed by the same AI evangelists that are always telling us to "not trust the AI blindly" in our many agentic coding meetings, with Claude's --dangerously-skip-permissions always enabled. Why have I even had to attend multiple of these meetings anyway? Surely vibe coding hasn’t changed that much in a month that I needed a refresher course.
The non-coding parts are also not that great in my opinion. We’re churning out so much AI-generated documentation that there’s zero chance that anyone with a beating heart will ever read them. Probably why the text reads like it’s written for other robots. I think the only place where most of the text is written by humans at this point is Slack, and it also has a bunch of AI transcriptions too.
Jira tickets are also hallucinated by LLMs. You know what’s a right laugh? Getting assigned a third-party integration ticket where all of the details on the external service have been made up by a GPU. Someone just had an AI fill out a whole document out of a single-line "Allow fetching data from [CRM service]". The only usable part was the name of the service. None of the use cases or API endpoints existed. It's fairy dust; it doesn't exist, it's not fucking real.
I'm not even that opposed to maintaining documentation, but I did try to use Claude to update the CHANGELOG.md file on my behalf. Unfortunately it managed to generate excruciatingly detailed technical descriptions of all of the changes, so I just deleted all of it, and rewrote it by hand. Hopefully this way the next developer can actually read through the changelog and not ask their AI agent to summarise the changes for them.
I’ve also noticed that trying to ask Claude about our codebases is also not that great. While debugging a problem that Claude had sidestepped by rigging the tests, I asked Claude to identify if we were validating the data we’re entering into our database. Claude then told me that we don’t have database constraints to validate anything, but we’re checking them in the code before it hits the database. I then looked and there was a function to validate the data, but absolutely nothing called it. And when I unrigged the tests, broken and unvalidated data was entered into the database. Shocking. Thankfully I had already determined the root cause before asking Claude, so I wasn’t led astray and only wasted some time looking up that unused code.
The single best use case for Claude that I've seen so far is code reviews. Before, I'd first review my changes myself, then I'd push it to remote and ask someone else to review it for me. Now I can first review it myself, then ask Claude what it thinks of my changes, and only then waste another human's time on it. The reason why I like it is that it has good upsides and minor downsides; if it notices something important, great, I can change it, and if it gets an absolutely abhorrent idea for a change, I can just ignore it.
And since LLMs can understand natural language and code, I can ask one to check if my documentation is technically accurate, which I can then fix myself. As much as I'm shitting on AI, I do actually find this useful. Although not useful enough that I'd actually pay Anthropic for a subscription, and I presume the current prices are heavily subsidized.
As far as the coding tasks that I entrust Claude, I'm currently keeping it for advanced copy-paste tasks only. By this, I mean that I might write five unit tests for a single model by hand, and then ask it to write tests for another model based on those previous handwritten tests. The way I'd have normally gone about it would've been to copy-paste the original tests anyway, so Claude can do it a bit smarter. Unfortunately, even when I've given it an explicit template for what it needs to write, it still has managed to conjure other test cases from outside my specific example, so even this use case is far from perfect.
I think code reviews also become quite interesting in terms of development team dynamics. I’ve seen pull requests co-authored by Claude with thousands upon thousands of lines of changes, and I don’t know if I can bring myself to review them. How much of my day am I going to devote to reviewing this change? And will I end up spending more time looking at this code than the guy who supposedly co-authored it with Claude?
I mean, I’ve definitely had to submit large changesets for review too, but at least I was not asking other people to work harder on reviewing the code than I worked writing it, since those large pull requests took real effort. And I definitely couldn’t open several of them per week.
So far I’ve been flying under the radar, trying to only snipe the smaller code reviews, and leaving the ones with tens of thousands of lines to other developers. I did start reading one larger change too, but someone approved it before I could really get going. Judging by how fast my coworkers get them approved, I can only assume that they’re checking the changes as rigorously as they are checking their own Claude’s work.
Unfortunately, I think AI has come to stay when it comes to software development. No idea what kind of a place the industry has for me in the future. The Internet keeps telling me that I am one of the ones who will be replaced. But I just can't imagine getting up in the morning and have the motivation to just prompt a model all day long. Personally, I’ll try to keep writing code by hand as long as I can, just like I write documentation and blog posts by hand, and enjoy the process and struggle.
I imagine my time at my current place of work is going to be relatively short. My tenure is measured in months, and I'm already pretty sick and tired of the culture. My feelings towards work are so bad that I'd probably sleep better if my employment was terminated tomorrow. Presumably for shipping code too slowly or something, although I don't think anyone has really made a note of how organic my code still is. The only thing that keeps me going is receiving a steady salary, and I've still considered just outright quitting before finding my next job.
The lesson learned from the current job? Figure out how any prospective employer is using AI before continuing, as nobody brought it up before my hiring, and I only discovered having joined a slophouse when I was already in it, and got told that I am expected to code using AI. Probably would've been better for both parties if they'd just told me ahead of time.
I absolutely adore the writing and the story in this game. It’s so masterfully done, both gritty and hilarious, and really pulls you into the story and the setting. This also applies to the characters, who all have excellent characterisation and dialogue, including your man Cole Phelps himself. Anyone who is a fan of detective novels, noir, or just great dialogue should definitely pick up L.A. Noire. I think the biggest negative thing I can say about any of it is that you can only experience it for the first time once.
The gameplay’s good too. Given the fact that this is an open-world game from Rockstar Games, one might assume that it’s basically a police version of Grand Theft Auto, but it really isn’t. You go around finding clues, interviewing witnesses and suspects, doing police legwork. Sure, some mechanics like tailing cars or engaging in firefights is similar to many other games, but it’s mostly a very unique, and detective-like gameplay experience. And really, even if it was just a lazier Grand Theft Auto clone, the story would still be enough to carry it.
I think the biggest criticism I’d have for the gameplay is that some of the interrogations require some pretty out-of-the-box thinking and leaps of logic. If someone completes this game with all of the right answers without looking them up, I’d have some concerns about their thought processes. Thankfully most of the interrogation questions are sane, and even the ones where Cole Phelps goes off on a limb do at least have some perverted logic behind it.
The facial capture technology in L.A. Noire is also quite fascinating. It really gives the characters a lot of emotion and range, a lot more than you’d expect out of a game from 2011. Quite impressively I was able to recognise some of the actors from their faces alone, so that should say something. Unfortunately the texture quality in the game isn’t the highest, so it does water down the magic of the technology a bit. I also don’t think Team Bondi really had to go so hard when it came to creating the facial capture technology, since a lot of the facial clues that you need to read aren’t that subtle. Thankfully too, since a lot of us are not trained detectives. However, hard they went and we got this very unique gaming experience that borders on a TV show as a result.
Apart from the super-unique facial scanning technology, graphics in the game are quite average. It’s not ugly, but shows its age. Unfortunately one of the bigger issues I have with the game is on the graphics side: the game is capped at a hard 30 frames per second. It also makes the game feel so sluggish that I took a long break in middle of my playthrough because it was so demotivating in an otherwise interesting game. Thankfully now you can play the game with the excellent V-Patch and enjoy a fairly problem-free 60 FPS, or even >100 FPS if you don’t mind having to turn it down for at least one of the cases. Personally, I stuck to 60 frames per second, as it seems to have no issues apart from more twitchy driving controls, and it fixed the game feeling so sluggish to play.
But even if the graphics aren’t that spectacular, you can really tell that a lot of effort was spent on the game. While I can’t speak for what Los Angeles was in 1947 (or even what it’s like even now), it does really feel like driving through a meticulous period piece. Team Bondi also really took the extra mile when it comes to the cars too. Despite featuring a lot of driving in an open world, L.A. Noire isn’t really a proper sandbox or driving game, and yet they saw it necessary to fit almost a hundred real-life cars from the period. I doubt that the illusion would’ve been broken if they had half the amount, especially given how often you’re using your police-issued vehicle, but they went hard here too.
The music and the sounds of the game are top notch too, and like the setting and cars, builds your immersion for a 1940s Los Angeles. I just had to go and listen the soundtrack after I ended my 48-hour completionist playthrough. And since the game is full of acting talent giving their motion and facial performances, the voice acting is also pretty much as good as it gets. Without going into spoilers, there is a particular scene where Aaron Staton gives off an emotional performance as Cole Phelps that I still think about.
L.A. Noire is such a special game and unfortunately we might not see another game quite like it ever again. The development of the game was tumultuous, Team Bondi folded after the development ended, and I have no idea if the motion capture technology they spent all of that time and money creating even exists anymore. And despite all of that, this wonderful, one-of-a-kind game got made. I would highly recommend everyone pick it up and experience it for themselves.
The story of Control is weird. You go into an empty government building to look for your little brother, meet a strange Finnish man, and get appointed as the director of the bureau within minutes. And unfortunately, it never gets any better. One of the characters you meet at the beginning of the game succinctly describes the story: "This is fucking unbelievable."
I feel like the story is trying way too hard to be weird and mysterious, to a point where I just completely lost interest. There's just a limit to weird cutscenes with strange double talk happening that I can process before I just zone off, and after that, there’s no amount of extreme close-ups of Jesse’s eyes that can make me start to care about her. I'd love a nice mystery story, but this isn't a mystery. It's just bizarre for the sake of being bizarre with zero intrigue except for "when does this start making sense?" But it really never does and has very little narrative closure.
Really, the only good thing going for the story is that there isn’t much of it. It took me eight and a half hours to beat the game from start to finish, and I wasn’t even playing particularly efficiently. Hell, considering I stopped playing the game for two years out of sheer disinterest, I even had time to forget how to play. But despite its short duration, it still feels padded out. The chapter Finnish Tango doesn’t seem to really have any plot significance, and just highlights how Remedy is a Finnish company and gives Martti Suosalo some work.
The combat starts off super disappointing when you're given the pistol. It's a total peashooter and feels super unsatisfying to use. Enemies will take at least three shots to the head with it before they even consider dying. I actually found it easier to rapid-fire enemies to the body than bother wasting your time aiming for the head. Your player character is also squishy, so being quick on the trigger was less likely to result in your untimely death. I'm surprised that a character in the story managed to commit suicide using this gun considering how little damage a shot to the head from it will do.
Thankfully the combat does actually open up considerably like 30 minutes afterwards when you're given your first magic power. I'm surprised that the developers decided to not give it to the players immediately considering how bad the firearm is. Or at the very least, do very minimal combat with the firearm before you're given the ability to magically hurl objects very fast at enemies. That being said, even with the magical powers, the combat is still not really amazing. It just stops being awful and starts being playable.
There’s a new collectible to grab like every 10 metres, but like the story and the characters, I’m struggling to find a reason to care about any of them. There’s also some form of crafting and side missions, and I’m not particularly interested in a game’s side hustles if the main parts are uninteresting. Apparently there’s also more magic powers than just throwing and levitation, but since those are not on the critical path, they’re surely not that important either.
I would otherwise consider the graphics to be a rare good thing about Control, but unfortunately the graphics are extremely buggy so I can’t even say that. Some enemies have the magic power of spawning pitch-black rectangles all over the screen, which makes the game very hard to play. I’m playing on an RTX 4090, so at least the issue isn’t a lack of horsepower. Thankfully I wasn’t wasteful enough to get the RTX 5090 when it came out, since apparently the game is even buggier on the 50-series cards than it is on my 40-series.
Not that the game needed to even really shine that much graphically given how bland the game is stylistically. Granted, it’s set inside a government building so blandness is on brand, but it doesn’t make it any more appealing to the eyes. You will probably not lose any enjoyment even if you can’t see the office fluorescent tube not shine off the marble floor.
At least the music’s good. I’d definitely listen to Take Control again. But that’s about the only thing Control that I want to take with me. At least I don’t have to feel miffed about not having access to Alan Wake II on Steam.
It’s frankly amazing how little effort PQube has put into this release while asking 49.99€ for it. The Steam version is shipped broken. As soon as you launch the game from the launcher, Steam will mark Punch Line as not running and thus will not update your gameplay time, won’t show you the Steam Overlay, and most importantly, will not allow you to use Steam Input. Seeing as I was trying to play this with my DualSense controller, not having my controller working at all was quite the issue.
Thankfully there is a workaround for this: open the game installation directory, remove the MAGESgamelauncher.exe executable, make a copy of the Game.exe executable and rename that copy to MAGESgamelauncher.exe. Congratulations, you can now use the Steam overlay, Steam Input and you will be correctly marked as playing Punch Line on your profile.
I’m guessing that this issue stems from the launcher launching the game executable as a completely separate process instead of a child process, so Steam never hooks the overlay/Steam Input into the Game.exe process and thus breaks everything. I’m also guessing that PQube has absolutely no quality assurance process since this should be the thing you notice immediately. Punch Line also used to have achievements on Steam – except those achievements never worked, so they “fixed” the issue by just removing them entirely. And they have the gall to ask 49.99€ for this release.
Once you’ve managed to actually get the game working, what you have is not that great. Ostensibly Punch Line is a visual novel with puzzle elements, but the puzzle elements in the game are so poor, annoying and unrewarding that the game would be better as a straight visual novel. There’s not really any way to figure out the puzzles, so you’ll just end up taking wild guesses and then doing a reset if you guessed wrong. There isn’t even any kind of fun interactions for non-progressing choices like there is for example in the Somnium segments in Uchikoshi’s AI – The Somnium Files games. The game is also around 12 hours long, and the annoying puzzle segments stop appearing around eight hours in – really for the game’s benefit.
Besides the obvious launcher issue, there are a bunch of smaller issues that pop up while playing through the game. While it never prevented me from finishing the game, the default controller layout seems to be broken (at least on a DualSense controller), with buttons that should do something not doing it. The background audio sometimes eclipses dialogue audio, even with the default volume settings. And perhaps the most annoying issue: unskippable anime-style opening and ending themes that you will have to watch about twenty times each. It’s possible that these smaller issues could be fixed by tweaking the settings, but I don’t want to spend all of my time fixing this game on PQube’s behalf.
Visually the game looks about okay – the character visuals are nice, but the environmental details leave room for improvement and the video segments, despite this having a tie-in anime, are kinda bad. I think they’ve had to retime the animations for the dialogue and it ends up looking more awkward than I would’ve imagined. At least during the final hour of the game, which is almost entirely video, the quality is at least decent. Thankfully there are no issues to point out on the music and voice acting side. Well, beyond the volume mixing issue noted earlier that is.
Really the only highlight of Punch Line is the story. It’s definitely not Uchikoshi’s finest work, but there’s a fairly interesting story with likeable characters. A lot more lighthearted than a lot Uchikoshi works too, considering how much of it revolves around panties. The writing also does a good job of resolving all of the mysteries that it sets up, so you won’t be left with nagging questions at the end.
I actually really liked the Punch Line anime when I watched it over a decade ago, and it made me want to play through the game. But since it took several years for the game to actually come out in the west, I kinda forgot about the whole thing and only got to it now that the anime’s already over ten years old. Perhaps that was for the best, since I’d already forgotten a lot of the details about it, and most of the story is the exact same as in the anime. Really what the game has going for it over the anime is that there’s a different (and better) ending, and the story is more expanded, as the game has about double the runtime.
But alas, it’s really hard to recommend going through the story through the game rather than by watching the anime. Punch Line is a good story hidden in a bad game. There’s just way too many issues with this game and they will never ever be resolved. If you absolutely insist on playing the game instead of watching the anime, buy this on a deep discount. PQube simply does not deserve the full price of admission for this mess.
When I started playing through Megadimension Neptunia VII, I thought that this was Neptunia finally reaching a level of effort and polish never before seen in the franchise. Character sprites were improved, there were levels that weren’t just recycled from the previous games, enemies were voiced and so on. Unfortunately, despite the seeming jump in quality, I started noticing a problem: the game wasn’t fun to play.
You know how in RPG games the game decides to take away your party members for plot reasons, leaving your party understaffed and making those characters underleveled when they rejoin your party? And how that is annoying as all hell? Well, Compile Heart decided that this annoying mechanic was worth making into an entire game. It took like 17 hours for my party to be at full fighting capacity, and they don’t even stick around after that. You really don’t want to get attached to any single party member because the chances are that they’re going to be missing for most of the story.
It also feels like they’ve completely forgotten about basic game design that previous titles in the Neptunia franchise understood. Things like having save points before a boss fight. Oh, you thought to challenge the boss after only 25 minutes of grinding? Time to lose your 25 minutes worth of progression. There’s also a point where you can get your character’s health too low for a boss fight, with no possibility to heal. Hopefully you didn’t overwrite your previous save, since you’ll have to reload at that point too. Compile Heart really has no considerations towards the value of your time in this game.
The overworld traversal is also a great example of just wasting your time. Movement between dungeons, cities and other points of interest involve moving on a gameboard-like map where every single space has a chance to randomly pull you into combat. Trying to get from one overworld plot event to another might require you to fight three random groups of enemies. And as you progress through the game, your party level far outstrips the enemies, requiring you to just waste time beating small fries for no good reason. Even if you don’t have to fight a single mob, the gameboard-like movement is just slow to watch. There is a fast-forward button but what the game needs is a skip button. And to further add insult to injury, you’re also expected to spend credits to build the paths to dungeons and other cities. Could I just pay a bit extra and build a path that isn’t infested with low-level mobs?
I think Compile Heart even realises how awful many of these things are since they allow disabling a lot of them once you reach New Game+. Personally, I’d design my game in such a way that you don’t want to disable a dozen "features" after a single playthrough.
VII might also have the worst achievements in franchise history. Card Master is an awful waste of time that will give you carpal tunnel, headache and a disdain for life itself. Treasure Hunter, while not reaching similar peak awfulness, is also a giant time sink without really any fun aspects to it. If you decide to challenge these, hopefully you have a second monitor to watch YouTube videos on as you painfully grind through them all. Completionists beware.
While there’s some good and interesting changes to the combat, like having a combo system that discourages just spamming the same attack again and again, combat feels worse than for example in the Rebirth titles. The timing for symbol attacks, trying to align your attacks to hit multiple enemies, having to chase after enemies dispersing all over the place. Everything just feels that much worse. Then there’s things like boss fights where the boss is immune to damage until you hit him enough times to break his damage immunity item – a process that takes way too long to be remotely fun. It’s not even a hard fight since the boss only does chip damage, meaning that at no point are you actually in danger.
As for the story, it’s just okay. Definitely not as good as Victory’s but it’s a serviceable Neptunia story, at least towards the end. I’m not really a fan of how the game is split into three different parts. It doesn’t really even add much to the story since it’s all continuous anyway. The conditions to reach the true ending are also annoying to fulfil and probably better saved for the New Game+ when the game stops doing its most annoying features.
The technical side of VII is nothing short of a train wreck. The sound settings are awful, you can’t have controller and keyboard input working at the same time, the game is capped at 60 FPS and 1920×1080 resolution and the battle load times are bad considering I have an SSD and the game isn’t that graphically intensive. I also had issues with the controller button mapping with my DualSense but maybe that was the fault of Steam Input. Hasn’t happened in other games though. And to put a cherry on top of everything, using my KVM switch (which unplugs the keyboard and mouse to use them with another PC) makes the game crash.
Thankfully there is a mod that allows you to run the game at such arbitrary resolutions as 2560×1440, and mercifully above 60 FPS. While my PC is quite powerful, I saw no negative effects from running this mod and would recommend it to anyone playing on hardware more advanced than a Steam Deck. And since the mod works so wonderfully, it really makes me wonder why Compile Heart couldn’t implement these features themselves. Although I think I know the answer: Compile Heart is a two-bit kusoge developer that wants PC gamer money for the least possible effort.
What a shame that the sequel to the best Neptunia game, Victory, and the only mainline Neptunia game we’ve gotten in the last decade, is such an awful game. And this is by Neptunia standards, which have historically not been that high anyway. I don't know why they decided that VII stands for "Victory II" with a quality gap like this. Usually with these things you can at least recommend the game to fans of the series, but I don’t think you can really even recommend it to them. The game just isn’t fun to play and the story’s not interesting enough to justify it. If you insist on having to experience the game for yourself, at least do yourself a favour and grab the game on a sale. Maybe that way you’ll feel less bad about playing it.
Everyone knows that Uchikoshi loves nothing more than non-linear stories, and Zero Time Dilemma might be him taking it to its ultimate conclusion. It feels like Uchikoshi wrote the final part of the Zero Escape series, fed most of it through a woodchipper and then made you scavenge the plot in a random order from the remains. The game lasts for about 20 hours and you’ll spend at least the first ten hours absolutely clueless as to what is happening – and when.
Even though the previous Zero Escape games are definitely non-linear stories too, at least you had a sense of belonging in the story. Most of the game feels like such a mess that I was worried if they could even to sort everything out in a manner that makes sense. Thankfully the story starts to mesh together in the end, at least to a certain point. But before that meshing starts to happen, you really need to cling onto your love for Zero Escape to get through the drudgery and confusion. I was sure that I’d recommend against this game during the first half.
The drudgery is not helped at all by the graphics. Technically speaking, the level of the 3D graphics has taken a significant leap since Virtue’s Last Reward. And yet, despite the objective advancements, the end result is arguably worse. Animation quality is laughably poor, especially the lip syncing, which amounts to random lip movements completely divorced from the spoken lines. It’s even worse than the sockpuppet-like mouth flapping of Virtue’s Last Reward, since at least there was some connection between the audio and the animation. The low-poly VLR models doing a Sesame Street impression didn’t look nearly as awkward as the more realistic models of Zero Time Dilemma. I think there was also an instance where a character was supposed to look shocked or mortified during a life-and-death scenario, and it ended up looking like he was visibly bored.
And even though the 3D quality is higher than Virtue’s Last Reward, it’s still not that stellar. However janky The Somnium Files games might look, they’re still doing circles around Zero Time Dilemma. I would’ve liked to see a lot more work towards the models and the textures. Presumably the developers either had limited skillsets for 3D work, or they simply didn’t have the budget to spend the time. But what didn’t need as much work as it did was the character designs: returning characters don’t look at all like their previous selves despite there only being some months since we’ve seen them in 999 or VLR. Akane and Junpei can really only be identified by their names and voices, as their looks are completely different, making it much harder to treat them as the old characters. Sigma and Phi aren’t quite as bad but the differences still catch the eye.
The redesign of the old characters isn’t helped at all that quite a few of the returning characters don’t really even act like themselves in the story. This is even noted in the game, especially with Junpei, but we never really get great explanations for it. Sigma also acts out in a pretty weird way considering the life he’s lived thus far as detailed in Virtue’s Last Reward. I was sorta expecting there to be a plot twist about one of the returning characters being an impostor, but no, these are the real deals.
And the graphics aren’t even the only problem! The resolution selection in the launcher didn’t seem to work so I just set the resolution by rescaling the window and ended with the uncommon resolution of 1729 × 979. Probably should’ve just edited the configuration file directly. They’ve also landed on a worse control scheme than Virtue’s Last Reward, so I ended up using the mouse for way more things than I wanted. In VLR, I used the keyboard a lot more for camera control and for inventory usage. And considering the sheer amount of text inputs in the game, was it that much hard to implement keyboard typing? You really can tell that this is a console port.
On the core gameplay front, gone is the visual novel style storytelling. The story is instead told as (poorly-animated) cutscenes from a third-person perspective. While I understand the reasoning for it, since it ties into the storytelling with the revolving character focus, I do still much prefer the first-person perspectives of the prior instalments. It made you feel like a participant in the games and not a mere observer, robbed of all agency.
The escape room segments aren’t too dissimilar from the previous entries. I’d say that the difficulty level is somewhere between the straight-and-direct 999 escape segments and the more obtuse VLR escapes. You probably still want to have a notebook at hand like with Virtue’s Last Reward, as there’s quite a few things to remember. What was surprising was that there’s actually not that many escape segments. I got the achievement for solving everything with like a quarter of the game left.
Once you get to a point where the story starts coming together, it does start to feel like proper good old Zero Escape. To a point where I reconsidered my earlier opinion and felt like it was a game worth playing, even if was clearly a much weaker presentation than the earlier titles. And as per usual in the franchise, you also get some plot twists to open up the mystery. And when I got to the big plot twist, my faith started faltering again. It just felt like it was reaching a bit too much, a bit of an asspull. At this point I was once again starting to reconsider whether or not I actually liked this game.
Quite honestly, I’m still not sure. While the big twist wasn’t completely without any setup, the foreshadowing was extremely subtle. In the previous Zero Escape titles, I felt like Uchikoshi had thoroughly bamboozled me when I got an unexpected twist, and here my first reaction was that he’d cheated me. Even after scouring the Internet for all of the foreshadowing in the game, it still feels more unfair than the foreshadowing in Virtue’s Last Reward. I guess it is at least thematically fitting, as the game is about the inherent unfairness of life. Maybe this game is just as unfair.
There’s also things about the story that feel like jumping the shark. The Zero Escape world has always been pretty out there, with elements of sci-fi, pseudoscience and the supernatural all thrown in there. Still, it feels like Zero Time Dilemma takes it maybe a bit too far, with new and even wilder things coming out on a regular basis. I can’t be too mad about it, but I also can’t overlook it. They just seem like extremely crazy concepts that are mostly there as convenient plot devices.
Zero Time Dilemma – the first half of the story is a mess, the visual presentation feels like a bad student project, technical side has significant regressions from the prior games, the storytelling gameplay is less engaging, the big twist is potentially maddening, and some developments take it a bit too far. This doesn’t sound like a winning combination to me. And yet, I feel like I can at least somewhat recommend playing through it, despite its many flaws. This assuming that you are already invested into the Zero Escape world. Anyone else should run away screaming.
While some of the developments range from confusing to maddening, there’s still an interesting plot once it starts to come together. Uchikoshi at his worst is still Uchikoshi. Zero Time Dilemma also manages to do the thing that I was most expecting from it: shed some light on what questions remained from Virtue’s Last Reward. After I finished playing through the game, I asked myself if I regretted spending 20 hours on it, and the answer was “no”. Thankfully, as the final entry in the Zero Escape trilogy, it also doesn’t leave us players hanging for a continuation. Probably for the best, since who knows how hard Uchikoshi would need to stretch to continue the story and to pull twists on us again.
The Somnium Files are back except this time instead of running around as Date Kaname, you’re now given control of two different protagonists, who split the game’s plot into two halves: the past and the present. The past/present split gives the game a more linear presentation than the first game’s non-linear, wilder branching. Otherwise the gist is the same: run around like crazy and investigate a complex and bizarre murder mystery. Tone of the game is also the same: can take itself seriously but trends towards wackier, light-hearted comedy whenever it can.
Narratively nirvanA Initiative is good. It’s not as good as the original Somnium Files, but it’s good and enjoyable. If you liked the first story, you'll probably like this one too. Granted, some of the elements like the mastermind is weaker and the game is definitely more abstract, but I wouldn’t say that any part of the game ruins the journey. I also give it credit for not just retreading the track that the original game took, so it doesn’t feel like you’re just doing a replay. The characters in Nirvana Initiative, both returning and new, are still very strong and definitely made me want to keep playing through. While I don’t think that either of the new protagonists is as fun and full of a package as Date was, I still liked both of them.
Some bits of the story definitely feel like retcons, but I found the explanations given to make sense in-universe, so it didn’t really bother me. Not that any of them were so major that they’d be worth getting upset over, even if they left them completely unexplained. I also think that the additional backstory that they gave to some of the characters really rounded up some of the mysteries of the first game, so I’m a-okay with them.
There are definitely flaws in the storytelling though. Some bits were seemingly brought up to set up another scene without really ever being explained later on. Uchikoshi could’ve definitely done an additional pass through the story to make sure he didn’t introduce something only to drop it in the second scene. There were also some red herrings that seemed to not matter in any way, and felt like they only existed to fuck with you without any explanation why they were there in the first place. Could be that there was some information to explain away these red herrings, but with the game being almost 30 hours long and clues being spread quite thinly throughout that duration, it’s not easy to ascertain that there wasn’t a throwaway line somewhere that explained it. And even if there were, I’d still consider it a demerit that it wasn’t brought up in a bigger way when they were showcased as an indication of some kind of a conspiracy.
Then there’s the big twist. I wasn’t sure how to feel about the big twist at first, since it initially really felt like an asspull. However, given a bit of time and letting the game explain itself, it did start to mesh together to a point where I appreciated it more. There were also at least some hinting toward it, although I think there could’ve been more (unless I was just stupid and missed all of it), as the indication of the twist really just manifested as a sensation of something being really wrong. I imagine that basically no one managed to guess it correctly during the story. The big twist is also quite different from the big twist of the first game, so it doesn’t feel like just a repeat of the first game. So different that it’s probably why a lot of people seem to be turned away from the game entirely. You either like it or you hate it, whereas I think almost everyone liked the original game’s big twist.
The gameplay is almost the same as with the previous instalment, with most of the time being spent in the visual novel investigatory sections with Somniums in between. The biggest core change is that they’ve replaced the evidence-showing interrogation sections with virtual reality puzzle-solving sections, where you explore the scene of the crime in a 3D space and create a reconstruction of the events. Sorta like a mini-Somnium with a pop quiz at the end. I quite liked these VR sections, but I would’ve liked to have some of the evidence sections of the previous game too. Would’ve made it feel a bit more like you were doing proper detective work.
Quick-time events are still very much present, much to many people’s chagrin I imagine. They shouldn’t take anyone by surprise at this point, given the amount of Pavlovian conditioning the series has done by playing the same battle music every time you’re expected to hit random buttons in rhythm. I don’t exactly hate the QTE sections, but they’re hardly the best and brightest gameplay available. At best they're a change of pace from the point-and-clicking adventuring. At least the battle choreography for these sections is fun to watch.
While not drastically different, the Somnium sections seem like a step up from the first game. Not only do the actions feel less like taking stabs in the dark, but there’s also more actual puzzle-solving. Less of “I guessed it right” and more of “I solved the riddle.” Granted, there is still some amount of brute-forcing that you can do and I managed to send one Somnium by guessing the final answer, as my guess was just the correct size to fit the answer box. Still, the amount of variance in the Somnium gameplay mechanics really does make them feel more thought out than previously.
However, I’m not sure if there really was a need for so many Somnium segments, since a few of them didn’t really move the plot forwards all that much. Maybe they just wanted to break up the visual novel sections a bit more?
On the technical side, nirvanA Initiative is not much different from the first game. I could’ve done with better anti-aliasing and improved support for using a controller and keyboard simultaneously. The Somnium transition video is also compressed to hell and back. Surely there was no need to squeeze down the game’s install size this much? Thankfully this time they managed to make the ending movie play in the correct language, so there’s been at least some progress. A passing grade.
Localisation side is good but not perfect. I played through with the Japanese audio track (as one should) and some of the translated jokes really did not land as well in English as they did in the original script. I understand that some of the puns are definitely somewhere between hard and impossible to get right, but there were definitely places where there was a way to deliver the joke better. Some of it might’ve been due to trying to gear the game towards western audiences, with the expectation that the average player wouldn’t really get any references.
All in all, while not quite as brilliant as the first game, nirvanA Initiative is still a fun and interesting sequel, and one that I would recommend to anyone who liked the first game. If you didn't like it, you won't like this one either. And if you haven't played the original? Go play it instead. It’s a sequel, what’d you expect?
If they ever make a third game, aI’ll be sure to cobble up that one as well. A day one purchase, most likely.
I recently got convinced to switch from rTorrent to qBittorrent for my Linux-downloading needs. Tried it out for a bit and decided that it was indeed the superior of the two, meaning I fully migrated all of my Linux ISOs to it. Unfortunately, I didn't really like any of the available clients for it, so I asked myself what any good software engineer would at that point: how hard can it be?
I first thought about writing one in Rust, since I've had some experience with it lately, first building an automatic file sorter and later an anime scrobbler, but after a cursory look at Cursive and Ratatui, I buried that idea. Cursive just seemed like a bad fit and while Ratatui seemed like it could be a good fit for building my application, the tutorials were in middle of a rewrite and I didn't want to spend too much time learning from examples that were already abandoned. Rust and TUIs are tricky enough as they are without a bunch of outdated documentation.
Since I didn't want to make yet another web client for qBittorrent, and I'd preferably have something with at least decent performance, I figured I could try writing a native desktop application. And since I'm on macOS and Swift is actually a really good language, Swift it was. I've had a bit of experience in writing Mac apps in Swift before, but I figured that while I was pursuing my half-baked idea, I might as well learn something new while I was at it.
Cue SwiftUI, Apple's hot new declarative user interface framework that's actually not that new since it was released almost five years ago. Not sure if it's even that hot since I never hear anyone actually using it to ship products. Well, it's the thought that counts anyways.
Getting started
The first thing I noticed about SwiftUI applications is that they take absolutely ages to build. People complaining about Rust having long compile times should try SwiftUI to get some perspective. Granted, most Rust programs don’t have graphical user interfaces, but it’s still quite a grating development experience. Change one line and it takes like 30–120 seconds to get a new debug build running on an M1 Max. I’ve done some stuff with Storyboard and XIB interfaces before and I do not remember any of them being nearly as bad.
I first thought that the long build times were just a me problem, but after bitching about it on Bluesky, I did get another developer chime in to confirm that yes, SwiftUI takes a long time to build.
The other thing that I quite quickly noticed was that SwiftUI does produce quite a lot of annoying and obscure roadblocks for you to decipher. For example, if you fuck up your types and are writing views that have any girth to them at all, Swift will just shame you about your massive views without telling what type error you've produced. That's a super fun one. Apparently Swift's type checker is just so slow that it decides to mercy kill the whole operation.
I've also had Xcode hard crash quite a few times while building out my user interface. I imagine most of them have been triggered by the preview window rendering things. Thankfully I've never lost any work from these crashes, only wasted a bit of time.
Redraw woes
One thing about SwiftUI's automagic logic of updating your views whenever the application state changes is that it has some very definite flaws in it. I wanted to show my data in a table view where the table rows are selectable, but because I also have OCD, I want to be able to deselect rows to get my application back to a serene unselected state. In order to fulfil my desires, I figured that I'd add a keypress handler to my table so that every time Escape is pressed, the table's selection binding is cleared and my selection would be gone. Basically, like this:
struct TableView: View {
@EnvironmentObject var dataModel: DataModel
@State private var tableSelection = Set<RowObject.ID>()
var body: some View {
Table(dataModel.tableCollection, selection: $tableSelection) { ... }
.onKeyPress(.escape) {
tableSelection.removeAll()
return .handled
}
}
}
Good idea in theory. You update tableSelection that is the selection binding for the Table view and the table should redraw without any rows selected. Unfortunately, while it does clear the selection, the table won't show the selection disappearing until the data source for the table updates. Since I'm building a torrent client that fetches the current state every 𝓃 seconds, it might take up to 𝓃 seconds for the deselection to be visible to the user after they press it. Not great.
So, what if I just forced the data source to update? Would that work? Yeah, kinda.
struct TableView: View {
@EnvironmentObject var dataModel: DataModel
@State private var tableSelection = Set<RowObject.ID>()
var body: some View {
Table(dataModel.tableCollection, selection: $tableSelection) { ... }
.onKeyPress(.escape) {
tableSelection.removeAll()
dataModel.objectWillChange.send()
return .handled
}
}
}
By sending the objectWillChange event for my table's data source, I am able to force the Table to redraw, which will make the deselection action feel and look responsive. Worked absolutely perfectly in my testing. So where's the kinda part?
Unfortunately this (seemingly?) functional code is actually undefined behaviour and Xcode will give you a big warning saying it's verboten. So what can I do instead?
struct TableView: View {
@EnvironmentObject var dataModel: DataModel
@State private var tableSelection = Set<RowObject.ID>()
var body: some View {
Table(dataModel.tableCollection, selection: $tableSelection) { ... }
.onKeyPress(.escape) {
DispatchQueue.main.async {
tableSelection.removeAll()
dataModel.objectWillChange.send()
}
return .handled
}
}
}
Turns out that the solution is to wrap the code in DispatchQueue.main.async and you're golden. It works perfectly fine and gives zero errors. Not sure if this is actually much better than the previous solution, since we're just kicking off the code to be executed asynchronously in the main thread instead of doing it here and now. Shouldn't even be any difference in the thread since onKeyPress() handler should already be on the main thread, as calling DispatchQueue.main.sync instead will horribly crash your application. Would definitely be better if I could somehow either force the Table itself to update itself or have it update automatically whenever its selection changes, but at least there's some way to do it.
Is this supposed to work like this?
One thing that I wasn't really fond of in the default qBittorrent user interface is that you can't easily remove a torrent and delete its data at the same time, since you always have to separately click a checkbox to delete files. Hence, my client would have two separate actions: remove and remove + delete. And those actions should be possible with keyboard alone.
However, since it's quite risky to delete things without any confirmation, I decided to add a confirmation prompt before anything happens. Finder also does a very similar thing when you delete files on a network share, so it should feel pretty natural on macOS.
Thankfully this is fairly easy in SwiftUI. Just added a confirmationDialog to my view with a boolean binding that controls if the confirmation dialog is presented or not, with some customisations to boot.
The end result is very similar to what you get from Finder, except that my button has too much text for the buttons to fit on one row so they're stacked. Had I opted for a shorter "Delete" instead, I'd get them on one row.
Unfortunately what was less simple was the keyboard shortcuts. You get Escape as the cancel button for free by just having your cancel button defined with a "cancel" role, but the main action button isn't free. From what I've gathered is that you're supposed to use Button.keyboardShortcut(.defaultAction) to define that a button is the main button and the default key binding for the given scenario should be used. For a regular confirmation dialog, that button is naturally Enter.
.confirmationDialog(...) {
// This button is highlighted in blue and can be activated with Enter.
Button("Delete") { ... }.keyboardShortcut(.defaultAction)
// This button can be activated with Escape.
Button("Cancel", role: .cancel) { ... }
}
But if you want to create a destructive confirmation dialog like what you find in Finder, there seemingly does not exist a keybinding that would allow you to press that button – even though Finder lets you do it with ⌘D. At least I couldn't find one, and even if I did, it'd be so obscure that no Mac user would actually stumble upon it.
.confirmationDialog(...) {
// This button is highlighted in red and can be activated with ???.
Button("Delete", role: .destructive) {
...
}.keyboardShortcut(.defaultAction)
// This button can be activated with Escape.
Button("Cancel", role: .cancel) { ... }
}
I imagine that this is not intended, and that there should exist a .defaultAction shortcut that you can use with destructive confirmation dialogs. But since iOS is very light on keyboard shortcuts (I think you can use them if you have an iPad with a keyboard case?), I imagine that these sorts of things are just not a priority to Apple, as macOS is not where the money is. So after I'd determined that I probably wasn't just being dumb, I opted to just do the obvious thing instead:
.confirmationDialog(...) {
// This button is highlighted in red and can be activated with Cmd+D.
Button("Delete", role: .destructive) {
...
}.keyboardShortcut(KeyEquivalent("D"), modifiers: .command)
// This button can be activated with Escape.
Button("Cancel", role: .cancel) { ... }
}
Hope that Apple never changes how the one in Finder works.
It it any good? What's it for?
After I'd spent a week or two working on this project, I opened up an old Storyboard-based project of mine since I had to make some updates to it. More specifically, I had to add a new text field to a settings section, which was to be located underneath an existing text field. And while I'd always considered the Xcode Interface Builder to be fairly easy and fun, aligning things and adding constraints manually did feel extremely silly after having worked in SwiftUI. Why am I doing all of this silly manual work, dragging and dropping fields and setting alignment constraints, when I could just do this?
Form {
TextField(text: $old, prompt: Text("Old and busted")) {
Text("Old and busted")
}
TextField(text: $new, prompt: Text("New hotness")) {
Text("New hotness")
}
}
Fast, simple, concise, automatic alignment. What's not to love?
On the other hand, whenever I did something that SwiftUI wasn't really designed to do, like have multiple lists in a NavigationSplitView sidebar, then it's a massive pain. Just getting this janky piece of shit solution working took a long while:
It's kinda hard to see why SwiftUI exists in the first place. I don't actually have any data but I'm pretty sure native desktop and mobile applications are on a steep decline, with companies electing to ship Electron or React Native apps instead of writing native applications for macOS, Windows, Linux, iOS and Android separately. Writing a truly native application is only for the biggest of companies, who can afford to write a handful of separate applications, or small boutique shops, who are targetting a niche. Perhaps Apple engineers are just building it for themselves after getting tired of having to deal with multiple different UI systems at the same time, since everything they make needs to ship on iOS, iPadOS and macOS. They even ship the Calculator app on the iPadOS nowadays.
Glad that it exists though. For all the frustrations I had during my time working with SwiftUI, I did actually find it pretty nice and fun. My gaming PC went unused for days during my busiest days building my app. I definitely want to try doing more native Apple development in the future. I want to try and port my client to iPadOS at some point in the future too, since it shouldn't be a massive undertaking with SwiftUI. Not that I'd be able to distribute it unless there's some drastic changes to iPadOS distribution coming, since I don't pay for Apple's developer program. No signing or notarisation for my apps.
Although the next time I try to build anything that requires HTTP, I'm going to start immediately with Alamofire instead of trying to make do with URLSession. Anything more than GET and you'll find yourself building another Alamofire.
The app
All in all, I did manage to actually make something: a very bare-bones qBittorrent client called Dreadnought. I even managed to make a shitty application icon for it! Progress has been extremely slow lately though, since I hit a point where I managed to make an application that does like 80% of the things I need in the qBittorrent Web UI, so everything else is more work for increasingly smaller gains.
I do use it daily though, so it definitely serves someone's purpose, but it's hard to recommend to anyone that is not me, myself nor I. Someone might even call this a feature, since building apps with no general use means getting no support requests. I also wouldn't be surprised if the state updates are expensive as hell and if it was leaking memory, since I have done basically no performance testing. Well if nothing else, at least it ships without a copy of Chromium bundled inside it.
One of my least popular takes as a Professional Python Developer™ seems to be that I don’t actually consider the Python ecosystem and its tooling to be "hot garbage". I also haven’t really bothered with project management tools like Poetry, since I think you can use virtual environments and pip and be just fine. Really, the only tools in my arsenal that I consider to be absolutely essential have been pipx, as distributing Python tools is kind of a pain, and pyenv, because Homebrew will absolutely break your Python installation without any mercy if you so much as blink. Of course, I do also use other tools too, but I don't consider them as essential as those two.
However, I do recognise that there is some value in one-stop-shop tools like Rust’s Cargo, which bakes in a lot more project management stuff into your Rust development workflow and is generally pretty nice to use. For this reason I was quite intrigued by Rye after watching Armin Ronacher’s demonstration video, since it seems like a version of Cargo (and to some degree Rustup) for the Python world.
On a conceptual level, it sounds very good. Instead of installing a bunch of tools and managing/using them individually, Rye would bundle together all sorts of tools and functionality, and let me write Python software with basically Rye alone. The number of different tools that I use and Rye would replace is actually rather long:
current tool
The Rye Way
pyenv
Rye automatically installs whatever Python version your project uses.
pip
Rye installs and pins your dependencies as you add them to your project.
pipx
Rye allows installing Python tools in isolated environments just like pipx.
VirtualFish
Rye automatically sets up a virtualenv in your project directory.
Black
Rye bundles Ruff for formatting (rye fmt).
Ruff
Rye bundles Ruff for linting (rye lint).
I started off by migrating all of my pipx installs to Rye and it worked great. All tools were ready and usable on my $PATH and I could even add optional dependencies with --extra-requirement flags. Not sure if there’s a handy way of adding those extra requirements after the fact like you can in pipx with pipx inject though. I imagine I could always just manually install them into the virtual environment but I’d prefer if there was a more straightforward and obvious way. Would make it even easier to use Rye as a drop-in replacement for pipx.
As for actual development, I decided to start off small and convert a tiny Python script I use for fetching data from an API to work with Rye. Since this was just a tiny script that I run in a Docker container, it wasn't exactly organized in the way that Rye would organize a script project, so I decided to also make it a bit more structured with a script entry-point and everything. This isn't actually a requirement but I figured that I should try to do it like Rye would want me to. For what it's worth, the default Rye-tastic way of organizing code is fairly logical and probably good starting point for people new to Python. Personally, I've never really used /src/ directories in Python but I don't hate it either.
For this tiny project, everything worked very nicely. Granted, there was only one dependency (requests), so it wasn't exactly a demanding test. But Rye would bootstrap whatever Python version I had in the .python-version text file, create a virtualenv out of it, install my dependencies into it and then allow me to run my script with just python like I didn't have a virtualenv at all.
And if I upgraded or downgraded my Python version by changing the value in the .python-version file, it would bootstrap everything again to make sure my development environment matched the requirements. This would make version upgrades a lot easier for projects with lots of different developers, as you could be pretty confident in everyone having an up-to-date development environment if they just run rye sync. I've had to work on projects that upgraded between major versions of Python, and you'd always have at least some issues bringing everyone up to speed.
Emboldened by the success of using Rye with this tiny script, I moved onto the next trial: using Rye to develop a Django project. And this is where the pain begins, and not (just) because I picked an example with an ancient Django 2.0 codebase.
I bootstrapped my Django project as a virtual package with Rye and managed to install all of the dependencies for my Django project that are required to run the application locally. Then I started to add the dependencies I need to actually serve it on the web and I step on the big fat LEGO brick lying in wait: I just can't install uWSGI. The reason for this is actually quite simple: uWSGI needs to compile C extensions, Rye uses static builds of Python, and static Python builds + compiling C extensions is a known house of pain.
The Rye FAQ even admits how the situation when it comes to C extensions is a problem, and one with no real solution at the moment. It does however offer a workaround: register a non-static Python build that you've obtained from somewhere as a Rye toolchain. Since Homebrew has decided to grace my Mac Studio with a Python 3.9 installation suitable for my purposes*, I decided to register that.
* Technically Django 2.0 doesn't actually support Python 3.9 but fortune favours the bold.
$ which python3.9
/opt/homebrew/bin/python3.9
$ rye toolchain register --name=homebrew /opt/homebrew/bin/python3.9
Registered /opt/homebrew/bin/python3.9 as homebrew@3.9.18
Obviously I would never recommend actually using Homebrew-supplied Python installations for any kind of development work, but I figured it'd be fine for testing. Now I can just pin my custom toolchain to my project and create the virtualenv from that.
$ rye pin homebrew@3.9.18
pinned homebrew@3.9.18 in project/.python-version
$ rye sync
Python version mismatch (found cpython@3.9.18, expected homebrew@3.9.18), recreating.
error: failed fetching toolchain ahead of sync
Caused by:
unknown version homebrew-aarch64-macos@3.9.18
Oops.
Turns out that Rye is so set on managing your (static) Python toolchains that it will even attempt to download a toolchain already on your machine, and then fail because your custom toolchain doesn't actually exist beyond your machine. It won't even work even if you omit the custom name from rye toolchain register.
So yeah, Rye kinda sucks and is all sorts of broken if you ever need dependencies that compile C code. Obviously I didn't test out a wide array of C extensions, but considering that it's a very well-known problem with static builds, you'll probably trip up yourself sooner or later if you do a lot of Python development.
On one hand, I do understand that compiling software sucks because you need to hoard compile-time requirements and compiling software takes forever and makes your laptop burn your thighs and so on. However, I do think that compiling software like your Python development environment has some very significant upsides, such as the fact that your build of Python will reference paths actually present on your actual computer. Static Python builds are fine for some stuff but I feel like they're a bad idea to be the default and a truly awful idea to be the only option. The support for static compilation of Python just isn't there yet.
Even if the custom toolchain registering worked like it's supposed to, I find this design approach to completely water down the "one-stop-shop for all Python users" idea put forth by Rye, since I'm still required to install pyenv on my machine in order to have a toolchain to register in the first place. I think Rye would be a much better tool if it offered pyenv-style compilation of Python, and preferably made that the default option. Or at the very least ask me which style I prefer during the installation process. It already asks you how you want the python command work outside Rye-managed projects, so it's not unthinkable to have an option for compilation too.
In terms of ergonomics when it came to the Django project (the parts that worked), I also kinda miss the environment variable plugin that VirtualFish had, where it would set environment variables when you activated a virtualenv. As far as I can see, the way you're supposed to integrate environment variables as part of your Python development is by adding them to your Rye scripts. So if I want to set DJANGO_SETTINGS_MODULE for when I run any of the Django management commands, I'd chuck it in a separate file like .env and add this to my pyproject.toml file:
You could also add environment variables directly to the script definition but that sounds very not-portable if you work with other developers. But now I can run tests against my development settings module. I just need to use rye run manage to invoke the test management command instead of for example the manage.py script that Django will give you.
$ rye run manage test
Creating test database for alias 'default'...
System check identified no issues (0 silenced).
..................
---------------------------------------------------------------------
Ran 18 tests in 0.215s
OK
Destroying test database for alias 'default'...
I think this is totally reasonable and maybe juggling shell session environment variables isn't the job of a Python development tool anyway, but I do still kinda wish for there to be some kind of a better way for it.
All in all, I have mixed feelings about Rye. On one hand, many of the design ideas are good and I think I'd like to use a Cargo-but-for-Python to develop software, since I also like using Cargo. However, the insistence on using static Python builds leads to suffering and won't let me uninstall pyenv anyways. There's also some rough edges in implementation to cut yourself up further, like my attempt at getting Rye and Homebrew-Python to mingle shows. The fact that rye test is also just an alias to run pytest with no possibility for any other testing tool like the built-in unittest module, which is totally and absolutely a fine tool for testing that you get out of the box, seems very indicative of just how early in Rye's life we are at the moment.
If you have a working Python development environment, no matter how rudimentary its feature set might be, it might not be worth it to try out Rye now. Its best value at the moment would probably rather be for newbs, who can benefit from its easy setup and sane defaults for many things (like not using the system Python if you've got one). However, I think that there's also a good chance that Rye will actually be all-around great in like 12 months, so I also wouldn't invest too heavily in an alternative tool like Poetry if you're shopping around for project management tools.
Nevertheless, Python is still a fun programming language and the tooling is fine.
Like for many others, Persona 3 was my first foray into the Persona series, having played through FES back in 2011, and will always hold a special place in my heart. I have also been wanting for a Persona 3 remake since at least 2020, so I’ve been extremely excited to play through Persona 3 Reload ever since the rumours were confirmed to be real. And yet, with years of lofty expectations weighing down on it, I absolutely loved Persona 3 Reload.
Story-wise Persona 3 has always remained as my favourite one in the series. Granted, some of that might be nostalgia for my first Persona story, but it’s a feeling that’s stayed with me after Persona 4 and Persona 5. I just really enjoy the theme and overall feeling of the story. And now that I have experienced it again, I think I can still confidently claim it to be the superior of the three. Thankfully Reload manages to not alter the story even though it does ever-so-slightly expand it here and there. Not changing the story is also to its benefit since ATLUS’ efforts to add new story content for re-releases have often had very mixed results. And despite knowing the story, although fuzzy with some details after 13 years, the emotional highs still bit hard.
The gameplay side of things has a lot more changes, but it still very much feels like Persona 3. I was actually quite surprised that they didn’t relax the social link calendar to make it more like Persona 5 Royal. Getting all of my bonds maxed out in a single playthrough was definitely challenging with few days to spare. Granted, the social links are still considerably easier than in the original, but not to the point of feeling like the other games. The overall package does feel like a very good balance of the old experience and new quality of life.
The new Rewind system is excellent. I imagine it’s going to become a new staple feature in every single Persona game going forwards and reduce some of the save overwrite anxiety. Would’ve definitely saved me some time in prior titles. It’s also nice to see that despite trying to be faithful to the original game, they’re not beyond innovating when they have good new ideas, especially when they’re ones that do not dilute the original experience.
The combat side of things has definitely been Persona 5-ified – but actually less so than I imagined. I fully expected them to just clone the Royal combat system and call it a day, but they actually made a system unique to Reload. Familiar, but unique. And much like Persona 5’s combat system, it very much works. It’s fluid and fun. I always remembered Tartarus being the worst part of the original game, but somehow it doesn’t actually even as bad as I remembered in Reload. Either I am completely misremembering the experience or Reload has just improved it that much. Of course, it is still very repetitive labyrinth of procedural generation, but still a decent climb. The visual update also does help to make it less bleak and boring. The background music is however very faithful to the original, as in, not very inspiring. Might’ve listened to a couple of podcasts while I was at it.
The new visuals do look and feel excellent. It’s not exactly the sharpest edge of graphics technology and not all of the art assets are polished to the highest degree, but the style does make up for a lot of it. Especially when the only available versions of Persona 3 have been extremely low fidelity, Reload just feels extremely pretty in comparison. And I just absolutely love all of the character models. I wanted them to remake Persona 3 with the character models from Dancing in Moonlight when that came out, but Reload’s models are even better. And in a series first, there’s ray-traced reflections, which do look really nice.
The biggest downside might be the performance though – there’s some definite Unreal stutter happening even on powerful hardware. Maybe they’ll patch that out in due time, although I wouldn’t count on it. There’s definitely room for improvement on that department.
There’s still 2D-animated cutscenes, although some cutscenes that I remember being 2D animation in the original are now 3D live-rendered cutscenes. While the quality of the 2D animation has never been the highest, both in the original and Reload, I’m of two minds about this. The cutscenes in Reload are definitely good, but there’s just something irresistible about 2D cutscenes.
The voice acting in Persona 3 Reload is immaculate. First of all, almost all of the original cast from Persona 3 returns to redo their lines and they all sound exactly how I remember them sounding over a decade ago. The only exception that I am aware is Tanonaka, who passed away in 2010, and they got a fairly alright soundalike to voice Igor in his stead. And to make things even better, they got a bunch of voice actors for all of the social links, with every single rank of every single social link (bar the obvious silent one) being fully voiced! I really hope that we can expect all new Persona games to keep up this level of quality and quantity when it comes to voice acting.
The music side initially feels rather uncanny. Everything sounds so familiar to the songs I’ve had in my music library for the past ten years and yet are strikingly different. However, that feeling of uncanniness did wear off as I kept playing, and the music does feel like Persona 3 – even if served with the whiff of a bootleg recording. The couple of pivotal songs during critical parts of the game are also kept and sound very similar to the originals despite Kawamura Yumi being replaced with a new face in Takahashi Azumi. There are also a couple of original bangers, so I’ll definitely be adding Reload’s soundtrack to my usual music rotation in the future.
Yet I unfortunately can’t fully shake the feeling that maybe a small bit of Persona 3’s identity has been lost in the Persona 5-ification of the game. Not to a degree where I’d want to not play this or not recommend it to others, but just enough to make me think that there’s still a place in the universe for the original. I know that there’s definitely still a place for Portable, since the alternative perspective story for the female protagonist is not part of a Persona 3 remake. A lot of people were very vocal about this omission, but I don’t consider it critical: Portable always made it very clear that the female protagonist route was an alternative one. A new perspective, if you will. And who knows, maybe ATLUS will find the strength and courage to take our money again in the form of DLC for new perspectives.
Remakes are always a contentious topic; if it’s okay to replace part of the history with the new and shiny. However, I do think that Persona 3 Reload is a product of exquisite quality and that it’s definitely a great version of Persona 3 and more accessible for fans both new and old who’ve gotten used to the new way of things. However, that is what I think it should be considered as: a new and different version of Persona 3, and not as the new Persona 3 to replace the old. A great way to experience the original story, even if wrapped in a remixed experience. It’s not like hordes of people were going to dust off their PlayStation 2 to experience Persona 3, especially when the games are often going for what they retailed for originally – or possibly even more.
As an old fan of the series, I do highly recommend Persona 3 Reload. Whatever annoyance I felt before for the increased 70€ price tag has been replaced by the sheer joy that I got from my 100-hour journey into an old friend. I also imagine that people completely unfamiliar with Persona 3 will also enjoy it, as it is definitely a great game.
PS: Don't pay extra for the digital soundtrack. You'll hate yourself if you do. Just wait for a real soundtrack release.
Everyone’s favourite Scooby-Doo game on the PlayStation Vita.
Y(o)u're a Japanese youth, kicked from the big city into the boonies, when shit kicks off and people start dying mysteriously in an otherwise quiet small town – and the only one who can reach out to the truth is you and your merry gang of superpowered high-school students.
A role-playing game lives or dies by its story and thankfully the story in Persona 4 Golden is good. Not maybe the most clever or through-provoking one, especially for supposedly being a murder mystery, but still a good experience. Granted, I knew the general gist of the story since I’ve played through Persona 4 on the PlayStation 2 over a decade ago. But despite knowing what generally will happen, it was still a fun time reading through. And since it’s been a while since I last read through it, and I’d already forgotten some parts of the Persona 4 story, I could experience some things with “fresh” eyes, which was also nice. I imagine any players without any knowledge about Persona 4 would enjoy it even further.
The chemistry between the character in the story is also good. By the end of the game, it really feels like you’ve made unbreakable bonds with your social links. I also enjoyed the humorous slice of life events that didn’t carry the story forwards. It’s just a fairly good time in general.
Most of the game is spent living the life of the most popular and proactive student in Japan by improving yourself and socialising with your friends and neighbours. These day-to-day activities can be quite overwhelming, as you might have half a dozen citizens of Inaba vying for your attention at any given moment. It does very much feel like there is not enough time to get everything done within a single playthrough, and I did actually fail to fully nurture two social links in a single playthrough. However, it’s also somehow so incredibly addicting for reasons completely oblivious to me. At times I found it hard to stop playing even though I really had to.
This addictive gameplay also extends to the actual dungeon crawling part of the game: it’s really fun and at times it was hard to put down my Steam Deck to go do other things. Maybe the only thing I have against it after Persona 5 is that the battle interface feels clunky and outdated. Not to a point where it would’ve stopped me from playing, but definitely had some more friction in the beginning. I also know that some people do not like the dungeons in Persona 4, but I actually didn’t have a problem with them. Most of them have pretty nice designs and soundtracks, so I can’t hate them, and since they’re randomly generated, you can’t just memorise and blast through them.
I can’t really remember exactly how all of the aspects in the original Persona 4 worked, but Golden does feel a lot more polished than the original. Easier too, which is seemingly a thing that always happens when they re-release any of the Persona games. However, the mechanics still feel quite stiff to all of the modernisations they stuck into Persona 5, so I definitely recommend people interested in the Persona series to start off with Persona 4 Golden instead of Persona 5 or Persona 5 Royal. The transition from P4G to P5R is definitely smoother than going in reverse.
Not completely sure about the third semester / epilogue though. It seems kinda of an unnecessary addition, and mainly revolves around a character who seems quite forcibly added to an existing and complete story and one that you’ll probably struggle to care about. That being said, it’s not exactly long to get through, so it kinda slipped past me anyway.
Unfortunately, Persona 4 Golden is showing its age in the graphics department though. We’re talking about a PlayStation 2 game that was polished up for the portable PlayStation Vita and is then rendered at a slightly higher resolution on the PC. That being said, it’s still absolutely fine. The art style does make up for a lot of the low fidelity and the 2D sprites are ageless. Still, the 15-year patina is another reason I’d recommend people to start off with Persona 4 Golden before moving onto flashier titles like Persona 5 Royal or Persona 3 Reload. It’s not like Persona games have any kind of interconnected stories, so you can freely choose what order you play them in.
Thankfully the soundtrack still absolutely holds up. Never has there been a Persona game with a bad soundtrack and you are definitely going to be listening to the soundtrack outside of the game too. Guaranteed.
And as you might expect from a PlayStation Vita title, this game feels tailor-made for the Steam Deck. The gameplay really feels at home on a platform where you can suspend and resume the game as you please. The only thing that gives off the feeling that it’s not a native title for the device are these small freezes that occur here and there. I’m talking like a half second freeze for half an hour of gameplay. Small hiccups, not exactly something to worry about. Otherwise the performance is rock-solid, even if I did often limit the game to 60 FPS instead of the possible 90 FPS (which was smooth and solid) for some battery life improvements. There’s even proper Steam Deck button graphics in the game!
Actually, the game looks downright silly on a 27-inch monitor, and really shows how this game was designed as a portable title. All of the user interface elements look XXL-sized and the low-fidelity assets are really laid out bare. I also can’t get the vibrant colours and inky blacks on my PC monitor as I could with my OLED Steam Deck screen, which made it feel like I was playing it on a supersized PlayStation Vita. Granted, I do have a pretty awful PC monitor, but so do a lot of people. I’d maybe even go as far as to recommend this game a little bit less if you cannot play it on a Steam Deck. Not that a Steam Deck is absolutely necessary – but I’d argue that it’s the best way to experience this game. Persona 4 Golden was a PlayStation Vita must-have, and now it’s a Steam Deck must-have.
I must also confession something: even though I had Persona 4 Golden on my Vita for the longest time, I never actually managed to finish it there. Kind of a PlayStation Vita sin, not finishing one of its greatest hits. I guess I was saving it for when I needed a portable time-killer and never actually got to the end despite starting the adventure. But maybe that was for the best – since I think that the ultimate way of enjoying this game is not on the poor old PlayStation Vita, but rather on the Steam Deck.
Persona 4 Golden is, despite its obvious veteran age, still a fantastic and fun game and it should probably be the first Persona game for newcomers for the series. An absolute must-have gem of a game.
VA-11 Hall-A is basically a cyberpunk visual novel about a bartender and her clients, with a bartending minigame firmly attached to it. A relatively simple concept really. I'd actually cleared this game once already on the PlayStation Vita, but decided to get another playthrough in it for the yearly Daily VA-11 Hall-A, which I guess is already indicative of the game's quality.
While there is an underlying story in VA-11 Hall-A, it’s not really a plot game. Instead, it’s more of a character and slice of life game about a bar. And thankfully it does this part quite well. The characters are well written and it does generally feel like people are just trying to live their lives in the cyberpunkian shithole city they happen to live in. And to top it off, you do get a satisfactory jumping off point for the story – with some epilogue stuff sprinkled on top of it if you managed to do your day job properly.
There is also another strong point for VA-11 Hall-A: the art in this game is lovely. What it maybe lacks in resolution (since the game runs in a 1280x720 window) it more than enough makes up in style and atmosphere. Granted, there’s not really much event CG like you might have in over visual novels, but I don’t think it really matters here. There’s also a satisfying amount of animation for the characters, so it’s definitely an enjoyable visual experience.
The music is also pretty good, although I do have a problem with it: I don’t like the jukebox mechanic. Why am I in charge of picking the music? Can’t you pick appropriate music for me instead so I don’t have to pick something incredibly inappropriate for whatever worries my clientele carries in through the front door?
I do also have to criticise the Steam Deck experience too. Even though this game has been released on the PlayStation Vita and the Nintendo Switch, there’s no official controller layout. This means that it’s kind of a pain to play on the Steam Deck, and I couldn’t really find a controller layout that worked nicely. Touch screen controls for the bartending interface didn’t really work either. And even more annoyingly, even though the game advertises Steam Cloud support, it doesn’t sync save data between Windows and Linux versions. This game is far from a Steam Deck verification and anyone thinking of playing this on one should be aware that there are definitely issues here.
I also have to say that I hate the Model Warrior Julianne minigame and the fact that being it is part of getting all of the achievements. For a while I considered somehow cheating through it but thankfully I managed to eventually get through it by abusing every available cheese. I mean, it is an optional minigame so I can’t hate it too much, but the completionist in me compelled me to get through it and I definitely could’ve lived without it.
VA-11 Hall-A is a fairly short and simple slice of life bartending game, and an enjoyable one at that. If that sounds even remotely like an interesting concept to you, I have a hard time believing you wouldn’t get your money’s worth even at full price.
(Just don't get your hopes up for getting more of it, since the sequel seems to be stuck in development hell.)
The very first impression I had with Starfield was that the world design looked really good. The art direction of Starfield is great. I'm a big fan of the "NASA punk". However, the further I progressed in the game, the less impressed I got with the world. While the individual aspects of it might look good, it's way too repetitive. You land on a single wasteland planet and you've really landed on all of them. And as far as I can tell, the few points of interest in them are repeated throughout the known universe.
The cities are pathetic too. There's a couple of different cities with vastly different designs. They do however have one thing in common: the scale. All of the cities feel woefully small. It feels like I'm walking through a movie set than an actual city. And despite the lack of scale, there's loading screens absolutely everywhere. Get out of your space ship – loading screen. Go to a small residential area – loading screen. Go to the weapon store inside the small residential area – loading screen. Thankfully the missions aren't as full of loading screens but it seems strange that these small cities are in a big-budget 2023 release.
Traversal between places is also deeply unsatisfying. I had to go from a city on a planet, to a bandit camp on the same planet. I opened the extremely poor planet map, selected the bandit camp, and tried to transport there. Sorry, the game said, you can't fast travel to a place you haven't discovered yet. Then it forced me to fast travel to my spaceship and then land my spaceship at the bandit camp. And what does "landing your spaceship" actually entail? A loading screen and a jump cut to the spaceship landing, which all takes around 10 seconds. Almost all traversal in this game is fast travel and yet it still does this idiotic gatekeeping for what it thinks "fast travel" is.
The space bit in Starfield doesn't really improve the situation at all. You can either talk to whatever NPCs happen to be around you, or engage them in battle. That's it. Despite the lack of anything, you'll still be going to space very often since you need to spend at least a few seconds there between fast travels. Actually, the NPCs either reaching out or attacking got very tiresome since I just wanted to travel to my destination and not play tag with pirates.
The repetition in the world design also comes across in the story campaign. The story had me go look for Andreja, and I found her in an enemy-infested mine, and at the end of the mine, I found an artefact. Then later in the story, I was tasked to go look for more artefacts, one of which was at the end of an enemy-infested mine. A mine with the exact same layout, except this one was on a different planet and lacked one Andreja. Both of these mines also had a locked locker room, with contraband for the taking after you picked open the lock. While I can understand that the planets look similar when you have such a vast number of them, I expected a lot more from the main story locations in a AAA game.
Really feels like Bethesda is just taking the piss with the repetition considering missions are actually named "Into the Unknown" and "Further Into the Unknown". A bit on the nose there Todd.
As for the story itself, I didn't really care for it. First of all, I found the premise of how you became an adventurer quite contrived. And while I think they want to make Constellation feel like some sort of an elite gentleman explorer's club, most of the story missions made me feel like their glorified errand girl instead. For every interesting mission, like Unearthed or Entangled, there were two or three missions where I just had to fly from one remote planet to another, literally digging for quest items. Surely there must've been a more interesting story to write in the Starfield universe, right? And despite the fact that the story involves life and death situations, the writing is just so devoid of any edge or character. The whole thing just feels like such a lifeless and watered down experience.
Guess it didn't help that I couldn't care for any of the characters, including my own. The only (human) member of Constellation that I didn't have any active contempt towards was Walter. Vasco was cool too, and got promoted to my primary companion after Sarah got very angry at me for a heinous murder where I shot a man after he and his friends unloaded every bullet they had towards our direction. I guess Sarah was looking to end her miserable existence and I rudely interrupted it by defending myself.
I've heard that some of the side content is much superior but unfortunately I seem to have never encountered it. I guess I did mainly focus on the main quest so maybe that's on me. However, I did encounter side content that I didn't care for.
One example is a mission where settlers task me to drive out a corp from a planet that they consider their birthright and one that they seem willing to fight for. I go talk to the corp and they want me to either blow up the settlers or convince them to become serfs on their luxury resort planet. I decided to not blow up these poor settlers and instead opt to tell them about this indentured servitude proposal. Considering how defiant the settler captain acted, I expected this to lead into a war where they take the planet by force. Nope, the settlers turn heel and agree to become slaves. They they asked me to bring them a bunch of materials, presumably so they can make their own leg irons to really enjoy their bondage. I considered blowing them up for their own good but settled to just leave instead.
I did enjoy the ending though, at least on a conceptual and storytelling level. Unfortunately the gameplay side of it is extremely herky-jerky, and is not at all enhanced by the space combat or waves of trash mobs you also have to deal with. And right at the last second, the story really just raises more questions than it answers. At least the ending was pretty.
As a game, Starfield is a rather unfocused affair. It's so chock-full of different systems that it feels very overwhelming when you get started. Maybe veterans to Bethesda RPGs have already learned to juggle these systems in prior games, but as a newcomer, you need to ingest a lot of (unhelpful) tutorials at the beginning. You have ground traversal, space traversal, ground combat, ship combat, ship building, outpost building, cooking, crafting, spacesuit upgrading, weapon customisation, mystical powers, lockpicking, persuasion, scanning, mining, stealth, encumbrance, commerce, status effects and so on. "A mile wide and an inch deep" is the best quote I have heard to describe this game. Many of these systems are very badly taught to you, and even when you learn them, they're still not that great. Quite possibly why it also took me around ten hours to actually get into the game properly.
The ground combat feels fine. There's a decent selection of firearms available to you and they do feel pretty distinct from each other. It's definitely not very flashy, and there's very limited amount of specialisation you can do with the combat but I did generally enjoy engaging the enemy. The boost pack also livens up the movement for the combat sections too.
The mystical space powers feel like a complete waste of time, at least for enhancing the combat. I really only ever used the unlimited stamina power and otherwise ignored them. And the way that you obtain all of these various powers is so needlessly repetitive. You have to literally jump through hoops for them, again and again and again. It feels more like an elaborate prank than game design.
Space combat is less fine. First of all, whenever you engage in ship combat, you were most likely just on your way to do something completely different until someone decided to harass you. And when they do, they usually do it with their pals. In three-dimensional space. I feel like most of the time in combat is spent around doing loops in order to just find one of the three hostile ships so you can blast them, while trying to keep their two friends from doing the same to you. Even though I rarely struggled, I still did not enjoy it.
Ship building seemed pretty cool, at least for someone that enjoys that kind of a thing. I mainly used it pretty pragmatically, so instead of trying to replicate famous designs or come up with my own, I just used it to upgrade an existing design. It does take some getting used to, since the game doesn't really teach you how to do it and I had to get supplementary lessons from Google on how I can get ship pieces to connect how I want. And unfortunately changing your ship in any way completely destroys any interior decor you might have going on (for understandable technical reasons), so it definitely kept me from doing it just for fun. Maybe people that are heavily into this would find more to criticise about it but I thought that the ship building was cool overall.
Outpost building is also something that is not extremely well taught to you. And it doesn't really feel like you would ever need to do it. However, once I learned a bit on how it works, I got a frenzy of finding resource-heavy planets, building outposts on them all, and connecting them all together to a master outpost. I got pretty absorbed into it until I had all of my outposts connected. Then I realised that I didn't really have any kind of use for it. It just kinda is there, filling up my inventory and few resource crates with a bunch of minerals that I don't really have an use for. I did my outpost fairly early in my run and had no reason to touch it afterwards.
Inventory management is also subpar. For some reason Bethesda decided that the majority of the screen should be taken by 3D representations of your items instead of useful information. While the gun models look nice, I don't find myself that interested in knowing what copper ore looks like. I'd much rather have a side-by-side view of my inventory and a vendor's inventory instead of having to flip between two or three different inventories as I unload my stuff. The vendors will also instantly run out of money when you go sell things, so you need to go do your tedious inventory management with every single vendor in the game. At least I did get used to the inventory, so the user interface is not a constant source of pain, but I still pine for something better.
The lockpicking minigame is actually rather fun but it takes home the prize for the worst explained system in the game. I had no clue on how to do it and was just randomly trying to get through it. Then I watched a short tutorial video on YouTube and got it. Now I'm pretty good at picking open the locks and can actually enjoy some momentary tinkering with locks. What a shame the game doesn't want to teach you about it.
Not a big fan of character creation either. It feels like some of the aspects of it are very varied and some are barely varied at all. I would've also enjoyed having a voice for my character. There's one part in the story that actually makes me feel like it would've been very reasonable and feasible thing even. And I really don't understand the point of the traits. I remember people criticising the lifepaths in Cyberpunk 2077 for having little impact on the game, but I feel like the traits in Starfield have even less of an impact. I was a Neon Street Rat and it just gave me a couple of new choices in one particular part of the game.
Don't really like the skill trees either, especially how it locks some basic abilities in the game behind a skill point. Or many skill points for things in the advanced tiers of the skill tree. I'd expect to be able to do at least some stealth without investing points to a stealth skill but alas. You also need to level up your skills, so don't even think about reaching rank 4 in Medicine without spending time harming yourself and then ingesting health kits by the dozen.
Gatekeeping parts of the game by hiding them in the skill tree are also why I never really got into the crafting and upgrading aspect of the game. You have a few things open at the start of the game but anything beyond that requires investing your few skill points. A shame, since I would've liked to upgrade my weapons more, but Weapon Engineering is an advanced Science skill, so I'd need to invest and level up my science skills before I even got a single point in.
Technical aspects of the game are a bit of a mixed bag. The game was surprisingly stable and bug-free for a new AAA release, although neither aspect was perfect. My crew sank into my spaceship more times than I remember to count and when the game decided to crash, it screwed me out of a lot of progress. At least Cyberpunk 2077 was polite enough autosave very frequently to make up for its lack of stability, and wouldn't make me lose 30 minutes of gameplay with a single hard crash. The performance seems relatively poor considering the lack of ray tracing. Considering how Cyberpunk 2077 looks a lot better with crazy ray tracing, Starfield doesn't deliver that much of a higher frame rate.
It does seem like they did manage to improve it a bit with the recent patch though, but I've yet to clock any serious amount of time in it to make a judgement. Fingers crossed!
I tend to avoid pre-release marketing since I want to temper my expectations for new stuff, but I'm still left feeling disappointed. It's not that Starfield is an awful game. I did manage to beat it after all. It just feels really mediocre and undaring. Perhaps it's worth taking a stab at if you were already paying for Game Pass, or found it in a bargain bin, but for its pain points and overwhelming sense of mediocrity, it's easy to skip and hard to recommend.
CHAOS;HEAD is a really interesting story. On the surface it appears to be a murder mystery story but it's not really that. There are murders and mysteries, but they don't meet in the same way that they would in a whodunit. I guess I'd rather call it a psychological horror, even if it's not really outright horror. And much like STEINS;GATE, the entry in the series that people are probably the most familiar with, it is rife with conspiracies and larger implications. Really enjoyed myself as I read through it for the first ending.
The meat of the story is really in the common route, which then unlocks the other routes and endings after having completed it once. The character-specific routes aren't that special even if they do flesh out some of the characters a bit more and have some good elements. The most outstanding one of them is the Rimi route, which gives the most exposition for the common route, although I did also enjoy the Sena one too as a Sena fan. The endings also felt a bit lacking compared to the emotions I felt from the common route. Even the true ending didn't really change things up that much or be more detailed than the base ending. Thankfully the common route was interesting so I don't feel soured by these issues.
The protagonist Takumi is a flawed but interesting character. An anti-social, delusional otaku hiding from the world in his shipping container home. And because of those delusions, he becomes an unreliable narrator for the story, making the reader unsure what parts of the story are real and what are delusions. He enters a delusion, then something unexpected happens and you can't know if it's still a delusion or not. Or he encounters a stranger that supposedly knows him and you can't know which is true. The unreliability of his delusions really fits the whole mysterious and disturbing theme of the story very well.
The delusion triggers are pretty interesting system that allow having a positive delusion, a negative delusion, or no delusion. The delusions will then trigger either a small scene with no relevance to the plot, or change the actual routing (at least if you answer a quiz correctly). While the game does have an explanation as to why this would happen, it's still pretty hard to figure out why having a negative delusion in a given scenario would result in a different ending. Not quite as opaque of a routing system as the text message system in STEINS;GATE was, but still mixed up a bit. You'll probably want to look up a route guide for this one.
Art in CHAOS;HEAD is just about alright. Some of the character sprites and CG images have a derpy feeling to them, with the characters looking like their facial features are trying to escape their faces. I didn't find the art fundamentally flawed but it did definitely stick out, even in some quite pivotal parts of the story. The characters also have animated mouths but there's not much animation beyond that and some camera pans. And while the CG did have some quality issues, the amount of it was just fine. I guess this isn't really the most visual of packages.
But what was actually superb was the sound design. The sound design in this game is fantastic. Takumi is fully voiced and Yoshino does a superb job at voicing his uncertainties, fears, delusions, paranoia and other ailments. Maybe his performance was actually too good, since it overshadowed some of the side characters, even though there were also good performances there. Like for example Takahashi Chiaki's performance as Yua. The music and sound effects are also to the point when it comes to producing a creepy, supernatural and mysterious atmosphere. There's very few times when I'm impressed to this point by just the sound design in a visual novel. You'll definitely want to have audio on for this experience. Bonus points for an excellent insert song.
You’ll also want to have audio on because it’s very hard to figure out who’s talking without it. The user interface has no indication on who the speaker at the moment is, or if the text is just narration. If you’re deaf, you can go get fucked as far as CHAOS;HEAD is concerned. The interface is also kinda ugly in general and the text display is kinda rough. This is probably a function of the game’s age, as it’s already 15 years old. Not a dealbreaker but something I did definitely note during it.
For what it's worth, installing the Committee of Zero patch does clean up the text display so maybe consider doing that.
Whilst it's largely overshadowed by its excellent younger sibling STEINS;GATE, CHAOS;HEAD is still an interesting story and recommended reading despite some of its flaws.
Maybe I should now check why everyone hates the anime adaptation…
The story of Hyperdimension Neptunia Re;Birth1 isn't really anything special. It's ultimately just kind of a generic save-the-world adventure story and the actual substance is in the (comedic) character interactions more than anything else. I don't think I can claim to have loved the story, but I did find it charming in its own way. At least I don't regret reading through it. The English translation seems a bit liberal, but not to a degree that it'd completely ruin the entire thing either.
The combat is surprisingly good. There's kind of a good rhythm to the whole thing, where you first try to get a sneak attack in the overworld, then take turns in the battle to position yourself optimally, chip away at the enemy's shields and then deal heavy damage to them. During boss fights you would also look to optimise the mileage you get out of your powerful EX skills too. It's also fairly fast-paced, with the possibility to skip most of the animations, so you can really charge through your battles.
And boy is it a good thing the combat feels as good as it does since this game leans very heavily on grinding to pad out the game. I don't think there's any way to get through the story without grinding. You just won't get the levels to clear all the level check bosses. I had trouble beating a couple of the bosses, both early in the game and late in the game, and when I searched for tips in case I missed something, most of the advice on the Internet was "grind more". So if there's a way to beat the game without grinding, it went unnoticed by me and seemingly the rest of the Internet too.
The grinding also extends to the achievements. About half of the achievements for this game require grinding, be it for character levels or money. The amount of grinding is pretty mental. Really has this "low budget, must stretch out the game somehow" thing written all over it.
Not to say that there's no aspect for battle tactics and gameplay. For example, I lost to a boss, restarted the fight and then won the second bout. No grinding needed in the between the attempts. However, I'm not sure if the victory game down to skills or just luck. I feel like there's a limited amount of things you can do in regards to your tactics when the boss has the ability to insta-kill two members from your three-member party. It's quite possible that the boss just unleashed less of its wave of death area-of-effect attacks on my party on the latter attempt.
The rest of the game mechanics are kinda lame. The quests feel like they were added to check a checkbox, same as with the colosseum. And the share balancing act that you need to do in order to unlock the true ending or to unlock the little sisters is also not amazing, especially since it'll involve those quests and colosseum. Not massively time consuming thankfully, but still a bit of a mindless grind, doing the same things again and again to shift some percentages around.
Technically the game isn't the best. While I do like the art, despite the fact that the game reuses assets like it's going out of fashion, the graphics felt so bad and blurry at the beginning that I felt the need to install the Neptastic Mod on my Steam Deck. That did certainly help the game look nicer, although the FPS would not necessarily always be a stable 60 FPS. It also felt like the FPS would gradually get worse as I kept the game open as the Steam Deck was asleep. I think I went from around 40 FPS to 60 FPS in the same dungeon by just simply restarting the game. Deck users beware.
But in general, Hyperdimension Neptunia Re;Birth1 does run pretty well on a Steam Deck. Maybe not as well as you'd expect considering that this is originally a PlayStation Vita game, but well enough even with the modding. Combined with the fast-paced combat and heavy focus on grinding, it makes for a pretty decent portable time killer when you have a couple of minutes to do something. Not really sure if I'd ever want to actually sit in front of my gaming PC and play through this game. It just doesn't feel like that kind of a game (quite possibly due to its Vita roots), which is why I played it from start to finish on the Steam Deck. Fairly certain even that I beat most of the game while sitting on the toilet.
Hyperdimension Neptunia Re;Birth1 is a bit of a messy game with good core gameplay loop and enough charm to keep you around. For the couple of euro I paid for it, it did make for some good portable entertainment. Just don't really expect much beyond that from it.
AI: The Somnium Files is an interactive murder mystery story. You play as Date Kaname, an amnesiac special agent trying to track down a mysterious serial killer using an AI-powered eyeball and a machine that allows invading the dreams of investigation targets to unlock the secrets of their minds by solving puzzles.
But despite being a murder mystery, it kinda doesn't feel like one during a good chunk of it. The mood of it is kinda all over the place and quite a vast chunk of the game is spent engaging in various kinds of comedy routines. I imagine the humour won't land on everyone – it is quite Japanese and trends towards lower than higher end of the brow. I did generally like the comedic bits but it does induce a level of mood whiplash as you go from fawning over a porn magazine to gruesome murder in the span of 20 minutes. Thankfully it does manage to stay serious during the more serious and somber moments of the story, so it's not incredibly jarring.
The Somnium Files features a relatively unique way of branching the storyline. Taking certain actions in the game puts you in a different branch in the story, and some of the branches lock you out in middle of it, telling you to go finish other branches to unlock it. It even unlocks a route, lets you read through it for a bit and then locks it up again in order to throw you onto another, now-unlocked route. At first I wasn't really sure what to think of it, but now in hindsight, I do like the implementation. It does a good job at keeping you guessing for a good chunk of the story, preventing you from reaching the juiciest bits first while simultaneously dropping hints across the different branches.
As for the story itself, I really liked it. It's interesting, intriguing and engaging. It was fun to gather the clues throughout the story and work together the connections. And boy, are there connections. This thing is more interconnected than the World Wide Web. I'm proud that I was able to sleuth some of the plot points but I still didn't figure out the true culprit until the big reveal. As a novice to mysteries, the solution felt a bit out of left field, but not in a way that felt unsatisfying or unfair. And once you had the solution, a lot of the story prior to that just started clicking into places. Even some offhand remarks, made seemingly in jest, fit into the puzzle.
Beyond just the murder mystery, I also enjoyed overall character stories. Most of the early endings don't end you with solving the crime but still had satisfying and emotional finales to them. I may have even shed a tear or two for this perverted detective adventure. That being said, the story isn't all that realistic. You can probably figure that out within minutes when they tell you that you have an artificial eyeball that contains a highly sophisticated artificial intelligence. So there's definitely a degree of sci-fi going on here. However, it also has other very unrealistic elements too for you to suspend your disbelief. You might have a little girl with the power to take down a bear with her bare hands, or have a squadron of gunmen shooting worse than the freaking Stormtroopers. But if this kind of a thing doesn't faze you, there's a fun and interesting story waiting for you.
Most of the game is spent in interactive point-and-click segments where you explore the world and talk to the various characters. I don't think there's really any way to fail these, so it plays more like a visual novel with some leeway as to how many things you want to interact with. You also have some action bits with quick-time events and interrogation events where you need to pick evidence to show to a suspect to reel them in. The core gameplay is quite simple without really anything to particularly like or dislike, although I would've liked to be able to see what objects had new dialogue and which didn't. Could've saved me from clicking on Boss' computer for the ninth time to see if there was dialogue hidden behind it.
The other part of the game, and arguably the most video game part of the whole thing, are the titular Somnium segments. These are puzzle segments where you need to figure out how to unlock the secrets of the target's mind before you run out of time. But counter to this description, these are relatively slow affairs. You have 360 seconds to spare and each action you can take inside the dream takes up an arbitrary amount of time, and idling in place happens in super-slow motion. Running towards a door and kicking it open might be just a second, or it might be closer to a minute, depending on the stage and conditions. And if you run out of time, you'll fail and have to replay through most if not all of it again. These dream sequences are also where the actual branching happens, so you won't need a walkthrough for routing.
I'm a bit of two minds about the Somnium segments. On one hand, they were an interesting and unique aspect of the game, with the occasional moment of hilarity. And while some of the puzzles I could determine by remembering what clues I'd gathered during the normal gameplay and applying them in the dream, some of the aspects required just utterly pure guesswork. How am I supposed to know whether to stick my head or my hand inside a hole? And if you guess wrong and run out of time, you're gonna waste a bunch of time running through the same already-solved puzzles in order to progress through. These puzzles do definitely add to the game, but I just wish they didn't waste as much of my time if my guesses didn't work out. After a fail, it just becomes an exercise of mechanical repetition and holding down the skip button.
The graphics are so-so. The world detail leaves a lot of room for improvement but the character models are mostly good with excellent designs. Some of the facial animations look a bit iffy at times, but I did quite like the overall 3D character look. There's a sense of charm about it. I wasn't necessarily expecting that since I've usually preferred 2D sprites over 3D models in other games. Iris and Mizuki were both very cute in the game. And of course, with everyone being a 3D model, you could also have properly animated cutscenes in the game. Well, cutscene animation was also a bit strange at times too, but it's still a more visual affair than most visual novels.
No complaints on the sound side. Opting for the original Japanese voices, the voice acting was pretty good. Not its greatest asset but still good. The soundtrack was nothing special but it did its job providing the mood.
The technical side however is a bit sketchy. This is quite clearly a console port and has some basic graphics options inside a separate launcher that don't all even work properly. It also forces you to choose between keyboard/mouse and controller input in the launcher, even though you can still use a controller with keyboard/mouse controls. You just don't get the proper button prompts for a controller. Annoyingly it also ignored my choice for Japanese voice acting for the ending movie for reasons unknown. Granted, it never crashed on me or anything, but a bit more polish on the technical front would've been nice.
I would definitely recommend reading through AI: The Somnium Files. Its gameplay won't win it any recognition but story is interesting, rather fun, occasionally touching and most likely will keep you guessing until the end. I'm definitely going to be checking out the sequel, AI: The Somnium Files - Nirvana Initiative, at some point. I imagine I also need to check out other Uchikoshi production too.
I've poured about 400 hours towards Cyberpunk 2077 across all the platforms but I'm still finding it quite hard to write a review of the game. Cyberpunk 2077 has famously been kind of a mess, so why have I dedicated so much of my time towards it? I'm not sure if I know the answer.
I think one of the reasons why I like Cyberpunk 2077 is just the world. Night City is dense and grandiose stage that is essentially one of the main characters of the game with its overwhelming presence. There might be some issues when it comes to AI to make it appear like the city was full of real people going about their real lives akin to recent Grand Theft Auto games, it is still a sublime place to be in. I at least do not remember having as much fun exploring Los Santos in GTA V as I did when exploring the various alleys and roofs of Kabuki. Night City is called The City of Dreams and its design truly embodies it.
What didn't feel that strong was traversing the world though. You can fast travel between points you have already discovered, but I used it sparingly because I actually wanted to spend time in Night City. Sprinting around the city was actually a really valid option for moving short distances and I probably used it more than you'd think. However, for longer distances, you really need some kind of a motorised device. And the car physics in Cyberpunk 2077 are not the greatest. I've definitely encountered worse driving games but this isn't Grand Theft Cyberpunk.
However, I did discover the ultimate way to travel in Night City: the Yaiba Kusanagi CT-3X. While cars have a very funky handling model, zipping around on a motorcycle is actually super fun and the Kusanagi is the best of them. And in addition to being fun to drive, you can use the Kusanagi to go basically anywhere and cut through the Night City congestion. Really the only downside of the Kusanagi is that it's so much fun that it overshadows the rest of the vehicles. Kind of a shame, since the game is full of fun and cool vehicle designs. One of the cars even talks to you!
While exploring that world, it also becomes pretty evident that Cyberpunk 2077 is a part of a larger franchise. There's a shocking amount of lore stuffed into the game. So much so that I imagine that only few players are actually going to go through all of it. But even if you're not going to be engaging with all of it, it still has the effect of making Cyberpunk 2077 feel like it's an actual world and not just shallow set dressing. I've definitely taken an interest in the Cyberpunk franchise after playing through 2077. If only I had some friends to play the TTRPG with.
Another strength of the game is the writing. I feel like a lot of games these days have quite unnatural writing to them. My guess would be that they're trying to emulate the kind of "witty" writing that is present in current Hollywood blockbusters. Thankfully Cyberpunk 2077 doesn't feel like that even though they hired a Hollywood actor to appear in it. The dialogue seems a bit weird or otherworldly when compared to standard English though but I feel like that might be a desired effect. And I'm not referring to the specific 2077 slang words that are used. It would be kinda boring if people living in an alternative future spoke English exactly like people in our regular 2020 did.
I quite enjoyed the cast of characters too. You cannot have a huge influence over the kind of a person V is but I liked her (female V forever). She also meshes well with your main sidekick that's with you for most of the game and I enjoyed the interactions between the two. There's also quite a large array of interesting secondary characters and even tertiary characters have some characterisation. The only characters that I hated, I hated because they were awful people that could only be fixed with a bullet to the skull, not because they were badly written characters. I guess the only issue I had with the characters was that the voice actors were often voicing several characters, so you might encounter a shopkeeper that sounds exactly like your girlfriend.
The main story of the game is good although not necessarily amazing. I also feel like it's relatively short for a single-player roleplaying game. I think a main story only playthrough will only eat about 20 hours of your time. Perhaps it was kept short to give players a chance to finish it since relatively few people ever actually complete games. Personally I think I could've done with additional fleshing of the main story. I'm not sure if the Cyberpunk 2077 story is the most interesting story possible in the Cyberpunk universe, since there is quite a lot of material available, but it was interesting enough to make me want more of it.
However, to complement the relative brevity of the main story, the game has quite a lot of optional content. There's questlines for the main secondary characters, various side missions, gigs by the local fixers, cyberpsychos to subjugate and requests to "keep the peace" (neutralise targets) from the police. And the optional content is much higher quality than many other open world games where the additional content is mainly just collectibles. Some of my favourite missions in the game are in fact optional. Even the police requests, which are the lowest effort side content in the game, still have some amount of setting and variation to them. None of optional content felt like a chore or a lazy collectible like in some other games. In fact, I might take a detour while traveling from one story mission to another to just clear out a couple of NCPD murder requests.
The talent system allows quite personalised ways of playing as V. For my first playthrough, I went for a pretty generic jack of all trades gunner build that didn't really strike me in any way. After that I went for a more stealthy builds, which were much more to my liking. And on my fourth playthrough, I opted for a katana-wielding Sandevistan ninja build, which I really enjoyed. Never really tried out a serious gunner build nor have I tried a netrunner build, as those didn't seem as interesting. Maybe when I decide that it's time to complete Cyberpunk 2077 for a fifth time that I might experiment with those.
The way that you specialise your V alters the core gameplay quite a bit, and since I've never tried out all of the possible ways to play the game, I can't comment on all of them. But I do generally like the gunplay, the melee combat and the stealth mechanics. Even though I never committed properly to be a gunner, I always found it to be satisfying enough. And I quite liked the combination of stealth and hacking abilities to be able to circumvent cameras and people. And the Sandevistan katana build was just visceral catharsis with the way you could overpower your opponents.
I do find that the talent system does contribute to quite an uneven difficulty curve though. I always found the game to start off way harder than it ended, since at the start, you have so much worse gear, so much worse cybernetics and lacked all talents. When playing on the hard difficulty, it was quite easy to get yourself killed if you made any mistakes. And after you'd obtained your equipment and talents towards the end of the game, it was so much harder to die to a point of being a rare event. This is not exactly a rare effect with games built around equipment and talents, but for the sequel, I hope that CDPR flattens out the difficulty curve a bit.
Beyond just trying out different sorts of builds, Cyberpunk 2077 offers pretty good replayability in general. I should know, I've played it through four times already! I'm pretty sure I've discovered something new every time I've played through the game, and not just because they've kept patching the game ever since it launched in 2020. There's always been something new to discover in the environments, some kind of a way to complete a quest that I hadn't tried before, or something completely else. And judging by the stuff I've encountered on YouTube and various websites, there's still stuff that I've never seen before. There's usually a couple of ways to complete any given quest, so you might opt to go for a guns blazing approach on your first try and for a stealthy approach on the second. The various builds also unlock new ways to approach some of the quests.
The game is gorgeous too. I think this is still the apex of gaming graphics after two years of its release, and the system requirements really match that. Cranking the settings to their highest (ignoring path tracing because it's just a technology preview), I was able to get around 80 FPS at 1440p with an RTX 4090. Thankfully you can tweak the settings for a more optimised result on weaker systems, but this is really a game that benefits from beefy hardware. The beautiful graphics really just further enhance the beauty of Night City. Soundtrack of the game is also rather good, and I find myself listening to many of the tracks outside of the game too. Never Fade Away in particular is currently topping the charts on my Last.fm profile. I also loved the voice acting for female V, and really none of the cast really disappointed me with their performance.
The game isn't perfect though, as was quite widely covered when the game launched. Far from it. There were so many different glitches and bugs when the game came out that it was hard to go ten minutes without encountering something. Thankfully there's been over two and a half years of effort put towards the game to polish it, but it's still not quite enough. Luckily the effort has managed to fix the worst game-breaking issues that were present in the beginning. The kind of bugs that would prevent progress or otherwise really make you miserable. But you will definitely still encounter people continuing to walk around after being decapitated or see a car plow through a wall. Bugs that break your immersion but not your progression. I imagine that Cyberpunk 2077 will never truly be jank-free.
It also feels like they've managed to fix worst of the crashing too, since I think I only encountered a single crash on my new gaming PC. With my previous rig, I remember encountering several crashes in a single session. Not sure if that was an issue with the PC/installation I had at the time or if it was just the older versions. Hopefully newcomers to the game can enjoy it with only an occasional crash – since really no AAA game on the PC is truly 100% stable.
I'm not really sure how relevant my review is going to be though. I'm writing this review just a few weeks before the new expansion, Phantom Liberty, comes out and overhauls the game. I haven't actually looked at a detailed list on all of the changes because I want to experience them with fresh eyes once they come out, but the little I've seen online suggests that they're actually doing quite a lot of changes. Hopefully the changes are just for the better and don't ruin anything that I currently love in the game. But since Cyberpunk 2077 has been quite the enjoyable experience for me, I do have a level of trust in CD Projekt Red to not muck it up at this point.
Cyberpunk 2077 is kind of a rough product even after its polish, but one that has always captivated me and kept me playing, even though I still have a hard time explaining why. It's a fun game with a good story and writing, and it'd be hard to not recommend such a game.
Underwhelming spiritual successor to Jet Set Radio.
While I may not be the biggest Jet Set Radio fan out there, I still find it to be a pretty fun action/platform game. And when I saw that someone was going to create a Jet Set Radio style game for the modern age, I did get quite excited. However, having dipped my toes into the game now, I feel rather disappointed.
The movement in Bomb Rush Cyberfunk feels noticeably worse than what Jet Set Radio offered. And I think that's a two-pronged issue, one part being controls and one part being level design. On the control side, I had trouble getting my character to control as I wish. Feels like the characters can cut far too sharp in a direction when coming from a jump, and like it happens in an unpredictable fashion. And the control issues are not helped by the fact that the controls are more intricate than Jet Set Radio's, with many more movement possibilities. I could've taken less if the counterbalance was that it felt more intuitive to move.
And the early levels feel very sparse in comparison to Jet Set Radio with large open areas without any parts for grinding or jumping. In Jet Set Radio, if you saw a ramp going upwards, you could most likely grind against it. But in Bomb Rush Cyberfunk, it feels like I was constantly trying to grind against level elements that just didn't support it. And having to second-guess what parts of the level work for movement and which don't really kills the flow. The levels really didn't feel like they were designed to allow for stylishly combining various tricks to your heart's contents as long as you had the skills. Maybe I'd learn the movement controls in due time but I'm not sure if the level design would ever allow the same kind of free-flowing movement that Jet Set Radio had.
The approach to levels seems very different between the two. Jet Set Radio would throw you into a level with a map of all tag locations and big honking arrows pointing to the tags inside the game. Bomb Rush Cyberfunk meanwhile throws you into a map without any tag locations, no arrows and tells you to go spray enough of them to unlock the next objective. Jet Set Radio was more of a time attack whereas Bomb Rush Cyberfunk seems to be more of an exploration. And I don't think exploration really meshes well with the kind of movement-based gameplay associated with Jet Set Radio.
After I managed to clear the first couple of reputation gates by spraying tags, the gameplay really slowed down as I was now just trying to hunt down any tags I'd missed instead of doing stylish grinds and jumps towards my next objective. The only way to tell apart done and undone tags is with this small shimmering effect on top of the sprays, which is much harder to spot in movement. The minimap would only show you the location of the reputation gate, which would angrily send you back because you hadn't done enough spraying yet. You can't even use the spray button prompt to figure out if you've done a tag, since for some reason you can spray the same tag multiple times. As far as I could tell, there was no benefit in doing so, but it was for some reason still possible.
The spraying itself also makes little sense. Jet Set Radio showed you exactly what kind of joystick movements you should do, and if you screwed up the sequence of moves, you'd lose valuable spray paint and would need to go collect more. Here, you seem to be able to just wiggle the joystick around for about a second and you're done. No sense of accomplishment for a job well done, no punishment for spraying with your eyes closed. At least I was never able to screw up a single tag by just wiggling my joystick around without paying attention. You might as well just have the spraying happen automatically.
Bomb Rush Cyberfunk also differs from its spiritual forefather by having a combat system. A combat system that doesn't really feel fun. You might be thrown into a stage with a handful of mob cops and one flying cop, who will fly around and then deliver you heaven's punishment from above. And the game doesn't really feel like giving you any helping hand to figuring out where this flying cop is relative to your position, making it much harder to dodge the incoming attacks. And trying to take down the flying cop is also painful due to lacking any kind of a lock-on system, which seems like something a game throwing many opponents at you at once should have. And even when you are successfully taking down your opponents, it doesn't feel satisfying. More like just some flailing around. If the combat was completely optional, maybe in the style of Mirror's Edge, I might give it a pass, but as far as I can tell, some combat sections are not optional.
The combat sections also seem to suffer from the same lack of direction as the spraying part of the game. I got thrown into a combat section, and I didn't really know what was my objective. Should I just outrun and escape the cops? Should I beat up every single cop coming my way? Or should I just focus on the flying cop that was in the cutscene? Once again, the only direction that the game seemed to be telling me was "go earn more reputation". Guidance that was not really helpful when I was being hunted by the jetpack police.
The game does very much pull off the Jet Set Radio style. It does very much look like it was just another Jet Set Radio game released on the Dreamcast, for better and for worse. But while the graphics do look old, it is a very functional design. That was probably to be expected considering how well Jet Set Radio has aged. I might not like the levels from a gameplay perspective, I did enjoy them from an aesthetics perspective. Just maybe if they didn't feel so empty. Soundtrack seemed fine too, although it doesn't rival Jet Set Radio in the sound department. Then again, really no game can. Jet Set Radio's soundtrack was a just succession of bangers, whereas Bomb Rush Cyberfunk has good songs and then it has less good songs.
I really would've liked to love this game, but I don't think I can as it is. For the 40-euro price tag, it feels pretty underwhelming and undercooked. Some of the issues I had with the game could probably be improved, but I'm not sure if the core gameplay could be improved to a degree that made this feel like worth its price tag. Perhaps once it hits the bargain bin, I might give it another chance. But right now Bomb Rush Cyberfunk is unable to meet my expectations and I opted to refund my copy of the game instead. Not sure if I've ever opted to do that on Steam before. A shame.
I'd heard that Portal with RTX didn't run great, but I thought that when people said that, it meant that their RTX 3060 couldn't run the game. Thankfully, I have an RTX 4090, so I can just brute force through any bad optimisations. What I didn't expect was that the game runs like dogshit on whatever piece of technology you throw at it.
I launched the game and alt-tabbed before I even loaded in the first stage. The game crashed.
Relaunched the game and cranked the graphical settings. The game crashed.
Started the game for the third time, played through enough to get the half-powered portal gun and a bit more, and then alt-tabbed. The game crashed.
This was all within 10 minutes of playing. I cannot remember if I have ever had a game behave this badly on me before.
My 1% FPS during the start of the game was also around ten. With a 7800X3D and a 4090. I could run maxed out Cyberpunk 2077 without DLSS smoother than this game. If this kind of hardware cannot run this game without massive stuttering, there is surely no gaming PC on this planet that can run this game smoothly.
It's also not decisively better graphically, even though that's the only reason for this thing to exist. The environment lighting does look more natural than the 2007 version, but it still very much looks like a PlayStation 3 era game. They did change some assets but not everything, so you are still seeing 2007 quality, just with improved lighting. I guess the improvements are nice enough that were this a working product that I would do my Portal replay in this version. But considering how busted this version is, it's so not worth it to even try.
Thankfully this is just a free tech demo. Had they charged me even 50 cents for this experience, I would've asked for a refund. The only valid use for this game is to generate screenshots and video clips for Nvidia marketing material. Otherwise it's just an embarrassing and damaging showcase for their RTX Remix.
A first-person game about trying to break out of a time loop.
When DEATHLOOP was first shown to public audiences, I figured that it'd be a shooter game focused around on a time loop. And while it does involve both of those things, it's not really that. It's more of a repeating exploration game, where you go around in circles, collecting clues so that you can solve the temporal puzzle and bust your way out of the titular loop.
After laying down the premise and teaching you your basic abilities in a linear sequence of gameplay, the game becomes very non-linear by giving you some clues and telling you go to figure out how to solve things. It feels a bit daunting at first, and I waffled around exploring unnecessarily a bit before I understood that I could really just ignore everything but the visionary clues and I'd be golden. Once you understand to just follow those, the game doesn't feel like you're just supposed to explore around and take a stab in the dark. Trying to just blindly explore your way out of it is just gonna bury you in information. I still have absolutely no idea what I was supposed to do with some of the non-plot discoveries I made.
The story isn't really much anything. There's some exposition about what happened previously to begin the loop, but it's not really important and the game doesn't put much effort into it. Well, that or I just couldn't find the scraps of paper and tape recorders that explain the backstory. The few real revelations of the story also took place right before the end for me, since I guess I just played the non-linear story in that order. The meat of the game's story is really just the clues themselves, and not much anything. Granted, the overall puzzle is an enjoyable mystery but it's not a deep and involved plot. The ending is kinda weird and abrupt too.
The cast of characters is mainly you and the villains that you need to kill, with one villain being particularly chatty for plot reasons. The less chatty villains are basically all just massive unlikable narcissists for you to hate. This is probably for the best, since if you actually plan on finishing the game, you'll probably have a hard time counting how many times you've killed them. I don't want to claim that they're completely shallow or undeveloped, but quite a lot of them never really registered to me as actual people. More like objectives with legs. And the only non-villain character left over is your player character, who is actually a pretty distinct character and all around fun fella with a penchant for talking to himself.
For exploration, the main activity of the game, there's four different maps and four different times of day when you can go into them. And while the maps themselves don't really change, the intel you can gather out of them changes with the time of day. I don't think a fresh player can beat this game without coming and going to every single map so much that they should just install a revolving door. And I can't say that I'm really a fan of this. It just kinda feels like lazy level design to make us pour through the same four, relatively small maps with a fine comb. The maps really just aren't that big so you'll definitely get intimate with them in 5–10 hours of gameplay.
The gunplay in DEATHLOOP doesn't feel amazing. It's fine but not really anything more of that. I think that's because the game actually wants you to stealth through everything. The game does actually suggest to you that you're free to go through the maps as you please, including the possibility of going guns blazing, but the way things blow up in your face if you use firearms suggests that they would really prefer you to sneak around instead. And with the gunplay being what it is, it never really felt like "time for action, baby" as much as it felt like "oh, I fucked up, can I get out of this with these guns and the couple of bullets left in my pocket?" Also felt kinda weird that the magic powers you can get in the game seem very combat-focused when that's really second fiddle to stealth.
Much like the gunplay, stealth gameplay is "fine". It's not the best nor the worst stealth experience I've had in a game. My chance to go by undetected did get significantly better when I got access to the silenced SMG and I actually had the possibility of deleting enemies without groping their butt. Actually, the silenced SMG is so good that a more pro-stealth player might consider it overpowered, since it allows for insta-kills via precise headshots, and if you fuck up, silent rapid fire onto an enemy until it stops moving. Supposedly there were other silenced weapons to also find, but I didn't really need anything more than the SMG.
What actually felt a lot more satisfying than sneaking around was taking the high ground with a sniper rifle and taking potshots at the enemies from the distance. Position yourself far away enough and they won't be able to hear the gunshots, leaving them unable to hunt you down and helpless to protect their tender little heads. It also feels more in character as a big burly black ex-military dude to snipe down your enemies than try to sneak your way around them.
Occasionally the chatty villain also decides to invade the map you're in. The first time this happened, I flew into panic since I had no idea where I was getting shot at and had a hard time fighting her off. Pretty quickly after that I started to figure out her patterns and it became more of an annoyance than a massive threat. I even managed to sneak around her with invisibility, turn off my invisibility for a bit to lure her in, then go invisible again to knife her in the back for some sweet insta-kill revenge. You also need to go do some busywork to unlock your escape route after this invasion, so that really just doubled the annoyance of it all. Not really a big fan of this particular feature.
There's also a multiplayer aspect where you can play as the invader, but I never touched it. I want to play a single-player campaign that I can put on pause if I needed to go pee. Thankfully opting out of the multiplayer aspect is very much possible, although I believe it did automatically start off in online mode.
Possibly my favourite part of the game is just the presentation. DEATHLOOP is dripping with this cartoony 1960s vibe and it's fantastic. The sound design is also great. The stealth music is very good at building a tense atmosphere, and the shift in music when you go from stealth to combat, or from combat to stealth, also works well to indicate your status, so it also works from a gameplay perspective. Colt's voice acting is also good, although I'm not so sure about the rest of the voice acting. It at least didn't feel as strong. But the overall presentation of the game is still to its benefit.
Technically the game seems sound. I get pretty low system utilisation when running max settings at 1440p with a 240 FPS cap, so I assume that this game runs pretty fine on low-end hardware as well. Not a lot of bugs either, although I did manage to softlock several times in the inventory screen, and a couple of times an enemy body ragdolled wildly into space after I've killed them. I also read somewhere that the loading times were tied to your FPS. Can't really confirm that claim, but if that's true, then it sounds kinda bad. Thankfully my loading times were pretty decent with a 7800X3D, an RTX 4090 and a fast SSD. Overall, I don't have a lot of things to gripe about the technical side – a lot of games are put out in far worse state.
DEATHLOOP is kind of a mixed bag for me. The story didn't feel that special and the gameplay didn't feel that special either. The time loop mechanic, even though it was an interesting concept, didn't really make the game feel that special either. But despite none of the aspects of the game really grabbing me, I still felt compelled to pour about 23 hours into the game to solve the puzzle and reach the ending, so I guess DEATHLOOP was doing something right. I'd give the game a cautious recommendation if you enjoy exploration and stealth, but I wouldn't say that it's a must play title.
The Total Eclipse anime was actually my first experience with the entire Muv-Luv universe – which I imagine was the case for a lot of current anime-watching Muv-Luv fans. I actually don't remember anything about it other than the first two episodes (which are not even covered in the VN). And while it has been eleven years since the anime premiered, having a forgettable story is not exactly a good sign for the VN. Then again, this meant that I came to the visual novel version of Total Eclipse with pretty fresh eyes.
The visual novel story is entirely kinetic and feels a bit meandering, with the focus switching between political intrigue, life or death action, or high school romantic comedy, with a good chunk of the female cast competing who can have the most obnoxious teenage crush on the protagonist. Political intrigue across the major powers of the Alternative timeline does take the driver's seat for most of the story though. Even the action scenes are drenched in geopolitics.
Ostensibly the main objective of the story is bringing the XFJ Program to completion but it doesn't really feel like it. Rather, it feels like the story is about whatever is the current crisis that is currently unfolding and which may or may not relate to the XFJ program. I'm also not sold on the ending as it felt like the villain had better motives than the protagonist. Of course, the ending converges to work out somehow at the end because the writers didn't want to write a story where the protagonist does dumb things and meets a dumb end.
There's also some questions raised during the story that were not really ever answered. Perhaps the idea was to leave room to further explore some of the factions in the future. But since Total Eclipse as a franchise is already over 15 years old, I doubt we'll ever get some kind of a satisfactory explanation to some stuff. At the moment it just feels like they're introducing unknown characters of unknown factions with unknown motives and you just have to deal with it.
In general though, I felt at least engaged with the story overall and even actually captivated during certain parts of the story. And despite the ending involving some questionable developments, it does thankfully hit some good emotional tones and is ultimately something I felt somewhat satisfied with.
To anyone interested in Total Eclipse, I'd say that the requirement for actually enjoying the experience is a general interest in the Muv-Luv Alternative universe – the story takes place just before the events of Muv-Luv Alternative and only expands on a tiny part of the Alternative story. Standing on its own two feet, it's not really a strong story, making the decision of making Total Eclipse the first animated Muv-Luv story questionable. But as an additional story in the Muv-Luv Alternative universe for Muv-Luv fans, it's decent enough to recommend.
I did notice that the character arc for a lot of the characters followed the same pattern. They introduce a character as supremely unlikeable, throw them into the ringer for a bit and then have them calm down and become somewhere between tolerable and likeable. I noticed that they did it for the protagonist, Takamura, Cryska, the Chinese girl, and the other Japanese-American test pilot. Thankfully the protagonist and Takamura calm down relatively fast, otherwise the experience might've been quite a slog. I think they might've even tried doing it with Tarisa but she's far too adorable to be unlikeable. But despite the repeating character development pattern, the characters of Total Eclipse are ultimately likeable and interesting.
Presentation in Total Eclipse is the best that Muv-Luv has yet to offer. I did initially feel a bit weirded out by the colour palette, as it's very muted in comparison to many other entries in the series. However, after getting used to it, I can say that the art is very good. Both the sprites and CGs are high quality, and there's a good amount of CGs to boot with 195 entries in the album. Only very occasionally I saw something that looked weird or low quality.
They're also utilising a good amount of animation for the sprites in order to wring out a bit more life out of static images. And for the first time ever, Total Eclipse also utilises a decent amount of actual 3D animated cutscenes during the story. While they managed to give a good kinetic feeling to the mech action in Alternative, it's completely different to see TSF manoeuvre when it's given a fully animated short cutscene. The only downside of the 3D cutscenes is that there's not that many of them and it feels like they mostly happen during the first half. Hopefully the level of animation shown in Total Eclipse is the new standard for new Muv-Luv stories, as I am quite a fan of it.
On the sound side, there's not that much innovation. Yuuya is kind of a weird protagonist in that he's mostly not voiced but still is given a voice from time to time. At first I thought they only voiced him during flashbacks and when the story is not from his point-of-view, but later in the story, his lines are sometimes voiced and sometimes not. Not really sure the rationale for this, as THE DAY AFTER had Hibiki fully voiced. At least it's not super distracting. Otherwise the cast and voices are solid even if I can't really find anything special to say about it. What I did particularly enjoy however was the usage of insert songs during some bigger action events. I'm a huge sucker for insert songs for action scenes.
No issues with general ergonomics and display. A much better reading experience than with the old rUGP titles. I read through Total Eclipse exclusively on my Steam Deck and the performance was what I've come to expect from titles using the Ages Mark 2 engine, which was mostly solid. Like with the other games, my Steam Deck exhibited the same kind of audio crackling after resuming the game from sleep. And even though this is a Steam Deck verified game, there's still no official controller layout.
I finally however decide to test the Pause Games plugin and it seemed to make the whole Steam Deck audio experience a lot more solid. After installing the plugin and enabling the "pause before suspend" feature, it felt like I could resume Total Eclipse from sleep a lot better than previously without having massive and continuous audio crackling. Highly recommend this tweak to anyone experiencing audio crackling in Ages Mark 2 games. Wish I'd tried it back when I read through THE DAY AFTER series.
All in all, Total Eclipse is a decent 25-hour delve into the geopolitical and weapon development politics of the Muv-Luv Alternative world, with occasional hot-blooded action to balance things out. Just don't make the same mistake âge did and try to make it your introduction to the Muv-Luv world.
I've had my Steam Deck for six months now and I am surprised at just how transformative of a device it has been for me.
When the Steam Deck was first announced, I was very sceptical of it. Back in August 2021, I remarked in a conversation how I only saw the use case for the Steam Deck in emulators and visual novels. And in January 2022, I commented how I wasn't interested in buying one. To be fair, my reasoning for not buying one was pretty sound: I was almost never away from home where I keep my gaming PC, and I already had a Nintendo Switch and two PlayStation Vita. But in December 2022, I changed my mind: I thought it seemed neat, especially for visual novels. And about a month later, I picked up a second-hand 256 GB model for 400€.
Now six months later, I absolutely adore my little Steam Deck. It's a really cool gaming device. Unfortunately due to Steam breaking the hours per platform tracking that SteamDB had, I can no longer see how many hours I've played on Linux (Steam Deck being my only Linux gaming PC), but my current estimate is that I've spent over 120 hours playing on the Steam Deck. Around 20 hours per month is not nothing for me, especially since I'm rarely more than five kilometres away from home / my gaming PC.
UPDATE: Turns out that if you register your own Steam API key, you can fetch your own Linux-specific Steam playtime. My total Steam Deck (or rather Linux) playtime is currently 131 hours!
It's actually extremely easy to forget that the Steam Deck is a weird, specialised gaming PC that is running on top of modified Arch Linux even if almost all of my Steam library is just Windows games. You can see that a lot of work has gone into Proton to get it work as flawlessly as it does. It's of course not perfect, but it is downright sublime all things considered.
So far on my Deck, I've read through six Muv-Luv visual novels (with the seventh currently underway), played through Bastion and Jet Set Radio, tried out a bit of Forza Horizon 5, Grand Theft Auto V and San Andreas, and put many hours towards Grand Theft Auto IV, Tokyo Xanadu eX+ and Vampire Survivors. Not everything was a perfect fit on the Steam Deck, and not everything worked completely right, but I still feel extremely satisfied with what I've managed to get done with a Linux-based handheld gaming PC. Haven't managed to try emulation yet though since I've never been huge into emulation, but maybe I'll take a stab at it at some point as well. I've at least heard the emulation experience is quite good.
It's also quite easy to juggle a lot of games on the Steam Deck. Even though I had the 256 GB model, I immediately spruced it up with a microSD card. Flash storage has gotten so cheap that slapping an extra 512 GB of storage to my Steam Deck only cost me 38.10€. I actually had to check my receipts and discovered that's about the same that I paid for my 16 GB Vita memory card nine years ago. And now, thanks to my 685.7 GB of usable storage, I have 43 games installed on my Steam Deck with 115.7 GB left over. That's a lot of game library for such a small device.
I've also been pleasantly surprised at how manageable of a device the Steam Deck is. On paper, it seemed absolutely massive compared to my Nintendo Switch and PlayStation Vitas. But when I held it up for the first time, it didn't actually feel that massive or heavy. I was expecting it to feel like a brick but I think the weight is balanced well with its size, so it doesn't feel particularly dense. Navigation on the Steam Deck is also well-implemented, with its touchscreen and trackpads, the latter of which is particularly nice for the occasional text entry. While I haven't done any massive marathon sessions, I don't really have any issues with the ergonomics.
The Steam Deck has been so excellent that I think it has managed to make PC my primary gaming platform for the first time in a decade.
Back when the PlayStation 4 and the PlayStation Vita were new products, Sony touted the benefits of their ecosystem by promising that you could buy a digital copy of a game for both your PS4 and your Vita just once with Cross-Buy, and then sync your progress between the home console and the handheld console with Cross-Save. Unfortunately third-party support for these features, and the PlayStation Vita as a whole, never really took off, and Sony never had the fortitude to carry on with the experiment. Well, unless they manage to surprise with the upcoming Project Q.
However, it feels like Valve has managed to do what Sony couldn't. Since the Steam Deck is just a small PC and has direct access to every single game in my Steam Library, I can now buy a PC game just once and have it available both at home and abroad. A lot of games these days also support Steam Cloud, so I can realistically carry on where I left off on the other device – at least if the Steam Cloud support is properly implemented (I've seen it not be).
I think the biggest proof of my transformation is the fact that ever since January, I have not picked up a single game for any of my consoles. Contrast this to my estimate of having spent approximately $3800 on console games between January 2021 and January 2023. Ever since the Steam Deck, my interest in my gaming consoles has dropped off significantly, and my interest in PC gaming has increased dramatically. My Nintendo Switch might offer the portability of my Steam Deck and my PlayStation 5 might match my gaming PC in delivering high-fidelity experiences, but neither can offer me both. But PC gaming / Steam definitely now can.
I'm in fact currently waiting for a bunch of (very expensive) gaming computer parts to arrive, hoping to have a brand-new rig ready for my summer holiday. If I'm lucky, I'll have my hands on those tomorrow. Sort of hard to argue that I'm not a PC gamer at this point and it's at least partially my Steam Deck's fault.
I have to give it to Valve's business sense – releasing the Steam Deck, a good and affordable portable gaming PC, is probably a massive boon for their bottom line. I've kept track of how much money I've spent on PC games this year and the total comes to 572.88€. 451.59€ of that is directly within Steam and 121.29€ on third-party sites like Humble and Fanatical. Not sure what cut Valve gets from third-party sites that deliver games as Steam activation keys, but with a 30% cut inside their own store, they've made at least 130€ off my 2023 Steam purchases alone.
I'm actually pretty satisfied for having spent less than 600€ in about a half a year though. Compared to my previous console software spend, this is much cheaper. Possibly because I always opted for a physical console copy and that plastic ain't cheap. Maybe at some point my software savings will offset the huge amount of money that I'm throwing at the hardware.
And since Steam is the first-party store on this device, I'm definitely avoiding all stores that are not Steam or don't give out Steam keys. You can probably get GOG and Epic Games Store working on this device but I really can't be bothered to. Not when the first-party experience is so good. So far I've picked up Cyberpunk 2077 for -50% on Steam after having bought it in 2021 from GOG, and despite Epic Games Store giving out free copies of Control and Death Stranding, I bought a 14.12€ eight-game bundle mainly for Control and paid 9.99€ for a Humble Choice for Death Stranding Director's Cut. I'm spending money that I didn't have to, just to have my gaming life consolidated on this one platform as much as possible. This is some truly next-level evil genius financial brilliance on Valve's part.
But even though I've showered praise on Valve for the Steam Deck, there are still a couple of things about the Steam Deck that I don't like.
What might be the most annoying part of using the Steam Deck portably for me is the fact that the display polariser is 90 degrees wrong. This means that it's completely impossible to use the Steam Deck while wearing sunglasses and I like to protect my eyes when I drag myself outdoors. If I'm out and about, for example sitting in the commuter train, and I want to play my Steam Deck, I'd need to find some place where to put my sunglasses for the duration. That sounds like a great way to either lose or destroy my sunglasses, which is why I opted to read a book instead during my last train ride. Apple has managed to make my iPhone work while wearing sunglasses, so I'm sure Valve could fix it too.
The battery calibration also feels off on my device. At times it has sat at 99% charge for the longest time, and then started rapidly discharging afterwards. I've also had the device pop a low battery warning for 10% remaining, followed by the Steam Deck shutting itself off within 10 seconds. Afterwards it never charged the battery when I connected it to a power cord. It just stayed at 0% for over an hour and didn't start charging until I reconnected the power cable, forcibly shutting off the device in the process. Thankfully my Deck is never too far away from a power source in my usage so I've never suffered too badly, but this sort of a thing is still an annoying paper cut in the ownership experience.
And whilst this isn't really a problem caused by the Steam Deck or Valve, I've also been annoyed when it has come to patching games. Some visual novels really need translation patches for a proper experience, but the Wine-based game install environment doesn't really facilitate that. Attempting to patch my copy of CHAOS;HEAD NOAH on the Steam Deck was such a pain in the neck that I think I'm just going to try copying over the patched game installation from my gaming PC to my Deck. Maybe in the future these patches will be made with considerations for the Steam Deck, but right now it's an annoyance.
But still, for a 400-euro weird little first-generation gaming thingy, the Steam Deck is a hoot. I don't have any regrets about buying mine and I'd recommend it to anyone even remotely curious about getting one. During the latest Steam seasonal sale, the barrier of entry was even at just 377.10€, so the financial investment isn't too bad. Most likely you won't regret it and even if you do, you can probably load it off on the used market. Some weirdos do in fact buy used Steam Decks.
I recently ordered some manga from Book Depository right before they shut down all operations (RIP). In fact, I got five total volumes of My Dress-Up Darling in English, as the impending closure of the site prompted some good discounts: on average I paid 13.37 € per volume. On other online book retailers, each volume was closer to 17.50 €, and my previous Book Depository orders for the same series last year were also closer to 17 € than they were to 13 €.
My Dress-Up Darling is great. It's funny, it's cute, it's romantic, it's well drawn. It's just all around great. Whenever I start reading a volume, I just gobble it up like a greedy pig. So much so that I decided to take out a timer and see how long it took me to finish the fifth volume.
It took me just 35 minutes to finish the volume.
If I use this amount of time with my favourite way of evaluating entertainment, euro per hours of enjoyment, My Dress-Up Darling is about 23 €/h even with the cheap pricing I got with Book Depository. With the more expensive prices on other sites, it'd be closer to 30 €/h. I love the manga but this still seems like a pretty awful money-to-time ratio.
With this in mind, I decided to go through the list of games I've completed lately to get a comparison for the amount of entertainment I've sucked out of those games compared to the amount of money I paid for them.
Game
Price
Hours played
Price/hour
Bastion
0.00 €¹
17.7 h
0.00 €/h
Ghostwire: Tokyo
0.00 €¹
39.4 h
0.00 €/h
Jet Set Radio
1.19 €
4.5 h
0.26 €/h
Muv-Luv photonflowers*
10.49 €
10.1 h
1.04 €/h
Muv-Luv photonmelodies♮
12.49 €
29.1 h
0.43 €/h
NEEDY GIRL OVERDOSE
9.30 €
11.2 h
0.83 €/h
Synergia
~2.23 €²
3.7 h
0.60 €/h
THE DAY AFTER 00
8.99 €
4.5 h
2.00 €/h
THE DAY AFTER 01
15.11 €
8.2 h
1.84 €/h
THE DAY AFTER 02
15.11 €
6.7 h
2.26 €/h
THE DAY AFTER 03
15.11 €
7.7 h
1.96 €/h
Gift
Bought in a bundle
All of my recently finished games absolutely obliterate My Dress-Up Darling when it comes to the amount of entertainment for the price. However, they're all quite clearly old and cheap games. New games are expensive! Often going for $70 at the time as they're released!
But what if they were too? Well…
Game
Price/hour ($70)
Bastion
$3.96/h
Ghostwire: Tokyo
$1.78/h
Jet Set Radio
$15.56/h
Muv-Luv photonflowers*
$6.93/h
Muv-Luv photonmelodies♮
$2.41/h
NEEDY GIRL OVERDOSE
$6.25/h
Synergia
$18.92/h
THE DAY AFTER 00
$15.56/h
THE DAY AFTER 01
$8.54/h
THE DAY AFTER 02
$10.45/h
THE DAY AFTER 03
$9.09/h
At $70 per game, absolutely every single one of them becomes much worse value than what I actually paid for them. But even if we assume naive $1 = 1€ conversion, all of them are still cheaper for the amount of time I got out of them than that single volume of My Dress-Up Darling. Even Synergia, which I think is not worth the money or the time.
I also have a bunch of unfinished games that I've dabbled with during the year. Let's see how they hold up in reality and against the $70 benchmark:
Game
Price
Hours played
Price/hour
Price/hour ($70)
CHAOS;HEAD NOAH
6.29 €
2.8 h
2.25 €/h
$25.00/h
Cyberpunk 2077 (Steam)
29.99 €
30.8 h
0.97 €/h
$2.27/h
Mass Effect Legendary Edition
49.79 €
31.3 h
1.59 €/h
$2.24/h
Mirror's Edge Catalyst
4.99 €
6.1 h
0.82 €/h
$11.48/h
Muv-Luv Alternative Total Eclipse
23.79 €
5.3 h
4.49 €/h
$13.21/h
Tokyo Xanadu eX+
~2.23 €¹
7.9 h
0.28 €/h
$8.86/h
Vampire Survivors
3.99 €
14.8 h
0.27 €/h
$4.73/h
Bought in a bundle
Those again feel pretty reasonable when it comes to value. CHAOS;HEAD NOAH, a visual novel that I've only read for a couple hours so far, is the worst performer on the list and fares quite badly against the $70 benchmark. And yet, they still all fare better than my manga. And according to HowLongToBeat, I can expect 20+ hours from CHAOS;HEAD NOAH even with my speedreading, bringing the benchmark value in the $3.50/h range (0.31 €/h real value). So unless I drop any of these titles, the cost-per-hour is only going to become more and more reasonable.
Granted, books do come with one big advantage: they don't require any additional purchases. It's a fully self-contained product. However, I feel like my manga collection still fares quite poorly value-wise even when accounting for the hardware. The last number I have for my Steam Deck playtime was around 110 hours. Having paid 400 € for mine, that translates to roughly 3.64 €/h. And that's with under half a year of ownership with a recluse that spends around 23 hours of their day at home. I imagine by the time that I replace the device, especially if I can still recoup some of its purchase price on the used market, that the final cost of ownership will become a rounding error. As long as I just don't break it.
I also recently calculated that were I to throw my gaming PC into the lake right now, it would have had a price of 2.60 €/hour of gaming. And if I were able to sell it for half what I paid for it (which seems like a reasonable expectation looking at prices for used gaming PCs) then it'd drop to 1.30 €/h. I'm also one of those weirdos that has a completely dedicated PC for gaming, as my primary computer of choice is a Mac. I imagine most people gaming on PC do actually use their gaming PC as their primary computer anyways, so the value of the hardware over time becomes negligible.
One additional piece of media that we can benchmark games against is movies. Finnkino movie tickets seem to range somewhere between 14.90 € and 22.00 €, and a movie is probably somewhere between 90 and 150 minutes in runtime. That gives us a range of 5.96 to 14.67 €/h. So had I paid $70 for my copy of NEEDY GIRL OVERDOSE, I'd still be looking at the cheaper end of the value spectrum when it comes to movie tickets. And Ghostwire: Tokyo pricing might not even cover the parking.
And the thing is: it's very easy to not pay $70 for a game. Not a single game that I've listed here was bought for anywhere near $70. In my opinion, $70 is mostly just a tax that is levied against the impatient who must get the latest AAA game right at launch. Games go frequently on discount, and if you're a PC gamer, you have a pretty decent selection of storefronts to shop in.
To quantify this in data, I arbitrarily chose some bigger releases from the past year and checked out what kind of savings are/were possible for them if you just waited for a bit using gg.deals (great website btw).
Game
Launch price
Current price
Historical low
Price drop
Atomic Heart
59.99 €
39.59 €
39.59 €
34%
Call of Duty: Modern Warfare II
69.99 €
38.49 €
38.49 €
45%
Dead Island 2
59.99 €
43.79 €
35.99 €
27% (40%)
Dead Space remake
59.99 €
41.99 €
35.99 €
30% (40%)
Diablo IV
69.99 €
69.99 €
69.99 €
0%
Forspoken
79.99 €
39.99 €
39.99 €
50%
Hogwarts Legacy
59.99 €
40.65 €
40.65 €
32%
Marvel's Spider-Man Remastered
59.99 €
30.43 €
29.73 €
49% (50%)
Redfall
69.99 €
55.74 €
52.49 €
20% (25%)
Resident Evil 4 remake
59.99 €
44.39 €
38.69 €
26% (36%)
Returnal
59.99 €
37.37 €
35.99 €
38% (40%)
Star Wars Jedi: Survivor
69.99 €
52.49 €
52.49 €
25%
The Callisto Protocol
59.99 €
29.99 €
26.99 €
50% (55%)
The Last of Us Part I
59.99 €
46.21 €
44.99 €
23% (25%)
Looking at this data, I'd say it becomes quite clear that games do get pretty decent discounts within just a year of coming out. The only game on my list of recent major titles that wasn't discounted was Diablo IV and that came out 25 days ago. Of course, that could also just a sign that the last year has had quite a lot of bad games (hello Forspoken and Redfall) and botched releases (looking at you, Star Wars Jedi: Survivor and The Last of Us Part I). But even the universally acclaimed Elden Ring (released 16 months ago) can be bought for 40% off its launch price right now. Just chip away at your backlog for some months and you can probably grab the next AAA release for a good discount.
As far as I'm concerned, gaming is still an incredible value for money, even with the new $70 price tag for AAA releases. Just as long as you just play the damn games.
Collecting the dues
There is however something that I find to be incredibly bad value: collector's editions.
For example, a copy of Tears of the Kingdom is currently available on a Finnish retailer's site for around 63 € but if you want the collector's edition, you need to pay 160 €. That is 97 € more for the collector's edition compared to the base game. Sure, Tears of the Kingdom might be a blockbuster that offers somewhere between 50 and 200 hours for a playthrough, but you can get that experience for just 63 €. So what does the 154% collector's edition markup give you?
A 204-page hardcover artbook, a steelbook, a poster and four pins.
Personally, I'm not really a fan of steelbooks, posters and pins. I think steelbooks look awkward among non-steelbooks on the shelf and they're not really even that valuable. You got a free steelbook upgrade if you pre-ordered Persona and got the launch day version. Not sure how many people like to hang posters but I'm definitely not one of them. And sure, this might be a premium steel poster, but how much can that be worth? It's not even that big. And do people outside of school children decorating their bags have a use for pins?
The only thing that strikes me as desirable is the artbook, but even then this collector's edition seems like bad value. I actually bought the Persona 5 artbook when it came out and it's very thick at 432 pages. And how much did I pay for it back in 2017? Around 24 €. And the Zelda one is half the size for an almost 100 € upsell. It's closer to what I got with Persona 4 Official Design Works, which was 192 pages for 34 €, and that's probably with a reasonable markup since I bought it six years after release. (These days the English version of the Persona 4 art book seems to be fetching over 100€.)
To me it seems like collector's editions are a way for game publishers to increase profit margins on games as base games have not kept up with inflation. The standard game price jumped to $60 during the PlayStation 3 era, and $60 in 2007 is worth about the same as $90 in 2023. Of course, they do need to make a couple more trinkets to stuff in those boxes for them to sell, but I've rarely seen a worthwhile collector's edition, at least one that doesn't cost an arm and a leg.
Now I have to admit that I have actually bought a collector's edition for a game in the recent years. It was the Cyberpunk 2077 Collector's Edition for the PS4. It does come with quite a lot of stuff, although the value of some of them can be debated. Inside the massive box, you get the game, a steelbook, a world compendium, some postcards, a map, a collection of stickers, a metal keychain, some metal pins, a guide leaflet, embroidered patches, a 216-page hardcover artbook and a large figurine of male V on a motorcycle.
I'm no expert on collector's editions but I reckon the Cyberpunk 2077 one includes a lot more stuff than your average collector's edition. It's somewhat debatable whether or not that stuff is good but it certainly has it. However, it's also quite a lot more expensive than your average collector's edition at 249 €. So why on earth did I get it? Well, because I got it for 79.95 € during a Black Friday sale. I guess the launch reception of the game really had an effect on its retail price. And since the game collector in me wanted to pick up a disc version of the atrocious 1.00 version of Cyberpunk 2077 anyways, I figured I might as well get the artbook, a cool keychain and a figurine at the same time. The base game hovered around 20 € at the time so it was really only 60 € more for those. What a bargain! Or so I tell myself at least.
Still, I have limited love for these expensive collections of trinkets. I think I'm going to keep avoiding collector's editions and instead put my money towards standard versions of games – which I'll continue to mostly grab from various sales.
Time for the second Muv-Luv side story collection. This time with three stories: Adoration, Resurrection and Altered Fable.
Adoration
Adoration is described as "a rare glimpse into what the world of Muv-Luv Alternative is like outside of Japan", taking place in Dover, England and focusing around the Cerberus battalion. However, it is not a hardboiled European military action as you might believe from the summary. Rather, it's a lighthearted comedy starring Makabe Seijuro, a man with a hundred elder brothers and an off-brand Lelouch.
That being said, Adoration is a fun read. I found myself grinning quite a bit when the serious Royal Guard cadet faces absurd situations with the German troops. Big-time Alternative fans looking for a serious character drama might not enjoy their stay, but if you enjoyed Extra, you might enjoy the adventures of Cadet Makabe. Budget Lelouch also encounters lucky pervert situations at a pace that might make Shirogane jealous, so you have some ecchi action to entertain you as well.
Nonetheless, there are actually serious moments of character development as well. It might be a fairly brief read, clocking at around four hours to completion, but Makabe does in fact grow as part of the story. And the rest of the cast, while not experiencing the same kind of a character arc, are extremely likable. Even the one "asshole" of the story is still a likable dude.
The writer also seems to have exercise every opportunity to cram in references to other in vogue series. I'm not calling Makabe an off-brand Lelouch for nothing.
Adoration is approximately four hours long, which is fairly good for a side story.
Resurrection
From the beginning of Adoration, I figured that it'd be a fairly serious and somber story, much akin to Rain Dancers in photonflowers*.
It's definitely not that. It's in fact probably the most moronic that Muv-Luv has ever gotten. It's exceedingly absurd and silly.
That being said, Resurrection is not a bad read. As long as you are not expecting a story as serious as the beginning would suggest, you'll probably find some amusement in it. However, it is very hard to take seriously – and the story tries to have some serious parts to it, which are pretty hard to digest given how stupid everything that became before it was.
Resurrection also features very good art and animation. Kozuki and the Valkyries play a pretty big part in the story, and they all get new sprites that I found vastly superior to their old ones. And animating the sprites when it came to stuff like throwing punches is definitely competing for the top places when it comes to Muv-Luv stories. I think the only other Muv-Luv story that might rival it is Total Eclipse (which I've sampled for a couple of hours in between reading photonmelodies♮).
Resurrection was approximately seven and a half hours long, so a good chunk longer than Adoration.
Altered Fable
Altered Fable is the meat and potatoes of the entire package, taking up over 12 hours of my time to finish. But while it might be the main dish of photonmelodies♮, it feels more like a dessert for Muv-Luv fans. And while it's the direct sequel to Muv-Luv Alternative (at least until Muv-Luv INTEGRATE comes out to muddy the waters), there's not really a central plot or discernible character development. It's more like a bunch of random events strung together. Definitely not required reading, but I wouldn't expect that from a side story collection anyways.
That is not to say that there's zero plot. There is an underlying story but it really only surfaces every once in a while with a conclusion at the end. Otherwise it's just random Extraverse romps, but now with a much larger cast than you had in Extra. At least that underlying story does explain a lot of events that happen during Altered Fable that otherwise seem like really convenient coincidences to tie it into Alternative.
Altered Fable also has pretty intricate routing for a side story, even though there's not really routes per se. The main story is fixed, and you can really only affect what small events play during it and which of the girls you get to spend time with. Personally, I would've loved a full on Kashiwagi route but I'll take the little scraps I get here.
Going for 100% completion was kind of a pain though, and took me over nine hours of additional reading/skipping, as there's quite a bit of additional content beyond a single playhrough and it's not really clear what choices unlock what. And while I couldn't find a route flowchart, there is thankfully a walkthrough. Trying to get the full gallery unlocked without it was a fool's errand.
But ignoring the pain of being a completionist, Altered Fable was definitely a treat. A moment of fan service to sunset the adventures of Takeru. Just don't take it too seriously.
photonmelodies♮
While photonmelodies♮ features less stories than the previous photonflowers*, it feels like a tighter package. Many of the stories in photonflowers* were very short, and you couldn't really get into the stories in such a time. In contrast the shortest story in photonmelodies♮ is a respectable four hours, and the total package has at least 20-30 hours of content.
Technically photonmelodies♮ is safe and sound. While I prefer the Ages Mark 2 engine (used in THE DAY AFTER and Total Eclipse remasters) for its text rendering and some of its options, the venerable rUGP performs fine and runs basically perfect on the Steam Deck. In fact, it doesn't even suffer from the audio crackling bug that I've experienced with the Ages Mark 2 titles.
Muv-Luv photonmelodies♮ is a pretty easy recommendation to fans of Muv-Luv, and especially for those that enjoyed photonflowers*.
The worst graphics card I've had the displeasure of owning.
Now, I'm not going to lie: I'm no graphics card expert. Before the RX 7900 XTX, the three graphics card that I've owned were the Nvidia RTX 3060, the Nvidia GTX 760 and the Nvidia GeForce 9800 GT. If I don't misremember, those were bought in 2021, 2014 and 2008 respectively. Having only had experience with four different GPUs over a span of 15 years definitely doesn't give me the credentials to review a graphics card properly.
However, I can still say that the AMD RX 7900 XTX is definitely the worst card I have ever owned. Why? Because it just doesn't work.
May
I bought my 7900 XTX at the beginning of May. Thankfully AMD.com uses DHL, my absolute favourite courier service in Finland, and I managed to get my grubby little hands on my GPU in mere five days after placing the order. After I got done with work, I painstakingly replaced the RTX 3060 inside my gaming PC (sandwich-style PC cases are not easy to work in), booted up and installed drivers. And then, I decided to boot up the most graphically intensive game I have: Cyberpunk 2077.
Turns out that the 7900 XTX could really only muster around 40 frames per second with the ray tracing settings I had. I would have expected a little more for a 1100-euro investment. But opening up the AMD performance overlay showed a culprit: the GPU junction temperature was at 110°C.
Well, nothing to do about it but contact AMD support. To their credit, they did immediately acknowledge that these products have a fault, although they seemed to be downplaying it to a degree ("may impact the performance but the card can still be used"). Nevertheless, I asked them to generate me an RMA so I could get my card switched for a working one. Then about a week passed before I got my shipping label because AMD support was kinda slow. Maybe they are dealing with a big influx of unhappy 7900 XTX owners?
Since my assumption was that it'd only take a couple of weeks to get a card back, and since working inside the mini-ITX case gets on my nerves, I decided that I wouldn't reinstall my old graphics card back. It'll be back in a jiffy and I can just sate my gaming desires with the Steam Deck in the meanwhile.
Unfortunately due to the poor pick-up schedules at my nearest DHL location, it took around a week from leaving the return package to it actually arriving at AMD's return location. At this point, it had already been two weeks since I got the initial card and that I submitted the RMA request. And then, AMD's partner kept sitting on the package for around week and a half, and so May turned to June.
June
Finally in June, I got my second DHL shipment notification and DHL then hurried over the package to my front door in less than 24 hours. Again, immediately after work I painstakingly stuffed the 7900 XTX into my computer case and launched Cyberpunk 2077 to finally see what kind of performance the GPU can dish when it's not throttling.
It overheated and throttled immediately.
Unfortunately the second card came just against the weekend, so I had no choice but to continue the already-long email chain with AMD customer support and try to get another RMA open. Although since I'd just spent the past three weeks looking at my half-finished gaming PC gathering dust, I was leaning more towards a refund than having to do the same process again.
Then Monday comes and AMD replies. They couldn't ascertain that the graphics card was overheating if I didn't send them measurements from the AMD Radeon software. This felt a bit odd considering the first RMA they didn't bother even asking for evidence but I still went ahead and gathered them their precious Radeon screenshots. At least it's not difficult to document when the card hits 110°C as soon as you load a Cyberpunk 2077 save.
Having proven to AMD that the card was in fact overheating using their own tools, AMD customer support had another quest for me: send system diagnostics reports of my gaming PC and prove to them that the GPU power cables (two 8-pin connectors) are connected. Again, I have a hard time understanding how this is necessary since a month ago the same customer support agent had told me that these cards have a known flaw.
Unfortunately that was not enough. What was now required of me was to update my BIOS, do a clean reinstall of all AMD drivers, send details of the power supply that I was using for the card, and install the graphics card without the riser card my mini-ITX case requires. There is no clearance to install any kind of a graphics card with the motherboard in the case, so doing this would have required completely disassembling my entire PC.
I'd had enough.
I told AMD customer support that I was done playing their games. No BIOS update was going to fix this issue and AMD's drivers cannot be so bad that they turn the GPU into a hot plate while it's throttling. But thankfully after being told by a friend and having been proven by der8auer, these cards actually have a single orientation in which they won't overheat. Unfortunately, that was not an orientation that was achievable mounting inside my case.
So I did the only sensible thing imaginable: I grabbed two rolls of toilet paper and very carefully perched my gaming PC on its back to get the card in a vertical orientation. And while the PC might have been very unstable when it came to sitting on those bog rolls, the card did stay at a stable 80°C while playing Cyberpunk 2077. Clocks were also staying up and the fans were running a whopping 1100 RPM slower.
And those stats stayed stable for over 20 minutes. Then I turned the computer back up again. Temperatures immediately skyrocketed. Not even a minute and the junction temperature was 110°C, forcing the card to throttle itself. So the cooler was most definitely busted for the second time in a row.
Thankfully, telling AMD customer support how fed up I was with their diagnosis for an issue they are already aware of and sending them the graphs of the vertical orientation test worked. After about a week of constant contact with AMD and I got my second RMA.
And so, today, I dispatched the AMD RX 7900 XTX towards AMD's fulfillment center. Hopefully in a couple of weeks time, I will have my money back and I can leave this episode behind me.
Future
I think there are two lessons that I've learned from this experience.
First of all, I really need to do the research before putting down this kind of money towards a product. On some level, I was aware that AMD had a overheating problem with some of its products. But that was in the news months ago. And it only affected some units (or so AMD claimed). That's surely not something that I'd encounter. Twice.
And secondly, maybe just avoid buying reference graphics cards. They're clearly not that good and at least when it comes to AMD, neither is the customer support. And based on the discussions I've had with people, it seems like Nvidia support is not world-class either.
But I'm still on the market for a new graphics card. The reason why I picked the RTX 3060 was because I hadn't owned a gaming PC for over seven years at the time, and I wanted to try out PC gaming again with a careful approach. But with the amount of PC gaming I've done lately, and the inherent low performance of the card, I want to upgrade. But to what?
One of the reasons why I went with the 7900 XTX was because I didn't want to give my money to Nvidia and support the insane prices that they charge for their cards. The AMD RX 7900 XTX and Nvidia RTX 4080 have comparable performance but the 7900 XTX is over 200€ cheaper. But now that I've put my money where my mouth is, and fought against the Nvidia hegemony, the only things I have left to show for it are a long email thread with AMD customer support and additional scratches in my GPU bracket.
So now I need to choose: do I go for a custom 7900 XTX design like the Sapphire PULSE 7900 XTX and pay a premium over the reference cooler design just to get a working card, or do go back to Team Green with my tail between my legs and pay hundreds of euro more, to a point where it makes sense to just go for the RTX 4090 instead? I feel like there are no good options here, but Nvidia has yet to burn me.
Promise updates, take the money, and then ask for more.
Secrets by Outer Corner is a password manager for macOS and iOS. They just released the latest version of the software, Secrets 4, today and rarely do I feel as outright scammed by software as I do now.
I decided to start using Secrets a couple of months ago after looking for a successor to my previous password manager. It had many good things going for it: it was a native macOS/iOS app, it didn't require a subscription, the data was hosted in iCloud, it had an integration for Windows, and so on.
No one wants to buy a product that is unsupported, especially when it comes to software. This is especially important on macOS and iOS side, as Apple is famous for breaking old software in yearly updates. Windows users do not suffer nearly as much from this, as it's not unthinkable to be able to run a Windows 95 application in Windows 11. No wonder Windows developers have been so slow to adopt 64-bit and ARM support.
So I did what any good consumer should do: I dug around some. More specifically, I dug around some on Twitter.
Okay, so Secrets has not seen a big update in over two years, but the developer has made multiple tweets promising an update, and a big one at that. So I would in fact not be buying an abandoned piece of software. So I decided to pull the trigger, and bought both the macOS and iOS versions of Secrets for about 40€ in total.
Today, Secrets 4 was released. The full and complete version of the universal app was offered to existing users as a launch day special for the low, low price of… 50€. That's after the considerable 50% discount. If you wanted to upgrade to the basic version of Secrets 4, you could get it for the low, low price of… 50€ again.
So, having paid 40€ for Secrets just two months prior with a promise of a big update, I am now being asked for somewhere between 50€ and 100€ to get that major update that was promised and advertised multiple times.
Ever since I bought Secrets, I have received a single update. The macOS version of Secrets was updated to version 3.8.5 about a month ago, which "fixed a couple of compatibility issues with macOS Ventura". The iOS version has not been updated in 10 months as of writing.
The developer Paulo Andrade has been deflecting criticism of the paid upgrade by telling people that Secrets users have been enjoying free updates for seven years. Good for those users, I guess. I sure didn't! The update that was coming my way, has been turned into a paid upgrade. I just got that single-sentence update in the two months.
You could argue that "update" does not mean that it's free of charge. However, I feel like "update" is an update to the existing software, and not a completely new application on the App Store and a separate purchase. And I feel like Outer Corner agrees with me, considering the announcement post for Secrets 4 only talks about an "upgrade", not an "update".
It's not like Secrets 3 is a perfect password manager as it is. I'm currently suffering from random freezes when invoking Secrets from the Firefox extension, the image cropper in the application does absolutely nothing, and the image rendering for the thumbnails on macOS is garbage. I always just figured that since the update was coming, I'd just bear with it and hope that the update fixes everything. Had I been told that the version of Secrets would be the end of the road, I'd probably have refunded my purchase within the 14-day cooling off period. Or just not upgraded from the trial to begin with.
Technically the version of Secrets that I have is not abandoned. Completely abandoned, that is. The developer has that they'll support Secrets 3 for "the near future". What is the definition of "near future"? Absolutely no idea, the developer refused to elaborate. My expectations of the future after all of this are very low. I imagine I will be relying on the good graces of Apple to not kill any of the API that Secrets 3 uses in order to use it on future versions of macOS and iOS, and not Outer Corner's support.
I don't like how almost every paid application these days is shifting to a subscription system. That was one of the reasons why I chose Secrets after all. However, the situation with Secrets is even worse than subscriptions – I'm currently doing like 20€ per month of active support, considering that Secrets 3 is now unceremoniously on life support. Secrets has been awful value for money. Even the Photoshop + Lightroom bundle is less than 20€/month, and Adobe software aren't known for being a bargain (at least for non-professionals).
I do understand that Secrets is made by a small indie development company whose developer needs to eat. However, I find the practice of selling an already-axed version of a software without any kind of a warning to be anti-consumer. I feel like companies have two very good alternatives to the way Outer Corner did things: offer a free upgrade to anyone who bought in the last [appropriate amount of time], or make it very clear that you are buying an unsupported product with no future.
If it's not clear already, I think you should not buy Secrets. But if you do, don't expect updates.
Time for another blast from the past, this time Bastion from 2011.
I actually played through Bastion for the first time way back in 2012, when it took me about 12 hours to beat the game and do whatever side tasks I did back then. A decade older and wiser, I got through my first playthrough in under eight hours and mopped the rest of the achievements in another ten.
Turns out that Bastion is still a nice game.
You play as The Kid, a silent protagonist who has to smash and fight through a bunch of isometric stages that build around you as you move about the place. The core gameplay loop of Bastion revolves around combat, which is handled using melee and ranged attacks, a shield and special attacks. There is a handful of weapons to choose from, which unlock as you get through the story, and you get a choice of upgrades to further custoize your combat experience.
The combat itself is not particularly challenging, although you definitely can make things harder for yourself using the Idol system. About the Author, which requires you to defeat 30 waves of enemies in the Stranger's Dream while using all of the hardenings available, definitely gives you a run for your money. Even though I managed to beat it in two attempts, the first attempt failing at the last wave, it was not without its share of panic, screaming and aching arms. But if you don't reach for the Idols, combat's more on the relaxing side of the difficulty spectrum.
Oh, and if you decide to go for About the Author: Galleon Mortar and dodge roll, dodge roll, dodge roll.
There is a fairly short and basic story in the game, although it's not particularly gripping. Most of the interesting stuff in the Bastion universe happened before you even came along, and you're mostly told about it from the narrator. If you were hoping for a story-heavy game, you might be disappointed, but it's still a decent, somber narrative.
Bastion's art is highly stylized and gorgeous, although not perfect; due to the isometric perspective and the way the levels are designed, it's hard to tell which parts of the level you can actually access and which you can't. It's also not the easiest to avoid falling entirely off the level, which will snip some off your health bar as a punishment. Still, the overall quality of the art definitely offsets those minor annoyances.
The sound design in the game is absolutely top notch. I'd like to meet the person who doesn't enjoy Logan Cunningham's voice-over narration, which there is a lot. And Bastion's soundtrack ranks in some of the best video game soundtracks that I know of. It's the kind of stuff that you'll want to listen outside the game as well. I know I definitely have.
For my 2023 playthrough of Bastion, I did it entirely on my Steam Deck, as my gaming PC was actually out of commission for the entirety of it. And there is good news on that front: Bastion has a native Linux version and works absolutely flawlessly on SteamOS. Since this game is designed for a controller and is not graphically demanding, it makes for a perfect Steam Deck title.
Considering for how little money you can pick up Bastion these days, and the abundance of good things about it, avoiding it would be foolish if you feel like it would appeal to you in any way.
yurizaki is a file copying and sorting daemon written in Rust, made specifically for anime and fansub releases. It's specifically meant to automatically upgrade files in your library with better release groups and newer versions. You can specify multiple release groups and it'll replace lower-ranked groups with higher-ranked groups when they are added to the source directory.
Everything is set up with a YAML file. For example:
source: /src
library: /dst
Go to Bed:
aliases:
- Kimi wa Nero
groups:
- Shino
- SupersDisplease
Then if you run yurizaki and start adding files to the source directory, you can see yurizaki matching, copying and removing files to keep the highest version of the highest-ranked release group in the library:
MATCH: "[SupersDisplease] Kimi wa Nero - 01 (1080p) [3838CF72].mkv" => Go to Bed
No other release
Copied "[SupersDisplease] Kimi wa Nero - 01 (1080p) [3838CF72].mkv" to "/dst/Go to Bed/[SupersDisplease] Kimi wa Nero - 01 (1080p) [3838CF72].mkv"
MATCH: "[Shino] Go to Bed 01 [1080p Hi10P AC3][F4D8E418].mkv" => Go to Bed
Inferior release found: "/dst/Go to Bed/[SupersDisplease] Kimi wa Nero - 01 (1080p) [3838CF72].mkv"
Copied "[Shino] Go to Bed 01 [1080p Hi10P AC3][F4D8E418].mkv" to "/dst/Go to Bed/[Shino] Go to Bed 01 [1080p Hi10P AC3][F4D8E418].mkv"
MATCH: "[SupersDisplease] Kimi wa Nero - 01v2 (1080p) [3838CF72].mkv" => Go to Bed
Superior release found: "/dst/Go to Bed/[Shino] Go to Bed 01 [1080p Hi10P AC3][F4D8E418].mkv"
MATCH: "[Shino] Go to Bed 01v2 [1080p Hi10P AC3][F4D8E418].mkv" => Go to Bed
Inferior release found: "/dst/Go to Bed/[Shino] Go to Bed 01 [1080p Hi10P AC3][F4D8E418].mkv"
Copied "[Shino] Go to Bed 01v2 [1080p Hi10P AC3][F4D8E418].mkv" to "/dst/Go to Bed/[Shino] Go to Bed 01v2 [1080p Hi10P AC3][F4D8E418].mkv"
The end result is that only one file is retained in the library:
$ tree /dst/
/dst/
└── Go to Bed
└── [Shino] Go to Bed 01v2 [1080p Hi10P AC3][F4D8E418].mkv
I started writing yurizaki sometime in 2020, or possibly even 2019. Then I got it to a reasonably working state and just started using it half-unfinished. After that time, I've actually published several Rust projects on GitHub, meaning that this is not actually my first Rust release. But since all of those were written after the conception of yurizaki, I'm still going to call it the first.
But despite being my first attempt at Rust, and being left unfinished literally for years, it has actually worked very well in my use. I guess that might be one of the benefits of Rust – if you get it to compile, and you didn't litter your code with shortcuts, it might reasonably handle any situation you throw at it. Although my code did definitely have some shortcuts, and still does, but it has still worked fine. I don't remember if I've ever yurizaki panic on me.
I do still have some things I want to add to it though, so it's definitely not completely finished. I for example want to be able to run custom commands after copying to automate other things. It'd also be beneficial to reduce the remaining unwrap() uses to improve stability. I've also noticed that anitomy doesn't always play nicely with certain perverted titles, so regular expression overrides might also be a good idea.
Oh yeah, and there definitely are no unit tests yet. I always find unit testing in Rust to be a pain in comparison to a more flexible language like Python.
(Named after Yurizaki Mira from Dimension W, an absolute cutie patootie.)
I've obtained quite a few games from various kinds of bundles, and I imagine that this was one of them since I cannot remember paying money for this one. And having played through it, I feel like that was the right choice.
The story and writing aren't really that good. I couldn't really get into the story and the characters, who don't really feel like actual people. I guess the androids get a pass for that, but the people definitely don't. You are given a couple of choices during the story, but it doesn't feel like they really do anything beyond the next couple of minutes of story. Feels like they were just thrown in there to be able to say the game has "choices" (plural).
The bad end feels like driving a car into a brick wall. I was shocked to see the credits roll. Completely feels like the story was just left unfinished. The good end felt less abrupt but still not exactly like a satisfying conclusion. Neither of them really left me satisfied and the epilogue didn't help things. And to top things off, there are issues with spelling and grammar. The script should be ran through a spell checker.
I don't really like the art. Most of it looks like they left in the placeholder art. I'm guessing that's a style that they're purposefully aiming for but it really doesn't click with me. There are a couple of hits when it comes to the CG, but most art were misses for me. There's also not a great variety of art on display.
The soundtrack is super simple but fitting. It's probably my favourite part of the game. I'm not sure if I would ever actually sit down and listen to it specifically, but it fits the general mood of the game. No voice acting but I'm not sure if I would've wanted any from this game anyways.
In addition to the issues with the visual and the novel aspects of this visual novel, it also seems to suffer from technical issues. When I read through it, there were these text transitions between scenes that flashed by my screen in like 0.2 seconds. Was this the intended effect or was it a bug? Could go either way but definitely felt like the latter. There's also a setting that I cannot figure out what it actually does.
At one point there's also a completely black screen, and you're just kinda supposed to figure out on your own that you need to advance manually to continue the story. And even when you are shown the dialogue UI, it's really not good. The font is also very small to a degree where it feels awful ergonomically. I do have to wonder what kind of visual novels the developers are used to reading. I at least don't remember another VN with such a poor user interface.
There's also no cross-saves, so I wouldn't try to split the read between your PC and Steam Deck. Not that you'd want to read through it on your Steam Deck anyways, since the text is tiny and there's no ready controller layout. You'd either have to make your own or stick to trackpad controls. For what it's worth though, it seems to run "fine" (the same way it does on Windows) on the Steam Deck.
There's really just not much to like here. While my monetary investment towards this product is minimal, I am still disappointed at what return I got for my time investment. Honestly speaking, I would not recommend this visual novel to anyone.
Just got out of the movie theatre after watching Suzume no Tojimari.
I've never really self-identified as a Shinkai Makoto fan, but if I keep going to the movie theatre to see his movies and love all of them, I guess I just might be one. Definitely feels like that keeps coming true ever since I watched Kimi no Na wa in a theatre back in 2017.
(I have seen Shinkai movies before Kimi no Na wa but not in theatres.)
If you've seen the two Shinkai movies before Suzume no Tojimari, you probably know what to expect: beautiful scenery, fluid animation, tons of details, the supernatural, comedy, drama, love and RADWIMPS. Stick to what you know, right?
The story was captivating as ever. Shinkai also writes great characters that you'll want to root for. That goes for the main characters and all of the side characters. Suzume was of course as adorable as Hina and Mitsuha before her. Tears were definitely also shed during. You can definitely see how the 2011 earthquakes shaped this story, and the suffering brought by it. You even see a diary entry for March 11th during a scene.
I think Suzume might be the first time Shinkai has crossed the two-hour threshold in runtime. It didn't really feel that long, and the narrative was quite packed. Wouldn't have minded at all if it was even longer, I wasn't checking my watch.
I feel like this time around though, they definitely toned down the RADWIMPS. No insert songs this time, which feels like a bit of a shame since 愛にできることはまだあるかい playing during Tenki no Ko still gives me shivers. Not that the audio side of Suzume was anything to scoff at. The soundtrack fits the storytelling like a glove and the sound effects kept me immersed. I'm actually pretty glad I saw it in a proper theatre, you just can't get those rich bass sounds in a home setup.
If I had to rank Shinkai's latest three movies, I'm not quite sure what I'd say. When I saw Tenki no Ko, I felt like I enjoyed it more than Kimi no Na wa, even though the popular opinion seemed to be the opposite. I in fact enjoyed it so much that I went to see it in the movies twice. Even dragged someone to go watch it with me. But I don't know how to rank Tenki no Ko and Suzume relative to each other. I feel like I love them both, but in different ways. Not sure if I'd go rewatch it in a threatre though – maybe if I got to drag someone with me again.
I don't think Shinkai has pulled a Miyazaki (announced some kind of a retirement that may or may not last) yet, so I guess I'll wait for the next beautiful supernatural adventure from him.
Muv-Luv photonflowers* is a collection of side stories from the Muv-Luv universe, both Extraverse and the Alternative timeline.
Extra
The Extra side of things contains one longer story and six short stories. In my opinion, the short stories are kind of a waste of time. They're not particularly interesting and don't really have any payoff. The correct time to read through them is probably right after finishing Muv-Luv Extra. I'd also read them before reading through the longer photonflowers* Extra story, mostly because they'd feel like a disappointment after.
The longer Extra story is luckily a lot better than the short stories. It continues the story Extra story from the Sumika ending and feels like a proper extension of Extra. Much like Extra, it contains a mix of humour and serious bits. I think it alone is a couple hours worth of reading and has a satisfying story arc with an emotional ending.
You do get full voice acting with the exception of Takeru, just like in Extra. Art department will also feel like Extra, since it's mostly recycled assets from Extra. I guess it would feel weird to have new character sprites but you can't praise them for going an extra mile here.
Alternative
The Alternative side of photonflowers* is split between five side stories from the Alternative timeline.
Two of the longer Alternative side stories, Confessions and Atonement, are focused on giving established characters a backstory, and are on the longer side. I found both of them to be enjoyable reads and I imagine that any other Muv-Luv fan would also find them enjoyable. They are very much the flagship titles of this collection.
The remaining three, Rain Dancers, Chicken Drivers and Inheritance are very short. I think all of them were around 15-30 minutes each. While none of the stories were bad, some of them weren't great.
Rain Dancers centers around characters that I'd never heard of, fighting at a front that I've never seen covered in Muv-Luv before, and didn't really have any kind of a discernible story arc. Chicken Drivers also mostly has never-seen characters, but at least they're fighting at a front already covered, and gives a bit more interesting backstory. Inheritance has the tightest coupling to the Alternative story, and is probably the best of these three.
The art on the Alternative side has a lot of new sprites and generally feels higher quality than the Extra side. Possibly because they're not twenty-year-old assets. Some good CG and sprites too. Even the shortest stories get pretty good art assets. Again, full voice acting included.
photonflowers*
Systems-wise, photonflowers* is basically the same as Muv-Luv and Muv-Luv Alternative. That is to say, the technical level of it is about two decades behind today. Coming to this after reading through THE DAY AFTER trilogy (tetralogy?), it definitely feels dated. I would much rather have the engine from THE DAY AFTER. As a speedreader, I especially missed the ability to set when voice lines are stopped. The dialogue display is also a lot cleaner in TDA.
Thankfully it's very playable on the Steam Deck. It actually feels like it suffered from less audio issues after you wake the Steam Deck from sleep with the game running. I definitely had to restart the game a couple of times to get everything working again. This is made much less painful by the fact that photonflowers* jumps right back to where you quicksaved, so restarting is maybe like a 5-10 second ordeal.
However, when it comes to the price, the value proposition doesn't feel that great. There's around ten hours worth of reading, and all of the stories aren't bangers. Nonetheless, you are asked 21€ for the package on Steam. I would personally not recommend photonflowers* for that price. However, I got my copy at a 50% discount, and for that price point, it seems like a decent package.
All in all, Muv-Luv photonflowers* is a decent read for the dedicated Muv-Luv fan – at least if they don't overpay for it.
Seeing as Twitter has been (unsurprisingly) in a constant backslide since last autumn, I've decided that I need to start diversifying where I output all of my thoughts and opinions. And considering what a craptastic thing Mastodon is, I'd rather not partake in it either.
Therefore, I have decided that I should have a web log. Just like in the good old days.
I have probably attempted to have a blog in the past too, without much success. I was much more of a fan of microblogging, which is probably why I've been on Twitter for the past 15 years. But since there's a massive vacuum in the microblogging space, I've decided to pivot into (macro?)blogging.
Quite honestly, I do not have a long-term vision for this blog beyond having a space to store whatever I feel like I need to release into the wildness of the Internet. Most likely it'll be just a pool of random thoughts about whatever interests I have at the time, without any kind of a specific audience in mind beyond myself. Possibly also quite a lot of grumbling and venting. The first posts are just short reviews of the games that I've played lately, reposted from my Steam profile.
Also, considering what has been dominating the tech news for the entirety of 2023, I'll start off by proclaiming that this blog will not be written by large language models, but by something resembling a human.
Technical
This entire site is a static site made with Lektor. I chose Lektor because it's built on Python and uses Jinja2 for templating, and I'm already familiar with both of those so it's easy to hack on. And I chose to use a static site because it's much simpler to run. No need to worry about your unpatched Wordpress installation becoming part of a botnet.
For the layout, I just hammered away on the default layout that it bootstraps until it looked like me. It probably has very little in common with the initial layout now. Not gonna lie, restricting yourself to black and white does make it a bit hard, but I think my hands were tied with that one when I decided to use this domain.
NEEDY GIRL OVERDOSE is definitely an interesting experience.
The gameplay is essentially a raising simulator, where you make decisions for OMGkawaiiAngel to make her the biggest streaming star on the Internet. It doesn’t take long to wrap your head around the available choices and the day-to-day rhythm. Advancing through a single day is also quite fast, so you can make through the game pretty fast. Nothing too advanced here, just a core gameplay loop that does what it needs to do to present the story.
Story-wise, it’s a bit hard to comment. The game doesn’t play like an ordinary visual novel with a branching storyline. It’s more of a Choose Your Own Adventure book where you’re constantly making decisions that affect what happens, and at the end, you’ll land on one of the endings. Hence, there is not really a “story”.
The endings vary from humorous ones to more serious ones, and there’s a couple of dozen of them. Not gonna lie, some of the endings did hit me like a ton of bricks. And luckily grinding them isn’t too arduous of a task. However, if you go for 100% completion, you’ll probably need a walkthrough – some of the endings are pretty opaque as to how you’ll reach them.
I also need to point attention towards the localisation: it's fantastic. It actually feels hard to imagine that it's been translated from Japanese. Feels like they definitely put in effort to make it work in English.
The art is perfection. From the character designs, to the ways the sprites are drawn, the various filters that happen when things start to go awry and the faux Windows 95 user interface… I can’t find a single thing that I do not love here. I’d say just getting to ogle at the game is well than worth the price of admission alone, considering that this is not an expensive title.
The music is also superb. The chiptone soundtrack fits the art and story and it definitely didn’t get old during my time with the game even though you will hear a lot of repetition. And the theme song is just a chef’s kiss on top of that. Naturally, you can listen to it in its entirety in the game, as well as hear a chiptone arrangement of it as one of the background tracks.
A fairly short and bittersweet, I’d definitely recommend NEEDY GIRL OVERDOSE to anyone with even the vaguest of interest in it.
Finally, the explosive finale to THE DAY AFTER series. Except it’s not really the finale – they’re just calling the continuation “Muv-Luv Resonative” instead of “THE DAY AFTER”, but I digress.
When I gathered my thoughts about THE DAY AFTER 02, I described it as "the best that THE DAY AFTER had to offer so far”. This was definitely correct – the best so far, as the best that THE DAY AFTER series has to offer is most definitely THE DAY AFTER 03.
The story feels packed for a relatively short read – I clocked 7.7 hours to complete it, or 30 minutes less than it took me to finish THE DAY AFTER 01. But unlike the other parts of THE DAY AFTER series, it feels like a significant increase in quality, and I never felt like I had to wait for the good part to begin like I did with the previous entries. It feels to the rest of THE DAY AFTER series what the Muv-Luv Alternative feels like its predecessors, although in condensed form.
Everything is still fully voiced with good voice acting. I don’t think I saw that many new sprites for returning characters. However, THE DAY AFTER 03 uses the old sprites in a much more animated way that makes everything feel that much more alive. In that aspect too, THE DAY AFTER 03 feels like a significant step up from the previous episodes. So while the old sprites may have not been anything amazing, it still gets a passing grade from me on that department.
On technical aspects, most stuff is fine. You’ll need a custom controller layout if you plan on using a controller on PC or reading it on a Steam Deck. And it reads just fine on the Steam Deck barring some post-sleep audio crackling that you can resolve by restarting the game. You should also stick to reading THE DAY AFTER 03 on one device – the way this game handles Steam Cloud saves makes switching between devices not impossible but very annoying.
THE DAY AFTER 03 is a definitive high note to end the series – at least until the fabled Muv-Luv Resonative comes out some fateful day.
The third instalment of the THE DAY AFTER series, unless you decided to skip THE DAY AFTER 00 for some reason. I read somewhere that THE DAY AFTER 02 was the best yet that the THE DAY AFTER series had to offer. To be honest, I didn’t really believe it for approximately half of the story. It didn’t feel like the story was really progressing, but rather that I was getting a beat down.
Luckily that changes somewhere around halfway through the story. After that, it definitely picks up. The story gets more stakes and you see a lot more action. THE DAY AFTER series also finally starts adjoining the prologue TDA 00 into the overall storyline. And as the story progresses towards the finale, I felt that it definitely did become the best that THE DAY AFTER had to offer so far.
On the character level, THE DAY AFTER 02 adds several interesting characters, like Ogami and Komaki. Maybe they won’t be everyone’s favourite characters, but I did find them to make the story more interesting. THE DAY AFTER 02 also continues re-introducing familiar faces. Whereas previous instalments of the series could haul the old-timers around in a car, THE DAY AFTER 02 will need a minibus to carry all of its familiar faces. So it’s also definitely the best that THE DAY AFTER has to offer for fans of the series who want to bridge the gap between Unlimited and THE DAY AFTER.
On the art and sound direction there isn’t that much new. You still have a bunch of the old sprites and the main characters also get some new sprites. Honestly speaking some of the new sprites didn’t really seem to match well with the old sprites, so it comes out as a bit off-kilter. That being said, not having the new sprites as well would’ve felt a bit like they were cheaping out. Some of the art is still too low resolution for the amounts of zoom they’re asked of as well. Everything is still fully voiced and the performances are good.
The span of the story gives me a bit of pause though. Steam shows me that it took me 8.2 hours to read through TDA 01 whereas TDA 02 was a 6.7-hour read to completion. That’s a full 90 minutes less reading for the same price point. And there was definitely less effort needed on the art side, since a bunch of the assets could be recycled. But then again, if getting length-parity with TDA 01 would’ve required padding out the beginning of the story, I definitely wouldn’t want it. Yet, it definitely stings the part of the brain that considers value for money, even if the overall story is better than TDA 01.
All in all, it’s very easy to recommend THE DAY AFTER 02. You’re probably not getting this far into the series without having an investment into it, and despite some of the initial drudgery, you definitely get a payoff at the end.
THE DAY AFTER 01 continues the story in the Unlimited timeline. Mood is a bit all over the place, ranging from very serious politics discussion on life-and-death matters in a ruined world to Extra-esque slapstick comedy moments where the main character accidentally fondles tits. The most lighthearted moments even come with complementary chibi art. While I found the lighthearted moments of the story fun, I did feel like they dragged on a bit too long. But, much like with THE DAY AFTER 00, towards the end of the game I felt captivated to continue reading. Overall, THE DAY AFTER 01 does definitely deliver a more interesting story than its predecessor.
Most of the characters in THE DAY AFTER 01 are completely new. The main exception is Marimo-chan, who enjoys a lot of attention here. But those worried by the lack of familiar faces, don’t fret – the new cast is likeable. The main character does definitely have the kind of an idealistic view on world that some might find off-putting, but I didn't find his thinking to be too idealistic. If you’re anything like me, you’ll be cheering on the squad towards the end. I’m looking forwards to exploring more of their future, and their pasts.
As TDA 01 is roughly double the price of TDA 00, I expected the runtime being roughly double – and it is! However, it’s still quite short, just not as short as TDA 00. I read through TDA 00 in 4.4 hours and TDA 01 was 8.2 hours. However, as the regular price for TDA 01 is about $25, it’s definitely not a value purchase as far as visual novels come. Personally, I’d wait for a sale (and I did back in 2021).
Art and sound design are mostly good. Character sprites felt better than what TDA 00 had to offer. It also feels like there was less distracting zooming in on backgrounds that really just highlights any kind of artifacts. Either that, or I’ve already gotten used to it. Fully voiced story again with good character voices.
Translations at times dipped a bit too much on the liberal side for me. Luckily with the full voice acting, you can spot the differences if you have some amount of verbal Japanese knowledge. I also feel like they toned down some of the racist comments used in the story, going from “monkeys” to “Japs”. Not sure if they aimed for that, or if they just thought that the original comments wouldn’t have the wanted impact on English-speaking audiences.
Technical aspects are mostly fine. I read it from start to finish on my Steam Deck and there’s no blockers there. But you’ll have to use a custom controller layout to have a proper experience, and even then you’ll have to use mouse controls for some elements. Really wish they paid more attention to proper controller support.
I also observed the same kind of crackling after the VN has been running for a while / being suspended in sleep on my Steam Deck as I did with THE DAY AFTER 00. I feel like switching from the regular Proton to Proton-GE helped to alleviate the crackling issue, so maybe try that. Of course, you can just relaunch the game to get the audio in perfect condition again.
Overall, THE DAY AFTER 01 is an enjoyable story despite its relatively short length and smaller issues. I imagine I’ll crack open THE DAY AFTER 02 tomorrow to find out what happens after the world ends.
First things first: the environmental design in Ghostwire: Tokyo is great. You have an expansive virtual Tokyo to explore, and not just on the ground level, as you also have access sprawling underground systems and rooftops to explore. And not only is the world expansive to explore, but it’s also dressed up in this spooky veneer. And in parts of the game you also get these trippy sequences where the world just goes completely haywire and basically just dances around you. It’s great. The game world is definitely the best part of the game, and I can imagine someone getting their money’s worth just from the eerie Tokyo experience.
The combat is fine. If one were to pick up this game for the combat, I imagine disappointment would be almost guaranteed. Not that it’s bad, but it’s clearly not where the game shines. You have couple of basic attacks with slightly different behaviours, but at the end of the day, you aim at an enemy and shoot at them until they stop. Stealth mechanics are also not the best that I’ve encountered and there’s a couple of forced stealth segments. Nothing unbearable, but also not my favourite parts of the game.
There’s some degree of RPG elements in the gameplay, but not enough to actually make it an RPG. You have levels, and gaining a level nets you skill points that you can spend to improve your abilities. However, the combat always feels like the enemies always have the same level as you do, so you’re at least spared from having to grind levels to beat the game. Going for full completion will also net you all of the skills, so there’s no any kind of specialisation, mainly just a matter of deciding which order to acquire them.
The story is also just fine. The main story definitely isn’t too intricate or captivating and there’s no extensive lore to explore if that’s your thing. Again, nothing terrible here, but clearly this is not a story-first product. The story does luckily manage to hit some good emotional notes along the way. The side quests also don’t have any deep story, but do offer some small slice of life segments featuring ghosts and yokai that I found as quite enjoyable bite-sized experiences. The main sidekick that accompanies you throughout the game is also a fun character, and I enjoyed the interactions between them and the player character.
And while the game might be tagged as “horror”, I can personally guarantee as a certified coward that the game isn’t actually scary. It’s mostly “creepy”, “spooky” or “eerie”, but not outright scary or horrific. You’re not gonna get jumped on out of the blue or be forced to crawl through dark rooms. It’s definitely light enough of a “horror” experience that I was able to dig into it.
If you’re a completionist, be prepared: the game is full of collectibles. They want you to go collect landmarks, spirits, yokai, raccoon dogs, statues, shrines, beads, toys, foods, music tracks, outfits, and emotes, as well as complete all of the side quests. It’s actually wild how many different collectibles they put into the game, Shinji Mikami really got sick with Open-Worlditis here. The saving grace for the collectibles is that they’re spread out the amazing landscape, so wondering around for collectibles can be quite scenic. Thankfully there is a button that highlights nearby items to you, so be prepared to use that a lot if you’re one to go for all the achievements.
If you’re not a completionist, maybe temper your expectations; the main story takes about 10 hours to clear. Completionists will be in it for 30–40 hours, depending on your efficiency and luck. No wonder they added so many side activities, since just 10 hours of content would’ve been brutal for a full-priced title.
Ghostwire: Tokyo is a beautiful supernatural rendering of Tokyo. While few of the game’s aspect shine bright, I still enjoyed my holiday in Shinji Mikami’s creepy playground.
Disclaimer: I did not pay a single dime for this game, as I received it as a Christmas gift from a friend.
Taking place after the events of Unlimited, the story has quite a different pace than Unlimited or Alternative with the BETA threat gone. I wasn’t completely sold on the story at first, but the characters are thankfully mostly likeable and interesting. However, I did find myself enjoying the read towards the tail end of the game. The tail end that won’t take too long to reach – it was around a four-and-a-half hour read for me. Obviously, as an “episode zero”, it is meant as a prologue to the rest of THE DAY AFTER story, and that’s probably why it’s not a very long and extremely captivating story.
The art is mostly satisfactory. The characters have some animation, and a good amount of expressiveness and perspectives. The biggest issue with the art is that some graphical assets are not high enough of a resolution for the amount of zoom levels they push, so you have occasional mosaics right in your face. Audio-wise all characters are fully voiced and you get some hits of nostalgia with some familiar background songs.
The game also plays nice on the Steam Deck, but you’re going to need a community layout. I don’t know about what goes into getting a game “Steam Deck Verified” as episode 00 is, but there’s no actual good Steam Deck layout. The community layout that I used allowed pretty comfortable UX but still required a bit of joystick mouse movement, so there’s definitely room for improvement. I did also notice some crackling issues with audio after waking the device from sleep without quitting the game, but that might be a SteamOS issue, is not really a major one and can be resolved by just restarting the game.
All in all, I would recommend THE DAY AFTER 00 for a fan of Muv-Luv – just temper your expectations, and maybe don’t pay full price. After all, if you’re interested in the THE DAY AFTER series, you’re probably not going to skip the prologue, right?
Two decades later and Jet Set Radio is still a great game. Luckily the visual style keeps the game looking quite fresh despite 23 years worth of patina. The movement and camera controls do show some age, but not to a degree that you can't get used to them. Once you do, the movement does actually become rather nice and ever-so stylish. And the soundtrack is still one of the all-time greats.
My second playthrough of the main story only took around 4.5 hours after having beaten it for the first time in 2014, so it's definitely not the longest. I imagine first-time players will progress through the game a bit slower than I did though. However, there is still more content and collectibles to grind out if the main story isn't enough skating action.
I would definitely recommend this piece of gaming history. It is, if nothing else, a stylish and fun little snack.
As an added bonus, played basically perfectly on my Steam Deck and with good battery life.