Hey guys affection here and before I start with my normal noobie question, I would like to say I'm sorry if anyone is starting to get annoyed by my noobie questions, I've self taught myself everything about programming that I know so I don't have a 'normal' understanding of this stuff so thats why ask ask such silly questions.
So as far as I can gather, Dreamcast uses fixed fuction shaders or something, that is like shaders but instead of being able to create new shaders, you are given pre-sets and you can only modify them. So what can the dreamcast's fixed function do exactly? I know it can do bloom and DoF but what else?
If I have got anuthing wrong here please correct me as I find this extremely intresting.
Fixed Fuction
-
affection56k
- https://www.artistsworkshop.eu/meble-kuchenne-na-wymiar-warszawa-gdzie-zamowic/
- Has thanked: 0
- Been thanked: 0
-
MisterDave
- DCEmu Freak
- Posts: 58
- Joined: Mon Apr 08, 2013 1:16 pm
- Has thanked: 0
- Been thanked: 0
Re: Fixed Fuction
Fixed function is similar to the older versions of OpenGL which doesn't have programmable vertex and pixel shaders. Probably the best example of what you can do with a fixed function pipeline is at http://nehe.gamedev.net
Hope that helps a bit.
Dave
Hope that helps a bit.
Dave
-
PH3NOM
- DC Developer
- Posts: 576
- Joined: Fri Jun 18, 2010 9:29 pm
- Has thanked: 0
- Been thanked: 5 times
Re: Fixed Fuction
I suppose the DC's fixed function does whatever you can program it to do, with a limitation of 200mHz CPU and a non-T&l capable GPU.
All transformations occur on the SH4(CPU), not the PVR(GPU), as done by modern hardware.
"Shaders", by modern terminology, usually refers to code that is executed on the GPU( which is extremely well suited for geometric / vector operations ), as opposed to code that is executed on the CPU ( which is typically designed for general-purpose operations ).
Previously I had theorized some sort of programmable pipeline on the DC, by interpreting the "shaders" into SH4 instructions, but in the end that would require a huge effort to support a large array of functions. This would essentially be a run-time compiler.
Anyway good luck, and I will link you to a resource that has taught me quite a bit:
http://http.developer.nvidia.com/CgTuto ... ter01.html
All transformations occur on the SH4(CPU), not the PVR(GPU), as done by modern hardware.
"Shaders", by modern terminology, usually refers to code that is executed on the GPU( which is extremely well suited for geometric / vector operations ), as opposed to code that is executed on the CPU ( which is typically designed for general-purpose operations ).
Previously I had theorized some sort of programmable pipeline on the DC, by interpreting the "shaders" into SH4 instructions, but in the end that would require a huge effort to support a large array of functions. This would essentially be a run-time compiler.
Anyway good luck, and I will link you to a resource that has taught me quite a bit:
http://http.developer.nvidia.com/CgTuto ... ter01.html