Comments on: Reading DX10 docs… http://aras-p.info/blog/2005/12/16/reading-dx10-docs/ Random thoughts of a triangle pusher Mon, 07 May 2012 11:19:17 +0000 hourly 1 http://wordpress.org/?v=3.3.2 By: Lost in the Triangles » Blog Archive » Speculation: pipelining geometry shaders http://aras-p.info/blog/2005/12/16/reading-dx10-docs/#comment-184 Lost in the Triangles » Blog Archive » Speculation: pipelining geometry shaders Sun, 25 Mar 2007 23:22:40 +0000 http://aras-p.info/blog/?p=81#comment-184 [...] « Reading DX10 docs… 64k coding continued [...] [...] « Reading DX10 docs… 64k coding continued [...]

]]> By: kris http://aras-p.info/blog/2005/12/16/reading-dx10-docs/#comment-105 kris Wed, 21 Dec 2005 11:31:00 +0000 http://aras-p.info/blog/?p=81#comment-105 ReJ: I'd rather wait for a first dx10 card. Besides, now I have one game project at work, one at home and university :). ReJ: I’d rather wait for a first dx10 card. Besides, now I have one game project at work, one at home and university :).

]]> By: ReJ http://aras-p.info/blog/2005/12/16/reading-dx10-docs/#comment-106 ReJ Wed, 21 Dec 2005 09:14:00 +0000 http://aras-p.info/blog/?p=81#comment-106 NeAraz: why separate VS/GS, let's do some wild guessing here:<BR/>a) easier to program I suppose, backward compatible in some sense (easier to reuse legacy vshaders)<BR/>b) maybe easier for compiler to schedule<BR/>c) more options for underlying hardware - could be one block, or could be separate blocks, but with 'multithreading' to hide stalls<BR/>d) ... something really important NeAraz: why separate VS/GS, let’s do some wild guessing here:
a) easier to program I suppose, backward compatible in some sense (easier to reuse legacy vshaders)
b) maybe easier for compiler to schedule
c) more options for underlying hardware – could be one block, or could be separate blocks, but with ‘multithreading’ to hide stalls
d) … something really important

]]> By: NeARAZ http://aras-p.info/blog/2005/12/16/reading-dx10-docs/#comment-107 NeARAZ Tue, 20 Dec 2005 19:18:00 +0000 http://aras-p.info/blog/?p=81#comment-107 ReJ: if VS+GS would end up as a single piece of hardware, then why would it be exposed as two separate stages?<BR/><BR/>Analogy with PS2 makes sense though :) ReJ: if VS+GS would end up as a single piece of hardware, then why would it be exposed as two separate stages?

Analogy with PS2 makes sense though :)

]]> By: ReJ http://aras-p.info/blog/2005/12/16/reading-dx10-docs/#comment-108 ReJ Tue, 20 Dec 2005 18:55:00 +0000 http://aras-p.info/blog/?p=81#comment-108 Well that works pretty well for PS2 :) ('well' is a relative term of course)<BR/><BR/>I suppose gpu already has very deep pipelines, plus you will specify your maximum amount of vertices statically - so gpu can schedule its work ahead pretty well.<BR/><BR/>And probably :) it's not impossible for gpu to have more than one independent vertex pipe running at the same time.<BR/><BR/>Or VS/GS could be linked into the one uber-shader underneath (much like PS2 vu1 code) and executed in the same block - could be no worries about the VS stalls then.<BR/><BR/>Kris: no need to wait - go write vu code :) Well that works pretty well for PS2 :) (‘well’ is a relative term of course)

I suppose gpu already has very deep pipelines, plus you will specify your maximum amount of vertices statically – so gpu can schedule its work ahead pretty well.

And probably :) it’s not impossible for gpu to have more than one independent vertex pipe running at the same time.

Or VS/GS could be linked into the one uber-shader underneath (much like PS2 vu1 code) and executed in the same block – could be no worries about the VS stalls then.

Kris: no need to wait – go write vu code :)

]]> By: NeARAZ http://aras-p.info/blog/2005/12/16/reading-dx10-docs/#comment-109 NeARAZ Sun, 18 Dec 2005 08:56:00 +0000 http://aras-p.info/blog/?p=81#comment-109 Yeah, GS is a very interesting piece...<BR/><BR/>Though I can't really imagine how the hardware will be able to pipeline everything efficiently (when each input primitive can end up with variable number of primitives).<BR/><BR/>The possibility to do this is great, but I think most of the interesting scenarios will be pretty slow as well. Of course, "slow" is a relative term (i.e. it'll probably be faster than the alternative methods of doing the same thing) Yeah, GS is a very interesting piece…

Though I can’t really imagine how the hardware will be able to pipeline everything efficiently (when each input primitive can end up with variable number of primitives).

The possibility to do this is great, but I think most of the interesting scenarios will be pretty slow as well. Of course, “slow” is a relative term (i.e. it’ll probably be faster than the alternative methods of doing the same thing)

]]> By: KriS http://aras-p.info/blog/2005/12/16/reading-dx10-docs/#comment-110 KriS Sun, 18 Dec 2005 00:18:00 +0000 http://aras-p.info/blog/?p=81#comment-110 I just can't wait to write geometry shaders :). I just can’t wait to write geometry shaders :).

]]>