Comment by bane | Hacker News Reader

Comment by bane | Hacker News Reader

Parent Full thread

The demoscene operates very differently from other software communities in that it's inherently competitive. That means that different coders, or groups, try to keep what they're doing (and how) a bit secret so as to gain a competitive edge.

Most of the coders I know who produce for the scene honestly start with books on graphics or sound and work their way up from there. It takes years to produce a collection of effects, techniques and code that you can pull from to make a great production. Early productions by people new to the scene are usually fairly simple with only a few effects or scenes, but if they keep at it will grow from there.

That being said there are some scene "tutorials" that can help:

https://github.com/psenough/teach_yourself_demoscene_in_14_d...

https://in4k.github.io/

http://www.sizecoding.org/wiki/Main_Page

ftp://ftp.scene.org/mirrors/hornet/code/3d/docs/

https://bitbucket.org/inclinescene/demosplash-2016/src/maste...

https://www.reddit.com/r/gamedev/comments/205r78/sight_light...

http://www.hugi.scene.org/online/hugi35/hugi%2035%20-%20demo...

https://4matprojects.blogspot.com/

https://www.pouet.net/topic.php?which=10882&page=1

This is largely not the case anymore. Lots of democoders include their sources or put them on github. Plenty even do in depth write-ups on how they made it.

In particular, nearly all PC 4k coders are super active on shadertoy and most of their tricks appear there first, all in the open, before they roll a full-fledged 4k with it. It

And I dare say that in the past, the competitiveness as a reason for not sharing code was mainly a cheap excuse. The real reason was that everybody was ashamed of the terrible messes their codebases had inevitably become.

> The real reason was that everybody was ashamed of the terrible messes their codebases had inevitably become.

Haha true, especially for code written at the partyplace in a rush.

I agree with you as well, the scene is definitely opening up and sharing more than in the past. There's some pretty good talks that occur at parties these days, not quite an academic conference, but getting there in some ways.

If you're talking about the past-past (say up until the mid-nineties) IMHO you're wrong. Pre-mass-internet there wasn't really anywhere to share code. People were swapping floppies, or dialing bulletin boards. The competetiveness was absolutely real.

Also, pride.. If you wanted to figure out how an effect was done, either you reverse engineered the binary, or you figured it out on your own, based on your own experience and the occasional hint.

You have to remember that the demoscene largely grew out of the cracking scene (well, that + Compunet) in the mid-eighties, where _being first_ to release a cracked version of a game was of great importance. Likewise, being first to create a new effect for a demo.

That piece of culture was extremely short lived, maybe just the late eighties and the early nineties.

Already in the mid to late nineties people started mocking this culture, eg the MFX Transgression intros which were just 3d polygon renderers but with some uglifying post processing and then they claimed it was super impressive real-time raytracing (which, for no context, was basically impossible to do full screen on computers of the days).

Some stuck up kids blamed them for cheating but generally everybody laughed at the prank. The demoscene has been open, friendly and collaborative much much longer than it was properly competitive. It's really a minor origin detail, not a key part of culture and ethos.

People do still like to make fun of it, i.e. pretend is still current. Eg common proclamations such as "Kewlers suck", or demos like "Regus Ademordna" which IMO is a particularly splendid example and worth checking out.

Reply on news.ycombinator.com