pouët.net

introspec: It's good to attract more graphicians and musicians to our scene. How could it be bad? But we need to pass the sacred torch of democoding to future generations to avoid our extinction as nerdspecies! That's why the focus in code is of capital importance when we face the problem of find new blood.

There, you said it! :)

You see, my personal experience must be somewhat different from yours. I did a lot of coding for ZX Spectrum in the 1990s, so I can safely say that I have had a first hand experience of proper oldschool coding. Of course, I also saw oldschool demos (mid 1990s on ZX Spectrum is mostly megademos and some early trackmos). And most of the time I just did not care for it! Surely, I was impressed by certain effects, but there was nothing in there for me. I learned a few programming tricks by watching the demos, that was it. That kind of stuff just failed to engage me. However, when I decided to start coding for ZX Spectrum again recently, I found a totally different demoscene, the scene that had heart and soul and brains and panache to put it all together into a small work of art. Now, I am wondering, could it be that I am not the only person who needs to have a bit more than a pure code in their demos? I put it this way explicitly, because the kind of demo that I like, it is not about just having a good graphician or musician (god, I hate these word). Ideally, it is about a true collaborative work involving coder and an _artist_ and a _composer_. Good code is still important, but it cannot really replace all of the above. It cannot replace the lack of clear and distinct idea.

My favourite comparison is to the director of photography. You cannot make a film without one. It is pretty common to have great films filmed by great directors of photography, but it is not actually a requirement. A decent one is often enough. At the same time, there are numerous examples of films made with a great directors of photography on the team, yet leading to very mediocre overall results. I believe that code is just that. It underpins good demos and it enables more to be done. But the code alone is not sufficient. Sorry! :)

Code is omnipotent but also is a baby learning to walk. We face the extinction of our art before it could grow and run.

Remember, executable graphics are made with code. Music can be made with code.

The omnipotence of mnemonics is indisputable.

Long live the Zilog Z80! :)

Thanks for that link! I didn't know Displayhack yet.
What other sites are there?

introspec is right. The limitations of the oldschool platforms are pretty well explored by now, nobody is going to be impressed by yet another tunnel or spinning object, so I guess the oldschool scenes are having to sort of resort to art.

Ultimately I think people should be into the demoscene because it's a place and an artform for people like them. If we're having trouble finding people who're nerds that like computers and want to learn, build their skills, enjoy and create technical art and design, and meet other people with a similar mindset then holy shit what are we doing? Where did all the engineers go?

err, reading that back by "sort of resort to art" I didn't mean to sound so dismissive, I meant they now have the freedom and impetus to focus more on artistic expression than on the purely technical side of things.

Some "new blood" here (I don't like this expression very much). I just started making demos together with some friends this year. I'm only 19 years old, born 1995 (the 2001-9-11 was my first day in school (yes.. you are that old :D)). By now I am totally thrilled by the whole demo part of the demoscene and the party side adds just perfectly. I love watching a really well done prod where every aspect just plays perfectly together with the other ones.

The first time a demo got my attention was when netzpolitik.org wrote about the revision and linked on the YouTube version of Gaia Machina. I was really stunned by this little animation, even I didn't realized the technical aspect. It was only then that some guy from a bigger demogroup and also from my hackerspace made a little demoshow at a small chaos-computer-club-event. He explained a lot of the technical details and the whole culture and also showed some really awesome prods.

The parties and people he talked about fascinated me. And so, after watching some more intros and the live stream from revision 2013, two friends and I decided, yes we can found a new demogroup, yes we can overcome those awkward graphic cards, yes next Easter we can be in Saarbrücken and yes yes yes we will have an ass-kicking release we could show off to the whole world.. or at least to pouët.

But, yes.. what should I say, have you ever written some OpenGl?! And that's a big part of the problem I think. Not OpenGl, but how to get started in the first place. How to open a window, how to draw some simple spheries, how to make some simple postprocessing (I needed a whole damn week to get this fucking framebuffer and texture thing working). I didn't know and haven't even got started with something that you could call a demo. I think what the most beginners would have loved to see are some more detailed instructions how to get started writing demos, some descriptions of the very basic mechanisms and most of all more code, a lot more code. Generally, share not only the beautiful results of what you're doing, share also the magic that creates this beauty! Show how you're creating the awesome! Let us learn from it! Coders share your shader and framework code, composers share your synthesizer instruments and projects and artists share your sketches, ideas and inspiration. We will find them, dive into them, take everything apart and create something new, something different from them. You just have to show us what and how. Because starting without an idea of what to do is the most difficult part. It's already sufficient if you just dump your work somewhere without any explanation. A simple "here, just take it!" is better than "naa, that makes no sense, you wouldn't even understand".

Some may say, there are already enough tutorials out there explaining how to render triangles or make a good soundtrack, this is right, but it doesn't help anything. Demos are not the same as fancy games or cgi animation movies, they are demos. They are different. They are more. And that's why we all love them.

tl;dr: The "right"/engaged people will find the demoscene by themself and decide that it would be awesome to make a little demo, too. At this point we must not lose them to the "I have totally no idea where to start". Explain more of the basics of your work and share how exactly you're doing it. That would be awesome!

Although the C64 scene manages to still make code relevant. While you think they have explored everything, at every next year, they even make new effects or crazy improvements. Of course, the X winner demo are also full of art and music too, so of course it's the total that matters (the top ten demos at pouet are in every aspect master pieces, but also have great code)

To put it in another way, when we ask for code in demos does it mean hightech? Good looking effects even if simple code behind? Or just realtime graphics?

I remember this discussion about Batman Forever and Vanity. And while some of the CPC scene was saying many of the effects are not very advanced CRTC stuff, I have seen great presentation and new and/or really well done effects in BF, that can impress even those who know nothing about the CPC. But even me who knows something, I loved BF more than everything. It might not have some hidden extreme CRTC tech, but has unique presentation and shows new effects. New or original effects is also code for me, even if the technique behind is simple.

Well, I do not want to appear argumentative, but it is a kinda yes and no, simultaneously :) People who follow old platforms closely know that these dogs keep learning new tricks all the time. New types of codes, new ways to abuse video or sounds chips, all kinds of tricks. Progress is still continuing.

The issue is different. This progress does not matter too much anymore, because the average level of sophistication of imagery that we see on a screen (film or tv) has exploded in the last two decades. Just watch reels of TV breaks or promos from the 1980s or 1990s. The competition has increased ten fold. An average viewer is used to sophistication and the things that were cool in the 1980s are not really cool anymore. A display hack in the 1980s was cool because it could pull off things comparable with what was happening on the TV. Your average ZX Spectrum cannot possibly produce an image that would be competitive nowadays. Hence, one has to somehow justify the loss of quality, i.e. one must introduce stylization. Immediately, this is not a problem of coding, this is a problem of design. The same can be said about the sound design.

So, a more artistic mode of expression does become a matter of necessity for older platforms. I do not see this as a bad thing though, I see this as an interesting creative challenge, creative in both technical and artistic sense, and the interview mentioned by 4mat is very indicative in this sense. Hexagon is a huge inspiration for me in terms of both game design and music. Implementing something like it on an 8-bit platform is a proper technical challenge. One can have it all, clearly!

Optimus, I think this can be better called as "effect design". E.g., as you are surely aware, many old 8-bitters have screen-mode limitations (so many colours per so many pixels on ZX Spectrum or C64 or limited frame rate of CPC etc etc), and designing effects around there limitations is a fascinating mixture that requires understanding of both code and design.

Of course, Batman Forever is fan-tas-tic. No other word. And the criticism it was receiving for e.g. its rotozoomer code is precisely the kind of attitude I tried to talk against in this thread.

second that!!!!!!!!!!

I didn't saw the rotozoomer code criticism, anyway BF got 3x or more thumb up and CDC (;P) than any other CPC demos. The author wrote in the demo that he didn't use the full CPC potential. (CRTC learning...) His goal was probably to crush the previous technical demos with a big production, making use of his Amiga experiment.

Doesn´t feel sceenish at all to me, and even the name sucks. Sorry but true... =/

And I doubt that it would have attracted me if I were a non-scener interested in gfx coding, cgi visuals or sth. like that - it´s neither a tutorial nor a frequent updated showroom or some kind of community.


Apart from that, there is a huge difference between just "making people aware of the demoscene" and actually getting them to be a real part of it.

excuse me maali, but you do indeed sound like a bitter old man. you're blaming metoikos for using someone else's demotool. wait wut?!?! aren't you the guy who has been doing exactly the same for over a decade now? if not please post links to some prods you made all by yourself... :)

There isn't even competition, at demoparty cheapish beer 24/7 vs. expensive beer 9:00 - 03:30 (at best, in finland)!

Serpent, oh, I get it, one just needs to introduce similar (or stricter) regulations for alcohol sales in other European countries and then the demoscene will be saved! :)

Yes, I think it helps a lot to know someone who has coded a demo before, who can teach you the basics and answer your questions.
Getting a good book on computer graphics can help as well (the classic was always [url-http://www.amazon.com/Computer-Graphics-Principles-Practice-Edition/dp/0201848406]Foley and Van Dam[/url], but that may be a bit outdated now, since we no longer render in software).



Agreed.
I have released the custom file format, 3dsmax exporter and a simple OpenGL-based renderer a few years ago: http://sourceforge.net/projects/bhmfileformat/
While it is not a full demosystem, it does show you how to export some less-than-trivial data (skeletal animation with skins and bones) from 3dsmax and render it in realtime.
You can also use the code to get started with OpenGL quickly. You already get a basic framework with functionalities to compile and run shaders, perform mathematics, and load textures.

Perhaps I should extend it with some render-to-texture functionality as well... It has no direct use for the current sample, but I have the code in my OpenGL-codebase anyway. That, and some streaming texture stuff as well, for animated/movie textures.

That would be Foley and Van Dam by the way.

Urg... Third try: Foley and Van Dam.

Not sure about that. There is already a lot of demo code out there, both for demos (e.g., from Preacher) and demotools (e.g., Werkkzeug). But that might not be easy to know/find if one is new to the scene.

Maybe what's missing on top of that are tutorials that explain the code? Although there are seminars on some of the basics as well (e.g., from Preacher again).

A couple of points:

1.) The demoscene used to be driven by competition rather than collaboration. Hence, "sharing" the tools and tricks of the trade are not part of the scene's DNA. That slowly begins to change.

2.) I think you overestimate what you could gain from (even more) source code. Most demo code only shows you how not to do it. The older the code, the more likely that is. And there's literally nothing in there that you can't find on the web, in books and tutorials. The only "scene-exclusive" tricks that I can think of are size-coding and/or oldschool platform related or very dated and irrelevant. But I agree that at least the important stuff should be open-source, if only for history's sake and proper archiving of what has been accomplished so far. That too, slowly begins to change. After all, it only took Mr. Lim 15 years to open up Impulse Tracker. ;)

can i just point out that there's a logical flaw in some of the arguments here that go like "my personal experience of joining the scene or attracting people to the scene is this so this would work to attract others too" etc.

you're thinking microscopically. if anything those approaches could actually be wrong, because they are already employed and have failed to attract significant numbers of new active productive people. those approaches are not working. the trickle that comes in is less than the stream that leaves, gets bored, gets too busy, becomes inactive etc.

In the golden days of the demoscene computers were so limited that you basically had to code to do anything worthwhile. Today computers and tools are powerful enough that there's no inherent reason to write code in order to be creative. The demoscene's mindset doesn't reflect the change however, it still obsesses over code. The message to outsiders is that the demoscene isn't about creativity, it's about code for code's sake. When a community is stuck in the past and hung up on traditions that don't have relevance to the outside world, outreach is the least of its problems. For the demoscene to appeal to creative youth it has to change and make room for them. A bunch of old farts who cross their arms and say "oh, but that's not a DEMO" when something doesn't conform to their sacred idea of form isn't going to win anyone over.

i.e. this