I was a 16-year old ambitious boy once, learning code, not visual editors. It's not hard to follow up a 25-chapter Lite-C workshop with a 7-chapter shader workshop -- especially since the first 4 chapters work well on their own (they were originally written to end there) and are what I learned with. I was stoked to find I could write shaders in the exact same environment I produced everything else with.
Shaders in Gamestudio are written in HLSL -- that's the industry standard for writing shaders in DirectX (along with Cg which is identical). The engine generally doesn't matter when it comes to writing shaders -- some are weird and have proprietary means to produce shaders, but they'll usually include options for HLSL or Cg.

If someone really wants a node-based editor, it wouldn't be hard to produce one. Make a user request or something and list the kind of nodes you want. I'm sure you'll find someone interested in making one.

Jibb

Again, while defending your position you always miss my points. I understand you perfectly and I also started as a programmer. I learned programming languages very fast and loved to code.

But this is not what I am talking about. When creating a game that earns you only a couple thousand dollars, you have to speed developing at every corner. So it is better to use ready-to-use shaders and to use tools that are fast and responsive, deliver real-time feedback and publish to different platforms (not only Windows, even when your CG=HLSL definition says otherwise, the entire product often has to be ported).

Also a shader is not the same on another rendering hardware, especially when you switch to consoles or mobile devices. I know that from the Vision manuals, that there are some special conditions when you release a product to the Wii, iOS or other platforms.

So what is your point? That Sama should use Unity? Or UDK?

As a Gamestudio user developing a game (between studying at university) that has consumed no more than a few days of shader programming over 8 months of game development (and I don't mean a few days of cumulative hours, I mean a few days during which I spent some time on shaders), I don't see why you're talking about platform portability and speedy development.

Next time you tell me I missed your point, please follow that up with a summary of what your point actually is. It'd make things easier for both of us.

Jibb

Wow. Talk about hijacking a thread to argue a point. Why not let Sama recruit for his game(s)?

What!? And not get the final say!?

No, fair enough. This definitely got out of hand. Sorry, Sama.

Jibb

lol, agreed.

I've been watching this thread more since the hijacking than when it had first started... (and before anyone points out the obvious, I know it's now got more posts!)

Yes, you are right. Sorry for hijacking. But if you still missed the point I will summarize it for you and I will give you a little outlook in the end:

I really believe that everything you wrote works fine for you. And I am sure that people like you, who love to play around with shaders from scratch will agree with you. I would do as well if that would be my focus.

But I saw you trying to convince new users like Sama all the time to write their own shaders. And that is fine. But often it will not happen. There are several reasons for it. Most users need to write shaders only once per project and it is not the top priority. They will not learn every hint and trick. A workshop does not teach them the background of 3d lighting theory, matrix manipulation, variable types and much more. Even when they successfully wrote their own shader, probably the workshop shader with only a little adjustments, then they will continue on their project and forget about that very fast. And next time they have to start from scratch again. So I doubt that they write shaders in minutes, even when that works for you.

Another reason is that there are 2 kind of beginners: Some want to make a small realistic project. They dont want to write excessive graphics features. They never care about these shaders. The other ones are dreamers and want to have a modern up-to-date project. And they dream of great shadows, post-processing, HDR, DOF SSAO and even global illumination. The workshop will not teach them how to write that and they will not accomplish something like that in minutes. But they might see that some technology offers this right at the beginning.

So what is it all about? Basically: Most users will not write the graphical effects they dream of as fast as you said. And they dont want to write another normal mapping shader. They expect something old like that as the basics already included. They dream of something more advanced and will probably not accomplish it.

If you are honest, you will see that exactly this happens around us. The more advanced shader developers like Hummel, BohHavoc, HeelX and Slin also did not make such great shaders only from a workshop. They learned alot from Nvidia or from PDF research documents. They learned much more than only HLSL. The programming language is always the smallest issue, the knowledge behind the algorithms, the theory and the optimization strategies are the real asset, that has to be learned over time. And you often picture it, as if they can get all this in a few hours.
But the point I wanted to make is that many users dont want to start a journey over months or years without progressing on the real game.

Another fact you brought up is, that it would be easy to write a node-based editor for Gamestudio. I dont know how long you have been a Gamestudio user, but I remember that there was an official project like that. There was a module-based shader editor. I was even able to test the beta version. It never saw the release day. Maybe it is not that easy.

But on the other hand, this would be a great product. I know from the sales figures (e.g. from the Unity Asset Store) that tools like this sell much better than model packs. I know people who just wrote a framework for an iOS engine and they quit their day job and live only from that. But this also depends on the amount of engine users. I have no figures how many active users develop with GS.

matrix, and vector math is a bitch.

I agree with Frank and btw: I offered jcl to write a node based editor, not only for shaders but also for an entire pipeline (setting up render targets and such stuff). But he refused. And I think I understand why. You can create advanced rendering frameworks with GS but eventually it lacks of engine internal optimisations. It simply doesnt make much sense. Acknex is thought to be an engine for small projects like casual games, simple shooters/RPGs or prototypes and jcl want to keep that and thats fine imo.

I can confirm this. I was working with two people on a sci-fi project, but we had to cancel it two weeks ago. We were using shaders for nearly all surfaces and gamestudio simply couldn't work with that with a good frame-rate. Support told us that we couldn't to that with Gamestudio. We tested a bigger scene also using shaders on all surfaces in Unity and it worked at more than 100 fps.