pouët.net

Hello there,

I recently discovered the demoscene which seems to be a fascinating universe. I'd like to be part of it by creating something. But as I think one life is not enough to learn code, graphics and music at the same time (starting from a newbie level in each area), I wonder which of these 3 areas lacks the demoscene? Knowing that they all interest me but I sadly don't have time enough to learn all of them.

Thank you for guiding me.

It may be just me but there never can be enough (3D) graphics artists.

Definitely not musicians. If anything there's too many of them :')

you already have the right nickname for 3d artist at least!

graphics artists for sure

first of all, it is about what YOU like to do. what do you think you are going to be good at. the demoscene seems to be lacking of 3d artists at the moment. but the demoscene is not a company where personnel is tasked to man a particular sector because there is lack in it. it is always up to you, your liking, your heart and present skills. (and yes i love to start my sentences with a lowercase letter)

Coders. Especially Amiga coders :)

Demosceners that actually make stuff.

Coders with good ideas and the ability to execute them (either alone or in cooperation with others).

It is hard to phrase it any better than okkie just did.
Just make a demo about it. Use a demotool for example to start with.

I think what the scene lacks besides more demos, is people who are good at making demos. Conceptual artists, directors who 'actually make stuff'.

What Okkie and numtek said.

Also what Defiance said; no use if you waste your coder talent by making bad 3D models or below average music, for example. I think you only get things done if you do what you like and are good at.

Demoscene needs more laser cutters. ;P

I agree with Defiance. What is your love? Or, where do you think you might have more talent?

I didn't expect so many answers, thank you all. Actually I have some basics in coding and drawing. I'm interested in these two areas and it's very hard for me to choose which one I'll lay aside.

Is there any reason for the lack of 3D artists on the demoscene? Looking at the parties results, there are lots of 2D masterpieces in the graphics competitions. So I guess there are enough 2D artists. Are 3D graphics less interesting to do or less "rewarding"?

About the creation of a demo, is it like video game creation: 2D artists drawing concept art which is modeled in 3D by 3D artists? Are there some kind of designers to imagine the demo direction?

When I watch demos credits, it's sometimes confusing between "visuals", "graphics", "art", "design"... is there any nuance I didn't catch?

I think it's that is harder, the tools are less intuitive, the constraints are tougher, and you get kidnapped by the pro-industry faster because they also are in desperate need.

Sometimes, yes, although that's not common. It really depends on what the given group of people are doing.

It's mostly just that - nuances. There's no rule of thumb really. "Visuals" usually means the whole package including animation / compositing, "graphics" and "art" usually means 2D stuff and "design" sometimes only means just a bunch of colors or font selection.

i would slightly rectify gargaj that design would also encompass the general artistic direction of the demo.

It's a bit surprising as I read at various places that 2D graphics require more drawing abilities than 3D. To me, 2D drawing seems more difficult than modeling as it involves perspective and shading. You don't need to master this for modeling (maybe I'm wrong). But I agree, Photoshop seems more intuitive than Blender.

What 3D software is commonly used in the demoscene for modeling? Blender? 3Dsmax? Maya? Special demoscene software?

Could you please give me some demos references in which concept artists and designers were involved in the direction?

The whole point for me at least is to explore various skills and be influenced by them. For example, even though I'm not in any way "traditionally" skilled in holding a brush, I've still found the characteristics of with paint-style coloring interesting and applied them to real-time rendering. Now, I wouldn't have probably tried to emulate them in the first place if I wasn't interested in all kinds of mediums and actually picked up a brush / pen / marker / spray can and played around with them.

Similarly I think it's always a good learning experience to pick up various instruments, synths and audio equipment, especially when nobody is there guide you and instead you've left alone to experiment with them. A good studio environment can be the most inspiring atmosphere there is.

Somehow this all relates to how I originally learned about 6502 / Z80 8-bit processors by tweaking memory values with a monitor, and stepped through some games to understand how they did things internally. The combination of the algorithm and the screen image being drawn really clicked with the way I approached rendering things the way I wanted.

try a bit of everything, stick with what you seem to do best or enjoy the best and preferably both.

Draw with code, then you won't need to lay aside either. And at the end, you'll find that your drawings are alive, you can tweak, animate... or just release it as a demo :)

This is just a theory of mine - and I guess any graphics artist can correct me if i'm wrong - but translating a natural talent for drawing from paper to digital is relatively straightforward, because you can pretty much use the same technique (with a stylus), plus digital gives you a bunch of opportunities to use layers, undo, tint, etc., but you're still just working with a 2D canvas with colors on it. 3D on the other hand is more abstract and super-technical, taking care of topology, unwrapping, baking, there's no real "natural talent" for that. There's a lot to be learned for both but I'd say 3D has a lot steeper learning curve than 2D.

I'd say MAX is the most popular, but that's just my observation, I might be way off. Self-written tools are mostly used for <64k stuff.

This is a thread where you can see some examples

i think it's a lot worse/more so when you add the "realtime" issue into the mix.
more work involved, even more technical skill required, loads of limitations, most of your favourite plugins & things in your favourite piece of 3d software can't be used, and then at the end of it your work gets butchered by whatever engine you're stuck with, and looks a lot worse than any offline render you could have done yourself in the software you modelled it in in the first place.
and after all that the coder will make you entirely retexture it to all fit on one texture / rebuild the skin/rig to use half the bones / make some spikeballs instead.

no wonder there aren't too many 3d artists around in the demoscene. :)

Code something, anything, submit it to a compo, and then you'll know much more than a thousand pouet posts can tell.

cutting: Im guessing you will find what you like the best of these fields as you go along exploring them and find what you skill the most. What anybody else has said about graphic-artist and 3d-artists.



All the above.
There was a time when 3D Studio 4 (for DOS) where used. You can find the imfamous yellow duck.3ds in many oldschool demos and intros. (<- a littlebit of history). But I wouldnt recommend starting on that now :P But it has somewhat the relevant simple theory behind it, which is: that the simplest data you will ever need is the vertex and face/edge- data of your polygons. Your exporter should take care of this. That is before exporting, merge (i cant remember if that's was the correct name) your object-data.

Another software was Lightwave, and mostly used on Amiga afaik. It seems relevant for PC also today though.

Interesting what Smash said. I tried to say something about but merged it above. The point is the engine of the coder probably dont have all the importing capabilities (import-plugins) your 3d-software uses. A coder wouldnt make imports for every feature of every 3d-software. Better talk to the coder about this, and/or learn what export/import plugins means for a coder. The key here is the i/o-interface. One rule of thumb would be something like: stick to pow2 sized textures and merge the objects (make objects the way that you can export the relevant vertex- and face data only, because this is the most important data), and dont use fancy plugins everywhere, because the engine to the coder might not have this.