galaxyBETA
explore
· ·
log in

log me in

log in forgot password

don't have an account?

sign up

comments for Squares

sort: go back

SecondTomorrow November 19, 2025
+43comment score: 43
Having passive points being determined by pure RNG in a game with no active portion to counter that feels like the gaming equivalent of hostile architecture.
Lustre November 19, 2025
+24comment score: 24
Why did you need AI to make this? It's so simple...
Termt November 19, 2025
+22comment score: 22
So far I feel like the "QOL indicators that show how lucky or unlucky your total multi is" are kinda... meaningless? I have no idea what average they're using, but even with a combination that is above average from what I've seen it has a red down arrow.
TerranClarenc November 19, 2025
+21comment score: 21
genuinely such a simple game, what was the ai used for, why was it needed?

developer response: It was not needed. Galaxy requires the gen-ai tag even if you only use claude code or codex to write some of the code.

classified39 November 19, 2025
+20comment score: 20
Some neat ideas, but the non-permanent spells doubling in cost kinda slows momentum to a crawl.
If you are wanting to avoid spell-spamming, maybe have them double UNTIL the manual collection button is clicked?

developer response: Their cost does reset when you collect pink squares.

yhvr owner November 21, 2025
+8comment score: 8
Shadowsand: we require it for code too. Apologies for the incorrect tag description

Usually there's no way we can tell if you've just used ChatGPT to help you fix a function or something, but some people can certainly tell if a game was vibe coded et al. (or at least, all the verifiers can after how many we've seen!) and not want to play it for that reason. The gen-ai tag is used as such so people can easily hide all such games with a couple clicks if they don't want any of it
sharpnova November 27, 2025
+7comment score: 7
I have to say.. the way the multiplier at the end of each row is randomly generated and has such a wide range.. is absolutely game killing for me. If you had allowed us to upgrade it (not just scale it because then same problem, just bigger #'s) but actually improve it so we raise the bottom ceiling on the possible multipliers, then you'd have really found something clever here.
numbclicker November 19, 2025
+6comment score: 6
Feels like the longer you play, the harder you will be punished for playing
toasttaster November 21, 2025
+6comment score: 6
I like the idea a lot. Some comments:
- The rolling of mults is kinda fun, it makes me reset until I get a bunch of good ones in the start. However it would be nice if I could somehow reroll them when getting a lousy mult.
- In the red layer my blue layer income is really poor, seems to me that the mult I got from blue layer remains active in the red layer and there is no way to reset blue layer anymore except by resetting red.
- Red layer does not have enough updates.
Overall a pretty nice start, looking forward to more.
BlueMagma November 19, 2025
+5comment score: 5
An export to clipboard option would be nice. I vastly prefer using galaxy's cloud saves.
dudeforlifez November 21, 2025
+5comment score: 5
I like the concept but the actual gameplay is pretty slow and boring. I suggest doing a huge update to make it better and i'd probably come back to it.
MarcElAcoplao gamedev November 19, 2025
+5comment score: 5
This game is not that bad, just slow, can have potential where each color square is a new layer.
-Before pink squares, the end is just a timewall, definitively needs to have something more, or simply better scaling with the upgrades (or a higher level cap/being able to afford more levels)
-Haste should be able to be bought multiple times to add the time (3 purchases = 1min 30s) to make it more afk-able later on
-Unless final combo upgrades are good, combo just feels like a sloppy rng wall where you just wait to purchase upgrades. There should be a blue/pink square upgrade to boost them imo.
-Buy max is needed (or auto upgrading)
MaZe_Quania November 21, 2025
+5comment score: 5
I like the idea of the game but the execution isn't the best. I don't think the RNG-aspect should be a part of the game as it's not at all forgiving. Especially when AFKing
sbloom85 November 19, 2025
+5comment score: 5
QoL suggestions: Be able to view Blue Square multiplier once reaching Pink Squares, view total speed percentage, upgrade minimum multiplier.
creyol November 19, 2025
+5comment score: 5
are these...squares filling boxes???
blobdin November 20, 2025
+4comment score: 4
Warning -- the section resets on point collection. The tab structure strongly implies that these are separate and there's no big warning.

I think the arrows are helpful. Allows for many quick resets to get a higher mult. Although it strongly appears below average is the average. Probably related to 1.5 being the min which has a much strong nerf to points than 3 being only 20% higher than 2.5. Basically, I think the wrong kind of mean (maybe need geometric mean? idk) is used here and/or the mult odds distribution isn't taken into account.
sharpnova November 27, 2025
+4comment score: 4
yep.. with later progression in the game, it never stopped feeling absolutely wretched that you were running with bad multipliers. and never stopped feeling like a chore having to reset after waiting. over. and over. and over. and over again until you got decent #'s.

i like the idea. i've liked it since the game you took the inspiration from: "pixels filling squares" did it.

it's spoiled rotten by the random multipliers. just makes the game stressful. if you fixed that part, i'd be a lot more interested in this.
yhvr owner November 22, 2025
+4comment score: 4
Shadowsand: the form for game submission has changed since you last went through it, and it now asks developers if they used generative AI while making their game. If they choose to say that they did, we ask them to elaborate on how they did so. At least for new submissions, we assign the tag on a case-by-case basis based on the severity of use. A case like what you outlined where helper functions or an engine was AI generated but not the UI or content would probably not warrant the tag, but it hasn't come up and if it did I'd probably have a discussion with other site staff.

The current usage of the tag regarding code has been if a player could reasonably infer that generative AI was used. It leaves traces of itself in almost everything it does, and there's often not a lot that's hidden from the user in incremental games so it ends up being applied often when it's used.

I tweaked the description of the tag, however it hasn't been pushed to prod yet: https://gitlab.com/yhvr/galaxy-astro/-/commit/264fda625167a6d61c446c266536079820ae8b0c I don't want anybody to clog up this comment section further with discussion not directly about the game, but I'm curious as to what you envision with two tags. Would you be willing to discuss more in site PMs or on Discord?

developer response: My 2-cents is that the tag will ultimately lose relevance. How many players actually use the tag for something other than a proxy for quality and depth? It is not a terrible proxy but it is getting less and less useful as generative AI gets better and serious developers start adopting it. Is it really interesting to have discussions about how much AI tool use warrants a tag or not? I don't see how it helps gamers.

batian November 22, 2025
+4comment score: 4
I'd say if a game was made using one of those ai website builders or if it's entire foundation was vibe coded, then it warrants an ai tag
Shadowsand gamedev November 21, 2025
+3comment score: 3
> Galaxy requires the gen-ai tag even if you only use claude code or codex to write some of the code.

The description of the gen-ai tag is "This game makes use of generative AI tools for assets (such as writing, images and audio)". It makes no mention of using gen AI to write the code.