Re: [PATCH 1/2] wined3d: Correct required extension for R8G8B8A8_UINT.

[Stefan Dösinger]

> Am 13.05.2020 um 17:40 schrieb Henri Verbeet <hverbeet(a)gmail.com>:
>
> Big-endian CPUs, probably. Not that that's particularly relevant these
> days, but IIRC that's the main reason to avoid GL_UNSIGNED_BYTE, at
> least traditionally.
> And at least once upon a time only packed types were fast and using a seemingly equivalent GL_UNSIGNED_BYTE format resulted in no-op conversion or PBOs not working properly or something along those lines.