modern 64k intro development
category: code [glöplog]
Watching the latest Revision entries got me thinking... what is the "best way" to do a 64k intro nowadays and how does a proper toolchain work in 2024? I did my share of intros back in the days long long time ago, but obviously the world has moved on since 2006.
A 4k intro is most often one shader plus a minimal framework to load it, but would that approach be feasible for a decent 64k intro too? A bunch of minified shaders bound together by a glue framework for windowing, syncing etc and a synth (what are good options available? sointu? 64klang? Others?). The other option is of course an editor that exports a datafile and a player that runs it. I guess nothing much has changed in this regard in recent years? I used to do content by just handcoding a bunh of stuff, but with modern approach to coding instead of just GL_LINE_SMOOTH and glBegin() it might get crowded pretty fast.
Basically, what I am asking is, assuming I'd be interested in trying out an intro, is there something obvious that I am missing? I haven't followed the current developments at all.
A 4k intro is most often one shader plus a minimal framework to load it, but would that approach be feasible for a decent 64k intro too? A bunch of minified shaders bound together by a glue framework for windowing, syncing etc and a synth (what are good options available? sointu? 64klang? Others?). The other option is of course an editor that exports a datafile and a player that runs it. I guess nothing much has changed in this regard in recent years? I used to do content by just handcoding a bunh of stuff, but with modern approach to coding instead of just GL_LINE_SMOOTH and glBegin() it might get crowded pretty fast.
Basically, what I am asking is, assuming I'd be interested in trying out an intro, is there something obvious that I am missing? I haven't followed the current developments at all.
Hey, im not a coder so cant answer you question.
But zavie recently put together a website dedicated to the 64k-scene that might point you in the right direction:
https://64k-scene.github.io/
And there is a sizecoding discord server: https://discord.gg/pZE5rAQrHx
But zavie recently put together a website dedicated to the 64k-scene that might point you in the right direction:
https://64k-scene.github.io/
And there is a sizecoding discord server: https://discord.gg/pZE5rAQrHx
Thank you! Joined the discord and the website looks like it has links to all the stuff I don't know about :)
There's also our old codebase to scavenge for parts: https://github.com/ConspiracyHu/apEx-public
"A bunch of minified shaders bound together by a glue framework for windowing, syncing etc"
isn't that, what the category 8k is for?
isn't that, what the category 8k is for?
Quote:
isn't that, what the category 8k is for?
No that's for a single shader that doesn't fit 4k anymore.
🔥
I started to play with SqrMelon by Tropical Trevor et al. (Poo-Brain) and it's quite nice https://github.com/trevorvanhoof/sqrmelon
It has some bugs here and there but they're pretty much easy to fix/workaround.
It has some bugs here and there but they're pretty much easy to fix/workaround.
Quote:
But zavie recently put together a website dedicated to the 64k-scene that might point you in the right direction:
https://64k-scene.github.io/
Just to be clear, this was a common effort with LLB, and to be fair, he put more work than I did.
The repository is public, and everyone in the 64k scene is invited to contribute:
https://github.com/64k-scene/64k-scene.github.io
Quote:
Just to be clear, this was a common effort with LLB, and to be fair, he put more work than I did.
Ah, my bad!
Some additional resources written up and published since Revision:
How to create a small windows executable without pulling out your hair.
How to store data efficiently in 64k intros.
The tool we developed for Clean Slate that let us perform deep analysis on the compression of the final release. (The latest rekkrunchy build that goes with this is included in the release)
How to create a small windows executable without pulling out your hair.
How to store data efficiently in 64k intros.
The tool we developed for Clean Slate that let us perform deep analysis on the compression of the final release. (The latest rekkrunchy build that goes with this is included in the release)
Really interesting articles!
Also:
Also:
I asked ChatGPT to write a haiku about AI in the demoscene.
Quote:
Code without spirit,
Pixels dance without feeling,
Art lost in circuits.
Whoops, wrong thread :D
human error.. the only thing that still separates us from machines!
