[Mesa-dev] [PATCH] r600g: Add support for ROUND

[Jose Fonseca]

----- Original Message -----
> On 8 August 2011 02:24, Jose Fonseca <jfonseca at vmware.com> wrote:
> > There's no "wrong" or "right" when there are two equidistant
> > integers -- it's all a matter of convention.
> >
> But note that rounding to nearest even is a slightly better
> convention
> in terms of rounding bias. I.e., not using RNDNE is both likely to be
> slower and likely to produce worse results.

It's subjective. It depends on the expected input distribution, which is effectively impossible to characterize in general. One can easily find datasets where one method gives biased results and the other not, and vice versa. And if one takes all possible numbers, they are equally good.

The only thing that could define what's wrong/right here in the OpenGL GLSL / Direct3D HLSL specs, but I don't think they go this level of detail. At least I couldn't find any mention last time I checked.

But I agree with you on the speed.

Jose