Comme=
nt # 65
on bug 11078=
1
from Rui Salvaterra
(In reply to Richard Thier from comment #64)
> Rui!
>=20
> Btw I am also running your patch already too and did not see much prob=
lems
> so far, but many of the stuff I am running was DX7 and DX8 level in gr=
aphics
> settings so I do not know how many times I actually tried your code
> literally, but it is built into my mesa too.
Good to know! It worked fine here, too.
> Also Mount&Blade is working well too now! It=
was misconfigured in the new
> wine to have CSMT and that was =C3=BCber-slow on my single core machin=
e leading
> to starvation and a lot of overhead that was even visible in the perf =
output.
>=20
> So far everything works at full speed now!
Great!
> I am thinking about maybe trying to figure out H=
yperZ on r300. According to
> the feature matrix it is untested and I see bad glitches but elevated
> performance. I already made some changes to try it...
>=20
> But if I continue with hyperZ stuff likely it will be another bug repo=
rt,
> just for sake of transparency about knowing this one is pretty much se=
ems
> solved completely...
Yes, I think it's a good idea to open a new bug.
By the way, I have a vague memory of HiZ on the R4xx being exclusive (i.e.,
only the first app to use the Z-buffer gets HiZ), but I could be completely
wrong.