It seems that IMatRenderContext.SetRenderTargetEx would be the answer to my
question, but can someone confirm this before a waste a whole lot of time by
being wrong? Precisely- If using a pixel shader output such as:
struct PS_OUTPUT
{
float4 PosInViewSpace : COLOR0;
float4 NormalInViewSpace : COLOR1;
};
can I expect that these will correctly end up in the render targets I set, for
example: SetRenderTargetEx(0, posTexture) and SetRenderTargetEx(1, normTexture)?
> From: [email protected]
>
> I'm looking into doing deferred lighting. I'm wondering if it is possible to
> use multiple render targets in source, ie. mutliple target, single pass. If
> not, and I use a separate pass for each element of the g-buffer, what kind of
> performance impact could I expect. The shaders for each pass are quite
> simple, so is most rendering incurred because of complex shaders (meaning the
> transforms and rasterization itself is fast) or is rasterization itself
> costly? Can I expect adequate performance doing 4 or maybe more geometry
> passes per frame if the shaders are very simple? Thanks for any info,
> thoughts, tips, etc.
_______________________________________________
To unsubscribe, edit your list preferences, or view the list archives, please
visit:
http://list.valvesoftware.com/mailman/listinfo/hlcoders
