itch.io

Hello itch.io community!

From what I read on some other posts, the use of AI to develop of a game is at best frowned upon and usually strongly criticized... Maybe there are some nuances to bring to the discussion and that's what I intend to do while uncovering honestly how I came to develop and release Bouncenoid recently, on a series of devlog posts. And like anyone releasing their first game, I'm both excited and proud about it 😅 And yet, yes, I had help from Claude AI for quite a lot of the python code (I used to be a developer a long time ago and I am a data scientist as well) and some help from ChatGPT to improve the initial graphics (I'm definitely not a graphic designer). 

Does it make the game unworthy? I don't think so. Does it make me an unworthy game dev? I don't think so either.

At the end of the day a real "good game" is not a quick demo, a proof of concept and or test, it's a product that should be polished to maximize the players' experience and joy, and I don't think releasing such a "good game" is related to using or not using AI. AI can generate code, AI can generate graphics, sounds, etc. but AI cannot play the game and feel what players would feel, it cannot feel how difficulty is crafted and how much frustration or happiness somebody would have on a given level. That's what makes us game developers, not the other way round where AI would be the game developer. But leveraging great tools to explore more ideas, to speed development is a normal thing. I mean we are all using so many packages, platforms, tools and software developed for us by talented people. Even using itch.io is leveraging a cool tool to improve our development and nobody is saying "the guy has posted a message using a forum he hasn't developed himself, don't read it".

So why is it different with AI?

One of the reaction I read was that devs using AI for their game are lazy. So in fact, it seems that what we don't like is not the AI result itself, it's the fact that somebody might not have put the expected and necessary efforts for their creation. I can relate to that. But I have yet to see and play a game that I feel is a great game, that somebody has put together with AI in no time, not caring much. Personally, the rude part here would not be "using AI", but "not caring". It doesn't matter what AI you use, if you want to craft an excellent game, you need to care about it and about the people playing it. If you don't care, you're probably going to waste my time with an unfinished product, and I  would call you lazy, but not for using AI.

To illustrate my point, I'd like to share a video from a French Youtuber I checkout now and then, called Aywen, it's the one called in French "J'ai laissé 24h à Claude pour créer GTA 6 (c'est une dinguerie)", "I gave Claude 24h to create GTA6 (the result is crazy)". The guy is not a game dev, but he's sharing unusual coding experiments and he released several videos, where he benchmarked different AI to recreate famous games. It's quite fun to watch and there are really impressive stuff about the models, which are able to recreate 3D worlds, AI Characters and sounds and much more, only from prompts (costing hours of compute and several hundreds of dollars). It's all great but nobody is fooled: mocking about with an AI to create a whole game for you is definitely not the way best way to create a really "good game". And Aywen is not saying the result is worth of a real game, it's more about evaluating models' performances, which is useful and interesting.

There, that should set the mood for my devlog series about Boucenoid. What I'd like to share in these future posts, is what I learnt developing a game with some AI help, what worked, what didn't work, what I gained, etc. But no matter what, at the end of the day, I certainly don't consider that Bouncenoid is an AI-generated game at all. It's all my ideas, my sweat, my late nights, my countless hours of testings, my decisions... except for the music that comes from one of my best friends and composer ;-) And all of that is 100% human!

What's your views and experience so far with AI-leveraged games?

JerkyHips

I'm starting this thread so that players can comment on their favorite levels or ask about how to beat difficult ones.

Rules: 

• 1 level commented by post,
• Specify the Zone, number and name in the title, ex: "Zone 2 - Level N15 Minefield" to make sure it's clear for everyone.
• Check whether the level hasn't got a dedicated post before posting a new message.

Hi everyone!

I'm JerkyHips, diving for the first time in the itch.io community for a splash of friendly exchange.  

I guess at nearly 50 I might be some sort of dinosaur here, but one that has just proudly released his first game (Bouncenoid) ;-) I never really stopped playing games (although with ups and downs), but going back to coding one brings a weird time-travel feeling, back in the days where I used to copy code in Basic (or Assembly!) directly from magazines I bought, for my Amstrad and my Apple II C or for a friend's Atari 520 ST... Just one mistake then and you'd get "syntax error" or more obscure messages and you had to try and find it, even if you didn't really understand the whole thing 😅

Fast forward 40 years, after a pro life of C# developer, IS consultant, data scientist and Bio tech program manager, and finding myself suddenly without a job, and in need of a side project to keep my sanity. And what was initially a small toy idea became a full blown game a few months later. And it feels good ;-) I had fun discovering what tools, languages and packages are available now and just tried to do my best.

I don't know yet where this will bring me, but I'm definitely happy to meet the community, try out games and potentially know that people would enjoy my first game.

Cheers,

JerkyHips

Bouncenoid

New players may find the game a bit difficult at first but it's really a matter of getting used to the physics, like many games. That said I think it's useful to point out a few things and provide tips to enjoy it even more.

• All levels are designed to be finished in 1 life only and without Extra-boost. Of course it's not easy to perform on all levels, but with a bit of practice, you'll work out a strategy for each level, and sometimes you will uncover a brand new and easier strategy, even after hours of playing. It happened to me, even watching beginners playing ;-) The Extra boost is nice, but it's actually a kind of "late" feature I added after designing all the levels, to help players. Initially there were only 3 extra-boost and nothing more for the whole game, but then I thought of adding a new type of brick to make the game even easier... And for those who don't use the boost, you'll get a lot of extra points at the end! Last, in all levels it's possible to finish in 1 life only, without Extra-Boost AND picking up all the bonus bricks, but it's tough and risky on some levels.

• Pay attention to your "launch meter" when you throw the Spark: at first you might not care too much for the exact power of your throw, but as you progress, you'll find useful to remember the power value you're using to maximize your chance. For some levels the right value range + / - 5 is the key to your best shot to finish the level in 1 life.

• Study the level before launching and aim for the top bricks: though many levels are obvious, it's a good idea to take a few seconds at the beginning of each level to define you bouncing strategy. As the ball loses height/energy after each bounce (unless bouncing on a special brick), it's usually better to aim for the top bricks first to ensure you'll be able to break them before you're too low. But keep in mind that they are many ways to finish a level, so you'll have to find the one that suits you.

• Arrow keys/ D-pad controls to steer the ball: at first you might press the arrow keys repeatedly to move the ball to the right or to the left, which is fine. But if you keep the keys down (or the D-pad pressed), you'll see that the ball accelerates to the sides, allowing you to reach much further. From the launch pad, it is possible to reach bricks located all the way on the left-hand side, provided the bricks aren't to high on the screen. Repeated key presses work better for small adjustments. Also when you stop pressing the arrow key, the horizontal speed goes down very quickly, allowing the ball to return to a vertical fall. So if you aim for a specific brick, you may stay pressed on the arrow key and release it as soon as you're above the brick to fall on it.

• Play with infinite lives: as the name suggests, the Infinite lives mode lets you play the game with infinite tries on any level, which could be a good way to get to know the physics better and the strategy for each level. Now infinite mode doesn't mean you're going to finish the game with absolute certainty as you might get stuck on a very rare levels if you have used up all your Extra-Boosts and destroyed some key bricks. It could happen on very few levels. So if you play with infinite lives, try not to use the Extra-boost unless absolutely necessary to view the official ending. Even when playing with Infinite lives mode, you'll earn trophies, which is satisfying.

• Practice levels you've already beaten: each time you finish a level during an official game (starting from level 1), the level becomes available to play directly through the "Select a level" option from the menu. This is a good way to improve your technique so that the next time you reach that level in a official game, you get through easily.

• Listen  to (or rather read) the Spark through the Intercom: the Spark is usually trying to help you through your journey as you progress. It communicates with you through the Intercom screen on the right-hand side of the game area. Whenever you start a new level, the Spark will comment it and usually provide some tips on how to approach the level. If it says "full throttle", it usually means that you should throw the ball with maximum speed to finish of level.

• Read the Help pages: all the rules and mechanics are described in the Help pages, so you may find it useful to have a quick glance in case you missed something.

And maybe above all, practice makes perfect, so don't be too impatient to become a Bouncenoid master ;-)  And even more importantly, enjoy the game no matter what.

Happy Bouncing!

JerkyHips

Hey itch.io community!

I'm very happy to share Bouncenoid my first video game! If you like classic 80s based pixel art arcade games, you might enjoy it ;-)

I had the original idea of a "reversed" brick-breaking" game years ago, and I never thought I'd actually develop it. But while struggling to find a new job (not in video games!), I needed a project to keep my sanity and this idea came back to me... It was a side-project at first, mainly to proof-test the idea and try some AI tools to evaluate how much help I could gain. And little by little the game got to me and I started to have so many ideas to add that I finally decided to go all the way and try and craft something a bit polished. And the other important thing is that my best friend and composer Hubert Razack offered to create the soundtrack for the game. From there I had no choice but to create a game that would live up to the music 😅

It's a pygame-based game for now, though I plan to release a web version at some point. Though it's inspired by 80s games, it comes with trophies, many levels, options and other features for casual and hard-core gamers alike. I'd love to receive some feedback and know that people are trying to beat the game and reach the end of the 48 levels, enjoying the live comments from "the Spark", the ball of energy you guide to uncover the mysterious distress signal...

Anyway, enjoy it! Happy bouncing ;-)

JerkyHips

----> Bouncenoid on itch.io