pouët.net

So I guess that makes:
4klang: 14
Oidos: 7
Clinkster: 2
Other: 2 (0x4015 and Juippi)

Interesting question: Is this reflective of the4k synth distribution overall?

BTW: Blueberry, Juippi's synth is named ghostsyn but since no-one else seems to be using it, I guess its 'Custom'.

I must say that there's some truth to the "they're using the same synths" complaint, even though it's not really the synths' fault. I can sympathize with people who just use the same presets, because creating good-sounding modular synth patches is too difficult for most musicians even in top-class commercial environments like Reaktor, and 4klang's user interface adds a layer of obfuscation to everything. It's a big investment in learning.

I'm trying to make even small usability improvements in my fork. To be honest, 4klang's user interface doesn't try very hard to make it easier to understand what's happening. ;) With relatively simple effort it would be possible to show the signals, what's coming from where and going to what unit. I have this on my todo list:

The indentation level comes from the unit stack height, minus UnitPreSignals[]. Arithmetic/FOP is a special case, but anyway. Indent the texts/boxes, draw lines and arrow heads, etc.
I don't really expect this to increase the amount of people making their own patches, but at least I'd like it for myself.

("unit stack height" should have been, fpu stack, but I'll fix that when we can edit posts)

Apart from SDF domain-repeating/fractalizing tricks, many intros might look the same, because of another hidden "enemy", which is grid-based noise (gradient or not, combined more or less obvious way, doesn't really matter, you can still clearly "smell" it in almost every scene).

As for synth - it's really similar. There is only so much you can do with few modular synth components, even if you add some post-fx, compressor, simplified Karplus-Strong etc.., even then you can still hear the good old saw wave buzzing somewhere in the background.

The final reason is that pretty much everything on a demoscene look and feels the same. It's what we call "demoscene art". It's about some common aesthetic choices. There is a reason topics like Demo'ish videos and Demo'ish objects are active for so long.

But still, every once in awhile, someone makes something new with the same simple components. The limitation is you. Spend more hours and quit whining.

yzi: limitation is the size, not a person ;)

And of course I agree with you that it is an art to put simple components in a new, creative way, but there is a realistic (or even combinatoric) limit to it. Hence, even if a scene is put together in fresh/creative way, there are usually some obvious signs it's just yet another variation of the same thing. Observing that, I've never said it's necessarily a bad thing (at least not for me). There is still plenty of challanges to make something impressive in 4k. And personally, I am actually often impressed and excited to watch 4k intros in recent years. The problem is that the group of people that can recognize the novelty and be truly impressed is shrinking to small group of insiders that are also working on small intros. And unfortunately, for everyone else, it might in fact look "all the same", which is of course a result of an oversimplified view, but nevertheless an honest feeling, which we should respect or even be concerned about.

I think that the limitation is really in the programmer. Check out this guy, he makes incredible stuff with FM synthesis.
https://www.youtube.com/watch?v=P6CgLXrkMXI
https://www.youtube.com/watch?v=vdDofVcAII4

What prevents you from creating a new rendering algorithm? Not the GLSL language or the size limitation, that's for sure. What prevents you from making better sounds in 4klang? What prevents you from modifying it to be more versatile, or new structural designs? Not the amount of bytes available. Even the current state of compression level isn't maxed out, because there are untried alternative ways to structure the synth data and shuffle things around. You just need to spend the hours really.

yzi: this FM synthesis is indeed incredible. Thanks for the link.

Of course you are right. I totally agree with you that you need to spend hours, if not many weeks/months to do something truly unique. But at the same time that kind of proves my point. GPU is mostly suited for trivially parallelizable algorithms, so if you don't try hard enough, you will end up with commonly used techniques that are proven to be lightweight, take little code to render complex scenes (i.e.fractals) and work relatively well in real-time (for small intro standards). And, as usual, the number of people that are trying harder is pretty low, so we end up with lots of 4ks that have similar look&feel.

One needs new ideas. Eureka.
Ideas to give direction and value to a production otherwise it is utterly pointless to bring a technical achievement that can’t be appreciated because it looks vague.

You already know this I know. <3

this is also Clinkster <3