Radeon HyperZ In Open-Source On Older Hardware

[phoronix]

Phoronix: Radeon HyperZ In Open-Source On Older Hardware

Recently there has been a lot of talk about Hierarchical-Z/HyperZ support for R600g since its set to provide a measurable performance benefit the open-source Radeon driver for modern GPUs. This support has still not been mainlined and there are still a few issues to work out, but at least the support is stable for the older "R300g" driver and has been around for quite a while at this point. For those wondering what to expect from HiZ/HyperZ performance boosts, here is a preview.

http://www.phoronix.com/vr.php?view=17084

[Adarion]

That's the good thing in free software. Even users of older chip get some new cookies from time to time.
And it shows that Hyper Z in use makes sense not only in theory.

Is known what caused the strange behaviour in Nexuiz?

[FireBurn]

Were there cards PCIExpress 2?

[jakubo]

Were there cards PCIExpress 2?

i think PCIe2.0 begins somwhere with HD3400 cards (see wikipedia)

actually calling it new cookies is... dunno...

its just reinventing the wheel. closed drivers had this feature since the r100 days.

yet im very thankful and happy for my old cards to work an a new OS with my old hardware. thanks to the devs!

[mattst88]

its just reinventing the wheel. closed drivers had this feature since the r100 days.

Sort of... if the old wheel had its tire slashed and then was thrown in the dump.

[russofris]

That's the good thing in free software. Even users of older chip get some new cookies from time to time.
And it shows that Hyper Z in use makes sense not only in theory.

Is known what caused the strange behaviour in Nexuiz?

Let me start by saying that I do not want to diminish the efforts and contributions of the developers that worked on adding support of this feature to the FOSS drivers, nor do I disagree with the content or spirit of your post. Their effort is to be applauded, and I agree that it is nice that users of older hardware can now benefit from this tech once again. I would like to clarify that the only thing "new" about this particular cookie is it's inclusion into these particular FOSS drivers for these older chips. The tech is by no means new and it's usage in conjunction with these chips is not new (even in terms of FOSS drivers).

Bravo.

F

[Paulie889]

I find it hard to believe that there were no corruptions or visual glitches in the tests. My RV530 is buggy as hell when RADEON_HYPERZ is set. However its possible that R520 works fine.
BTW why testing two almost identical cards?

[whizse]

I find it hard to believe that there were no corruptions or visual glitches in the tests. My RV530 is buggy as hell when RADEON_HYPERZ is set. However its possible that R520 works fine.
BTW why testing two almost identical cards?

On my RV570 it was a sort of mixed bag. A lot of games worked fine, others only with major glitches and some with only occasional problems.

Also, it should be pointed out that this is not exactly a new feature, I think most of the hyperz work in r300g was done by marek early/middle of last year?

[smitty3268]

On my RV570 it was a sort of mixed bag. A lot of games worked fine, others only with major glitches and some with only occasional problems.

Also, it should be pointed out that this is not exactly a new feature, I think most of the hyperz work in r300g was done by marek early/middle of last year?

Yeah, and he gave up on ever turning it on by default because he couldn't get to the bottom of the glitches across all the different hardware that's supported.

Hopefully r600g has better luck.

[keivan]

I have one of those r300 cards and I want to test HyperZ. Is there anybody knows how to enable it?
Thanks in advance