Re: [5/5] WineD3D: Software vertex blending

[H. Verbeet]

> -
>        for(i = 0; i < 16 /* Max vertex attribs */; i++) {

I think you should be using MAX_ATTRIBS there.

Changing the strided data like that in drawPrimitive looks rather hacky to me.

[5/5] WineD3D: Software vertex blending

[Ivan Gyurdiev]

H. Verbeet wrote:
> > -
> >        for(i = 0; i < 16 /* Max vertex attribs */; i++) {
>
> I think you should be using MAX_ATTRIBS there.
>
> Changing the strided data like that in drawPrimitive looks rather 
> hacky to me.

You cannot access the strided data by both named approach 
(u.s.something) and indexed approach.
The two are completely independent strided formats - one is used for the 
fixed pipeline, and the other is used for the shader pipeline.

I think both of you are wrong, and the correct number is 15 + 
WINED3DDP_MAXTEXCOORD, which is the bigger of the two unioned together.
Only modifying the first 16 will break: normal2, tangent, binormal, 
tessFactor, fog, depth, sample in the fixed pipeline case (true, I don't 
see these being used anywhere right now ...).

Please leave writing of the strided data to the functions designed for 
this purpose.

Ivan

[5/5] WineD3D: Software vertex blending

[H. Verbeet]

On 03/05/07, Ivan Gyurdiev  wrote:
> I think both of you are wrong, and the correct number is 15 +
> WINED3DDP_MAXTEXCOORD,

True, I only noticed the hardcoded 16, not that the code's actually
looping over the strided data rather than actual vertex attributes.