[ATrpms-users] 200 fps on my FX 5200
jarod at wilsonet.com
Fri Apr 8 03:35:18 CEST 2005
On Thursday 07 April 2005 17:38, Preet Khalsa wrote:
> Sorry Jarod for posting this to the list, but I need more eyes on this
Not a problem, best to have the conversation archived for the world to
reference anyhow. :-)
> Finally had a chance to reboot so here are some results.
> > Huh, wonder what the difference is w/the RHEL4 x86_64 kernel. Where do
> > you see that error? Maybe I'm not looking in the right place, and I'm not
> > using NvAGP, but I'd swear I am...
> NvAGP = 1 and dmesg gives me ...
> NVRM: not using NVAGP, an AGPGART backend is loaded!
I get nothing in the logs on RHEL4 x86_64, but performance is identical with
NvAGP =1 and NvAGP = 3. I'm guessing its actually AGPGART all the time, just
> Other people have mentioned this same problem on the Myth list.
> > > I am at the latest MSI BIOS. That was the first thing I did. However
> > > I am now going to poke around in the settings. There is a "Cool and
> > > Quiet" setting that I enabled because this is a 24x7 Myth box. Maybe
> > > it or something else is squelching the AGP. I know that when I tried
> > > to overclock this beast, the FX 5200 barfed and shut down. I had to
> > > reset the CMOS to get started again.
> > I suppose it could be something in the BIOS interfering... I presume you
> > have fast writes and all that good stuff enabled, but yeah, maybe
> > something with the cool and quiet setting mucks with agp...
> Turned Cool and Quiet off ... still no results. 199 fps at best.
> Any more ideas from people. This card should give at least 1,500 fps.
> Could is be a problem with the x86_64 platform in general? I switched
> expecting great results. So far ... heartache.
Its certainly possible x86_64 plays a part in it, but I'm getting over 7100
fps on an x86_64 box (w/a GF 6600 GT).
jarod at wilsonet.com
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Size: 189 bytes
Desc: not available
Url : http://lists.atrpms.net/pipermail/atrpms-users/attachments/20050407/08dc56ce/attachment.bin
More information about the atrpms-users