itch.io

I’m about to harp on the rhythm game in My Dystopian Robot Girlfriend pretty harshly. Don’t get me wrong: I’m extremely happy that the game even includes such a thing, and grateful it exists, even in its current state of wonk. Even so, it leaves a lot to be desired, and I hope that what I’ve written below can be understood as constructive criticism. I understand well that programming a game is by no means a simple task, and that the suggestions I make later on aren’t things that can be accomplished in a short time period by any means. Even so, I do strongly believe it would be worth cleaning up this aspect of the game, as right now, it feels rushed and somewhat poorly thought out, which is sad. Rhythm games combined with eroge are a rather rare breed, so believe me at least when I say that I’m happy this exists at all.

That said…

As an avid rhythm game player, I have to agree: it’s upsettingly bad. It desperately needs a calibration option—and, ideally, a user-friendly tool to assist with calibration. For me, just being able to adjust the timing would be “good enough,” but for others who aren’t too into rhythm games, a calibration tool is more or less an accessibility feature… in other words, it’s an absolute “must-have.” It’s a shame that even a year later this is still a problem.

It would also be nice to adjust the note “fall” speed. When using the highest speed increase modifier, many of the charts become visually incomprehensible due to their absolutely ludicrous speed, at least on a 60Hz monitor. Perhaps, were I playing this on my tablet (with its 120Hz screen), it wouldn’t be as prominent of a problem, but this is such a trivial thing to allow players to change, and the actual difficulty of a song’s chart shouldn’t be dependent on the player being able to process incoming notes within a half-second. Combined with the awful 100ms+ desync, it makes it damn near impossible to actually beat many of the charts, which really sucks given that I’ve played much harder rhythm game charts than these.

Also, the two-button input schema is not great. It’s difficult to become familiar with the controls. And the fact that the Z and X keys (seemingly) can’t be rebound, nor can they be bound to more than one key, additionally makes the higher-difficulty charts near impossible. On PC, at least, you can technically use LMB as an alternative for Z, but, what about X? RMB doesn’t do jack. :/

I’d love for the rhythm game stuff to get some attention. It’s pretty uncommon to see rhythm games added into eroge. I can only think of a small handful of eroge that are either rhythm games in and of themselves or at least include rhythm game minigames, and they almost always suffer from similar issues. Developers frequently overlook timing issues caused by differences in hardware, don’t properly playtest their charts, and end up creating scenarios that are near-impossible—or, at a minimum, so artificially difficult that it’s hard to convince oneself to actually bother playing them.

TL;DR: We need Calibration Tooling, Keybindings, and Note Fall Speed Options. 

If the dev(s) happen to stumble on this obnoxiously lengthy, ranting post of mine, here’s some actionable recommendations:

1. Calibration Tooling: This consists of the ability to set a precise +/- offset globally and per chart (as some charts have much worse desync than others). In a perfect world, the charts would be properly aligned, but for a small team who only added this as a minigame, it’s a big ask (as it’s extremely time consuming to do so). Additionally, for genuine accessibility purposes, a calibration tool must be made available. This should essentially ask the player to tap the input keys at the time they can hear the sound, and calculate an offset from an average of the taps. Importantly, do not limit the quantity of taps a player is allowed to enter into the calibration assistant tooling (e.g. by only allowing them to hit the key 5 times before averaging it out)., as smaller sample sets are more error prone. Players should be allowed to have the game calculate averages from as many keypresses as they wish. Far too many rhythm games make the mistake of neglecting their calibration tools, which inevitably leads to frustrating user experience down the line.
2. Keybindings: Allow us to rebind Z and X as both primary and alternate keys. By default, a better control scheme would be S and L for “Z,” and D and K for “X.” This way, a player can still one-hand it if they want to, or they can two-hand it (and can do so comfortably). This, on its own, would significantly improve the rhythm game’s playability.
3. Consider a Different View: Instead of a single track with both Z and X notes coming in from the right to the left, consider two tracks (ideally vertical, but horizontal would be acceptable; this is mostly a personal taste thing). Consider additionally using icons with higher variation in their visuals (significantly different shapes are a common choice, e.g. circle + triangle, with drastically different colors). Bonus points if you allow the player to adjust the color. For some, this is another accessibility need, in that some players will undoubtedly have one of the many types of color blindness.
4. Allow Changing Note Fall Speed: This refers to the speed at which a note moves down the track. In addition to allowing the player to arbitrarily adjust this per track, consider tidying up the maths that are being used to calculate the note fall speed changes from track speed increase modifiers. The highest (and second highest) speed mods especially can cause the notes to become utterly unreadable on 60Hz monitors. I imagine this problem would be even more godawful for players who are using low-end devices that can’t even reach 60 FPS (or, rather, whatever the hell their monitor’s refresh rate is). Notes should never fall so quickly that you can’t visually see, process, and (feasibly!) prepare for them—regardless of if you are an expert rhythm game player or a complete newbie. This one is crucial.
5. Fix the Bad Charting: This one would be easiest if we were talking about a 3- or 4-key layout, rather than a 2-key. There are numerous charts that, even with a proper KB layout (e.g. S+D/K+L), would still be obscenely difficult—and by “difficult” I don’t mean actually difficult, in the sense that their charts are well-made, representative of the corresponding audio, and include tricky patterns, but rather, I mean arbitrarily difficult by design. These tracks feature rapid-fire “Z Z Z Z X X X X Z X Z X Z Z Z X X X X” type patterns or “Z Z X Z Z+ = = = = Z- Z X+ = = X- X Z” types of patterns, wherein (when they are sped up 75–100%) the notes come so quickly that you end up wrecking your wrist trying to press the key fast enough. It’s not a matter of being “good” or “bad,” but instead that the design effectively prohibits playing these tracks with >+X% speed mods. There is, unfortunately, a hard limit on the difficulty level you can feasibly achieve with basic/simple 2-key rhythm games, and this  rhythm minigame’s higher difficulties swiftly zoom well past that limit. Either accept that the difficulty level needs to be lowered, add more keys, or add more contextual actions to the keys. I won’t get into that last one; see games like BIT TRIP RUNNER, Muse Dash, or the in-development (turns out, this releases literally tomorrow) Unbeatable (or Unbeatable [white label]) if you want to see how seemingly simplistic rhythm game “rules” can be pushed into high difficulty via context-reactive keypresses. That said, I’m personally preferential to the “add more keys” option—it’s the easiest way, code-wise, to “fix” this issue, as it doesn’t require much more complexity than the current code (at least, comparatively to the other options).