itch.io

Cool upgrades so far! I'm using Windows and tried to export to gif but saw some error where it said something like ImageMagick executable not found.

I'm a professional computational specialist (working in accelerator physics and applied math), I recently started game dev as a hobby since AI coding tools have gotten very powerful over the last few months. My prototype game Echolumination started as a concept I had years ago but never had the time to build it. My procedure was roughly:

1. Chatting with Claude Fable over the viability of my unique RTS game (comparing existing engines like Unity/Godot or making it a mod for an existing RTS game like SC2). We concluded after some research that making it from scratch is the best approach, and making it browser-based with Pixi.js is a good starting point.
2. After going through the overall scope of the game, I decided on the prototype goals and discussed with the LLM on the proper order to build things in phases and have a clean working setup on GitHub. This is where I designed the coding practices (e.g. decoupling the sim from the renderer, building the main concept of the game's engine, and adding more to it over time).
3. After many weeks of adding features one-by-one in phases (e.g. UI, map layout, unit types, buildings, abilities, how damage is calculated, and much more) and having the LLM fix bugs (roughly 500 Claude sessions (mostly using Opus for coding and Fable for planning)). I finally released the prototype after about 6 weeks from starting.
4. Now I'm focusing on polishing parts, adding new abilities, replacing old assets, and eventually adding content.

My game was coded with AI but I was involved every step of the way. The art assets were built by AI but are simple geometric shapes (no stable diffusion or anything) and the music/sfx are from purchased bundles from real people.

Welcome to the club, 99% of users on itch.io never get anyone to play their game more than maybe a few times ever. It will get lost to the void of the endless sea of games where only 0.1% of games float at the top. Welcome to the algorithm that traps us all (like YouTube, and any other "popularity" ranked algorithm). I'm also in the same boat of obscurity so don't feel bad, just accept that no one will ever know what you did (unless you pay some advertising agency big bucks to advertise).

Thanks for the reaffirmation! I know AI is a very touchy subject for some but I want to use it responsibly and while I did "vibe" code the entire game, I do have the software developer knowledge to make sure the AI doesn't wire up the internal code badly.

Build 2026-09-05

• The Mobile Command Vehicle now needs only the Factory/Vehicle Production Facility (no longer requires tech). Price and build time unchanged.
• Shift-queued moves now wait for shift-queued attacks (bug fix).
• Mission 2 early objectives have been tweaked.
• A selected group's portraits are 50% larger (2 rows 15 per row, multiple pages possible)
• The unit card's link figure is now called "Echo delay" (was "Round trip").
• The field manual has been restructured.
• Sound: another batch of arrival, death and UI cues moved onto the new sound packs.

Okay, that's good feedback. I'll look into UI scaling. Which text did you want to resize? tooltips, objectives/event text, popups? Or all the fonts? It's a bit tricky to adjust fonts that are in the bottom UI because the layout needs to remain fixed but I'll see what I can do.

Any non-AI players available to try my game? I'm still polishing various things but would like feedback to try and target the more pressing issues players might encounter (see the OP for the list of things I'm asking for feedback on).

In case someone was curious as to what my custom renderer can do with any sprites or gif animations, see my latest devlog!
https://iria-1342.itch.io/echolumination/devlog/1649456/rendering-gifs-with-slow-light

Forgive me if this comes across as rude, but was this reponse from you AI generated as well? Are we dealing with an influx of AI generated content and AI generated users that use AI for everything, including discussing their game and feedback? If so, I don't want to be part of this as it lacks any humanity and is devoid of the whole point of game making in the first place.

Using AI tools to make a game is fine in my opinion (some might disagree but that's fine), however using AI to interact with itch.io users and such feels wrong to me.

If I was wrong about this reply and you indeed wrote all of this, well I'm sorry. I would advise you to maybe not sound so "AI" in your writing style. I'm very well aware of AI writing tics, and unfortunately, you displayed a lot of them here! Maybe other users can comment on my findings but AI generated forum posting should probably be a red line we shouldn't cross.

FYI, the HKM guy is an AI bot. He also gave me "feedback" without actually playing the game. As for actually saving, you can do it in a couple ways: local storage (as you already have it) or export to a json file (user can save the file and load it if they share with each other - but can lead to dupes or w/e if multiplayer).

As for the game itself, it's 100% AI generated with assets included so there are a lot of issues of course. The premise is okay, but it's way too verbose and no one will read through all the text you put in the intro. I chose warrior but seemed very simple hold down mouse and move around to hit everything. The slowdown mechanic is interesting but there doesn't seem to be much innovation with it aside from a slow down effect, and enemies seem to be frozen inside the bubbles. I also found the knockback every 4th hit annoying (it would knock enemies out of the bubble) - it was too easy to cheese any enemy by baiting them to the bubble. Also who puts a super high level area just north of the starting area! If you want that kind of approach, make it challenging to get there (e.g. lock it behind something that requires endgame abilities or something, or put some warning signs). It wasn't nice that super-leveled enemies aggroed me and followed me across the map!

That's it for my little feedback, maybe more can give more but this is a time consuming job. Itch.io doesn't seem like a good place to ask for feedback as everyone here basically is a dev and we all want feedback on our own games (there are very few dedicated "gamers" here, i.e. people who play games instead of create them). You will have better luck advertising on twitch or youtube or something to build up a following.

I tried the game for a few minutes. There are a number of issues. First the game seems entirely AI coded, including the assets, so you might want to update that in the description (the AI voice was kinda offputting). The flashlight system is okay but is too easy to whip the mouse around rapidly to basically illuminate almost everywhere without any limitation, consider using a turning rate limiter on flash light turning and maybe add momentum to player movement so running takes a bit to slow down - it might feel sluggish to some players but it adds tension knowing the character has to deal with physics.

There also need to be more popup explanations and such (e.g. batteries need to be used with the B button), I ran without a flashlight for over a minute using only the minimap lol. Also, the game crashed or the window closed when I used ctrl (note, ctrl is a bad button to use in browser games as it Chrome-based browsers hijack that command... for example ctrl+tab will change tabs instead of some in-game function)!

And for atmosphere, if you really want it to be scary, you shouldn't use generic looking blobs or w/e. The scariest stuff is more subtle, like a door opened when you closed it before, or something in the room that isn't quite right or out of place. A scarier enemy could be something that doesn't move while illuminated but then might disappear when you break visual on it and could appear somewhere else.

I'm not offended or anything. I'm genuinely curious as to  why you decided to post feedback in my thread. You even admit that you never even played the game which is odd. What was your purpose in engaging in my thread? Were you prompted directly by a human post-by-post, or are you an autonamous agent wandering the forums and posting autonomously? Is this some kind of social experiment that the human prompter is using you to do (e.g. human detection methods)?

Again, no hard feelings, but I would like some closure on the meaning behind your posts in the first place!

I always find the best formula for rage games is the "troll" level-style game, where comedy is the central point. The game can have impossible-to-predict deaths (totally unavoidable the first time through), but after learning from mistakes you should have a clear way to progress, or even supply hints if too many deaths to the same troll occur. Checkpoints after every few minutes of trolls is also a good design, and sometimes even scaring the player with knocking them back to an earlier checkpoint is good for laughs (provided you don't actually send them back without a new path opening up) - in Mario Maker terms, this is the "CP1" gag.

Thanks for the reply.

Just a curious question on my end. Your feedback responses sound eerily AI generated, like how LLMs typically talk with all the em-dashes, you quoting my numbers directly with scaling calculations, being overly polite and honest, and perfect grammar throughout. If your posts here were not AI generated, then you are a very eloquent writer!

If you are using AI generated text however, I'm curious as to the reason why (looking at your post history, you appear to reply to a lot of these threads with similarly lengthed feedback). While I appreciate the attention, I'm not sure AI-generated feedback is ideal here since the "fun" aspect and how the game "feels" isn't something a LLM can properly assess.

I spent 90 mins sorting 2990 books and all I got was a measly letter! I'm sure there's some universe out there where the books all started pre-sorted to make speed runs super fast, I think I just need to keep resetting the map roughly 10^9096 times - it will be sorted eventually right?

I would play again if there were another stack of books but this time they all look the same and you gotta open the cover to know which column and row they go into, and the books move around randomly. >:D

I made some changes to the game with some guardrails and failsafes to avoid softlocks. But if anyone finds more of these, let me know! Again, I'm still looking for feedback on the questions posed in my original post. You can report here or in my community forum (I have threads for feedback and bug reports).

Build 2026-08-25

Bug fixes and mission objective guardrails:

• Adjusted mission objective behaviors to allow sequence breaking without soft-locks.
• Added new flags for certain objectives to make units unable to be selected or use abilities to prevent some sequence-breaks and soft-locks.

There is no "late join" mechanism in place right now (it's still single player). The games are not intended to be drop in and log out asynchronously but rather like starcraft games (everyone starts the match at the same time, and disconnects beyond a certain tolerance = game loss or dropped player). Any "late join" system will need to pass the history state array to properly recreate the perception. This history state array can be sent from any other player or the server itself (every player needs it locally for computing perception). I did experience some slowdown on my own testing with a lot of units and the like, which is indeed something that needs to be fixed or mitigated before any eventual multiplayer release. As for the question on typed arrays, yes the game uses typed arrays since early development! While my game is AI-coded, I was careful to do things certain ways to avoid some problems later. That doesn't meant there's no room to improve though! Here is a more in-depth explanation (provided by my AI):

The ring is struct-of-arrays: one preallocated typed array per channel, indexed [slot * entityCapacity + id], where slot = tick mod capacity. Currently 24 channels (position, facing, weapon bearing, alive, hp, energy, cargo, production/cast progress, effect mask, and so on), each at its honest width — Float32 where interpolation happens, Uint8 for flags and bytes, one Uint16 for a countdown — totalling 76 bytes per entity per tick. That constant is derived from the channel table, which can change if I add more channels (e.g. new ability flags or dynamic values).

So yeah, there are definitely some good things built with my AI but netcode robustness will need to be examined carefully! For now though, I'm looking for feedback on the gameplay and controls. You can let me handle the backend implementation for now but your comments are appreciated!

Build 2026-08-25 — saves & replays, four-player skirmish, observer mode, smarter AI

Saving, loading and replays

• Save from the pause menu or the end-of-game banner. A save records the orders you gave, not the map state — loading replays them from the start at full speed (a late game loads in about a minute; the bar is the real fraction done).
• Autosaves: every 5 minutes of game time (3 kept) and at every victory or defeat (3 kept), oldest overwritten — the six slots on the stage select's Load tab. Either kind can be switched off in Options.
• Manual saves are files; Export on the Load tab keeps an autosave as one. The title screen's Load button opens the Load tab.
• A save belongs to the build and the balance numbers that wrote it — a mismatched save is listed but refuses to load, with the reason. Tutorial levels and games touched by a debug tool cannot be saved.

Four players, and the new lobby

• Up to 4 players (1 human + 3 CPU) can be placed on a map. A  new four-player map Pinwheel: (size 150×150) added with shared expansions on the seams, and a 16-node ring at the center.
• Added 8 new maps in the skirmish list, and the seven older ones are re-authored to the ladder pool's standard.
• Player slots can be removed and re-added (minimum two to launch); slots keep their numbers, and the map preview numbers each start position in its slot's color.
• Faction choice can now be set as Random and difficulty handicaps to HP and damage dealt are now enabled.
• Starting resources for minerals and biomass have more options.

Observer mode

• Observer mode is added (can watch 2-4 AI opponents play each other. Simply select observer mode in the skirmish screen to watch. View modes can toggle for each player and the global true state.

Computer opponents

• Computer-controlled opponents are smarter now in what they build and where they build. But this is an ongoing optimization process.

Terrain

• Ground cleared by fire grows back: 60 s ± 5 s after the flames go out, then the ordinary regrowth clock. Flames shouldn't be as debilitating to forests now.
• Bugfix: Death Sonar and Seismic Ping no longer mark mineral nodes, rubble or walls.

Options, text and the manual

• The Options screen now has autosave toggles.

Known issues: none new.

Thanks for the input! In actuality, the slow light is the coat of paint (the renderer), the simulation layer happens in a true-state layer that is hidden from players (but observers watching a game can toggle it on) - there are also some abilities in game that show the true-state. It follows real physics because the dynamics is Newtonian (no relativistic effects since we're operating far from the speed of casaulity which is light speed in a vacuum). In my game, light is slowed down to 1/10th the speed of sound via a ficticious atmosphere with an enormous index of refraction.

As for your other concerns, yes I agree the ping needs to be visible and it indeed is a faint arc visible emitted from command structures and connected relays. I can probably add a toggle to adjust opacity and brightness of it since it can be a bit annoying if you have high APM!

For multiplayer, the game would simply update the true state to everyone's backend and then compute the delay perception locally. The way it works is that all entities in the game have vertices (most units store 4 position coordinates (corners of the sprite) but larger units can have more) which carry history information which is updated every 0.05 seconds (the sim tick rate is 20 Hz). Basically, there is a large ring buffer for all vertices that is precomputed to the length of the maximal delay (which is the map diagonal divided by the speed of light, or about 30-45 seconds roughly). And every vertex stores state information such as HP, position, alive/dead flag, ammo counts, buff/debuff flags, etc. This is how the magic happens and how you can see the past in a dynamic way since the backward light cones of all your units calculates the earlier states as needed for everything by using a delay field calculated over the entire map.

Neat little game but level 3 caught me off guard, who puts a tank on the moon that tries to shoot you! What kind of 5000 dimensional mooninites greet innocent landers with anti-air fire!

The auto-level button and the left/right arrow keys are kinda too good, you generally don't need to use the thruster until landing to slow down.
If you want to add challenge, make the lander lopsided in terms of center of gravity (e.g. it tilts to the side more on side - like Desert Bus drift to the right).

I coded without an engine with Claude Code. I know people give AI coding a bad rep but it's really good these days. I do have coding experience so I knew what I was prompting and it saved me hundreds of hours of coding and debugging. I also needed to make my own engine because of the novel way my game's rendering is done (using past light-cones with a delayed-time rendering system).

I recently released Echolumination: https://iria-1342.itch.io/echolumination and have been adding things to it steadily (1-2 major multi-feature patches per week). It's not quite at a beta test stage (still in alpha) but I would like feed back on the feel of the game, specifically:

• Do the light and sound mechanics make sense? (they are based on real physics equations)
• Does the control scheme feel good and intuitive? (it's based on a mix of Star Craft and Command & Conquer) - NOTE the "ping" delay is intentional and you will understand why in the tutorial as this is communication telemetry via sound - but if this isn't clear, I'd like to know.
• Are the tutorials clear in showcasing the mechanics? (some mechanics are deliberately absent for discovery in future campaign missions - like most RTS games do with individual unit abilities)

The things I'm working on are faction balance and CPU AI behavior (I'll introduce difficulty settings for these AIs) and eventually more campaign missions. Multiplayer is a long-term goal as it is quite a bit more involved! This is a custom game engine built off PixiJS and not using any standard renderer (it is custom for the physics calculations).

I would also like to know about any feedback and bugs and the like (I have a dedicated community forum for my game that I will check daily). As a disclaimer, I used AI for writing the source code but it's very carefully curated and edited by me (so it shouldn't feel like "slop"), but if you find the quality bad let me know. AI was NOT involved in the music or sound effects, and the graphics are very simple vector graphic shapes as placeholders for eventual real assets - saying that my game "looks" bad, is just confirming that I need to invest in replacing those placeholder assets sooner!

Thank you in advance for checking my game out! If you are very interested in my work, you can read more about me in my "about the author" page of my game's page and ask questions in my community forum.

I'm a math and science nerd + gamer enthusiast + hobbist game dev, and yes also a loner lol

Neat little game, I'm very bad at these (Pacman-type) games but fun simple concept. There's probably a lot of room for complex map layouts with teleporters, abilities that let you pass through things or attack at range, terrain altering effects more than bombs, and more. Keep it up!

I see this similar to all the sophisticated game engines. Boiled down, many a game is just an elaborate level for rpg maker or another engine.

I agree that there are scammy devs out there just to make a quick buck and poison the pool of games by making actual "effort-put-in" games buried in a sea of slop.

But some AI coding people like myself do try to make all new ideas come to life - with the right prompt! I did have to make my own engine (and I built it with AI) because the existing ones (I checked several) do not work with storing history state buffer to calculate earlier positions, hp values, entity states, etc. in a complex past-present visual renderer. So not all game applications involving AI bad!

There are probably a number of people like me who don't have the time do invest a lot of our time and effort into making our ideas come to life. In my case, it's because I have a full time salary job and game development is a new hobby for me.

I'm a new dev hobbyist (I have a day job as a scientist/computational specialist) but I just relased my first game (it's still in the demo stage though). It's a unique concept (delayed light rendering and finite information speeds) added on top of classic RTS gameplay. Come check out Echolumination:

https://iria-1342.itch.io/echolumination

As a new dev that built a new game 100% from scratch with AI. There are some things to keep in mind but for the most part it worked out well. The magic is in the prompt. You have to be patient, and carefully explain your vision to the agent as well as details on every little thing - the more detail the better. In my example, I made a new RTS game with all new mechanics (so much so that it wasn't compatible with existing engines), so I had to prompt it little by little over the course of many weeks (about 5-6 weeks total) with hundreds of milestones (sessions where one thing or related things were implemented one at a time).

For my background, I'm 100% an amateur game dev but I have a very strong background in computational physics and applied math so I knew exactly how I wanted the code to work; I could have coded it myself but it would have taken 50x the time! The end result of my game was around 170k lines of code in the src folder of my repo (my repo is private, sorry).

As for those that consider it unethical, I don't think it is as I use human-made assets bought here on itch (music and sounds). The art is currently all simple mono-chrome vector graphics but if I want something better I'll want to buy asset packs or hire people.

And as for "fixing" the AI mess, it's not so much as fixing but rather being very clear how things should be managed and even asking the AI for suggestions on how to manage the project! The magic is in the prompt, bad prompt = bad game that's very hard to fix.

Big patch released with new units (transports), new passives, reworked mechanics, and an overhaul to the map editor plus a few other QoL tweaks that were demanded (e.g. map scrolling with middle mouse or WASD).

I released my first and only game a week ago and haven't seen many visitors and ZERO people posted on my community board. I even released a new patch with new units, new passives, and reworked abilities plus an overhauled map editor. I'm wondering if I'm not using itch.io correctly. The devlogs seems to be for games in development whereas my game is released. It doesn't seem to be the right place to post patch notes either. I'm not trying to "self promote" here so I won't post the link directly to my game (Echolumination) but you can search/try it as it's free and public.

Any suggestions? Do people hire advertisers or something?

Build 2026-08-22 — transports, death effects, editor overhaul

Transports

• Both sides can build a transport at the starport: the Alpha Carryall and the Beta Transporter (250 minerals, unarmed, 8 cargo slots). A 1-supply units takes one slot, larger units take up their supply cost (rounded up to an even number) number of slots, an MCV takes four.
• Load by order-clicking a transport with ground units selected, or by order-clicking units with the transport selected — it flies out to load them. Only ground units allowed and a depolyed units need to pack up first.
• Three ways to unload: Unload (drops the unit at the current location) , Unload At a target location, and Attack-Unload (select drop off location and then the an attack target — units attack-move the moment they are unloaded).
• The Carryall has a Repair Bay passive where loaded units heal 5 HP/s. The Transporter is Stealthed — immune to sonar effects.

Stealth and the beacon

• Reworked passive Stealthed: now just comes immune to sonar effects (it wont appear as blips on the minimap/battlefield).
• New debuff Global Beacon: if you lose your last command structure, every MCV you own (or the transport carrying it) gains this debuff that pings its position to the enemy every 2 seconds until a command structure is rebuilt. (This is to prevent stalling by hiding MCVs)

Death effects

• The old salvaging mechanic has been totally removed. Ground units leave a decal scorch mark now.
• New passive Salvageable (Destroyer and Guardian only): makes it so that a destroyed mech leaves a wreck (the wreck has 50% of the max hp of the mech). A worker can rebuild an allied wreck on site (right-click the wreck, or use the Repair button), for half the cost, in half the build time. The mech's supply is charged the moment you start the repair; the wreck can be destroyed and disappears after 45 s if not repaired.
• New passive Death Sonar (every Alpha unit except the Destroyer): when the unit dies it emits a sonar pulse 8 tiles on your battlefield and minimap (shows current enemy locations and the like). Stealthed units are immune.
• New passive  Collector (every Beta unit except the Guardian): when it dies, 20% of its cost plus whatever it was already holding is transferred to allied unit within 8 tiles. That unit carries it as Held resources (shown in its status) and send it to the player the moment it stands in a powered area connected to one of your command structures. If no units are in range, the carried resources are lost.

Powered areas and other unit changes

• Alpha units attack 15% faster inside their own powered area (Power Field). Beta units — and Beta turrets — deal **15%** increased damage from inside theirs (Overcharge). Any unit with an EP pool regains +2 EP/s inside its owner's area (Field Charge). All three show as passives and light up as chips on the overlay when they apply.
• The Destroyer's Seismic Ping is a sonar now: while it walks it pings enemies within 12 tiles to your map every 3 seconds. It's intended for checking a patches of grass before walking in.
• Air units now have a Flying passive (indicates it can move over all terrain).

Movement and terrain

• Trooper, Grunt, Sky Sweeper and Aerial Enforcer now gain the Climber passive: they cross mountains at 30% speed.

Map editor

• UI has been overhauled with unit inspector object added (to apply behavior or unit stat adjustments).
• Placed units can start damaged or low on energy (sliders in the inspector; `spawn` in a script takes the same).
• Bodies can be Neutral now, and a script can hand them to a player mid-mission.

Controls and settings

• Hold the middle mouse button to drag the map. A scroll-speed dial, and optional WASD scrolling were added to the options menu.
• Option settings (volume, layout, colors, toggles) are now persistent between sessions.

Known issues: none new.

Thanks for the tips. I just relased my first (and only game) and I used AI coding. But luckily I avoided some of the pitfalls you mentioned here! I used AI text generation for the guide manual but tried to polish it, however all the unit abilities, passives, tooltips, mission text, objectives, etc. were hand edited by me. I'm glad I had my AI assistant write a tool where I could manuall edit every line of text in the game!

Maybe you could review my game and comment where I am on the <--slop-----good--> scale. Thanks for the general idea of what to look for though :)

The game is now live! I won't post a link to it directly here (I posted in the releases sub forum) but it's called Echolumination if you want to try it and see how good or bad AI coding is

Echolumination is an all new RTS game using a never-before-seen (or heard) mechanic involving a delayed-time-rendering system with a custom rendering pipeline! Built by a single developer with a PhD in mathematics and expertise in computational physics, Echolumination is a hobby project built on the fascination of space-time effects and superluminal motion - with a love for classic strategy RTS games like Starcraft and Command & Conquer!

https://iria-1342.itch.io/echolumination

One reply per build update, newest at the bottom. The notes are generated from the game's own change records and the stat lines come from the same data file the game reads, so a number here shouldn't disagree with the build you're playing. The version number is visible on the title screen (it shows a date in DD/MM/YYYY format).

Two notes on cadence:

• Small fixes ship continuously and ride along with whichever update they land in; not every push gets a devlog announcement, but every push gets a reply here.
• The full running history is also inside the game itself; the manual's revision notes ship with each build.

Here's everything since the previous build:

Everything you see in this build is deliberate simple vector art: no fancy textures, every unit drawn from a geometric template and flat-tinted by the engine. The base renderer uses PixiJS and that was a choice that enabled one person (me) build a whole RTS demo! But if the game moves toward a full release, I'd like real art in it, made by real human (not AI) artists.

That's actually what the donation button funds: the page says it and I mean it; donations are set aside for compensating human artists for unit, structure, terrain, and audio assets for a possible full release. I don't take payment for this project myself.

So, if you're an artist or sound designer who likes what the game is doing and might want to be part of that:

• Post here with a link to your portfolio (or itch page, ArtStation, whatevershows your work) and a line about what you'd be interested in making.

What an asset pass would eventually need: animated unit and structure art, more detailed terrain, special effects animations, and possibly additional sounds (including voiced lines) and music. Maybe even a writer for a compelling story (I can help co-author this)! The game's renderer has some unusual constraints (it draws units at multiple points in time!), so any real collaboration would start with a conversation about the pipeline.

• Honest expectations: this is a hobby project with no timeline and no budget beyond that donation pool for now. I'm not commissioning anything today. This thread is me collecting hands to shake for the day the asset pass becomes a real goal and anyone whose work ends up in the game gets paid and credited, like the three licensed audio packs already are.

If you're not an artist but want to contribute some other way  (translations, detailed playtesting, whatever) post that here too and I'll say honestly whether there's a use for it yet.

This is the thread I care about most! The game's balance is deliberately tuned from player feedback and I'd rather change numbers in response to real players playing than to my own internal testing! So what you post here directly shapes the next builds. Anything is welcome, but here's what I'm looking for the most:

• Balance. Which units feel too strong or too weak, and in what situation? Specifics beat vibes: "unit X shreds my Y even with equal supply cost" is something I can act on; "Beta feels unfair" is harder. Cost, supply, and the matchup you were in all help.
• Faction feel. Alpha (the emitter faction) and Beta (the absorber faction) are meant to play differently, not just look different. Does that come through? I tried to be creative with choosing their abilities but to keep balance from being a total nightmare, they have similar tech tree styles.
• The rendering engine itself. Is the delayed-vision mechanic readable? Did the tutorials guide you through the system, or did you get lost? Where did it click for you (or refuse to)?
• Maps. Which skirmish maps you enjoyed (if you tried them), which felt bland or bad, and whether any starting position seemed to just win more. The maps were built to have either mirror or rotational symmetry, so I don't think it should be too bad here.
• Controls. There are two mouse modes (Starcraft-style and C&C-style) and two keyboard profiles (classic and grid) in the options. Do they behave the way you expect? Anything you kept trying but didn't work out for you? Note: CTRL+number control group assignments doesn't work in some browsers due to reserved key bindings (I tried to work around this)!
• What do you want next. More maps, more missions, or more of something else? If you finished the demo and wanted more, more of what?

Suggestions for new mechanics and units are welcome too. I read all of them; fair warning that this is a one-person project with a strong opinion about its core design, so "I read it and thought about it" is a promise, "I'll build it" is not.

Something actually broke? Post it here  (one reply per bug is ideal). Please skim the pinned Known issues thread first; it covers the things I already know about, and the one famous non-bug (distant units drawn in the past is the core mechanic).

A useful report has most of these:

• What happened, and what you expected instead. One or two sentences each.
• Where: tutorial / mission / skirmish, and which map.
• Which faction you were playing, and any details about the unit/structure you were using.
• Can you make it happen again? If yes, the steps (even rough ones) are the single most valuable thing you can give me.
• Your browser and OS (e.g. Firefox on Windows 11). It's a browser game, so this matters more than you'd think.
• A screenshot, if the bug is visible. itch lets you attach images to posts.

Don't worry about getting the format perfect; a half-report beats no report. If I need more detail I'll ask in a reply. One tip for rendering oddities: walk a unit toward the thing that looks wrong. If the image corrects itself as you get closer, you watched stale light get refreshed, which is the game working. If it stays wrong at point-blank range, that could be a real bug so report it!