DynamicClocks on R200

[Patrick McFarland]

On Sunday 24 July 2005 03:22 pm, Alex Deucher wrote:
> On 7/24/05, Patrick McFarland <pmcfarland at downeast.net> wrote:
> > On Sunday 24 July 2005 11:21 am, Alex Deucher wrote:
> > > On 7/24/05, Patrick McFarland <pmcfarland at downeast.net> wrote:
> > > > I was reading the radeon manpage, and I saw DynamicClocks, and it
> > > > said I could save power, but it possibly could decrease speed in some
> > > > cases. So, I thought to my self, "my machine uses a lot of power as
> > > > it is, so lets turn that on."
> > > >
> > > > Before turning it on, I ran the good ole glxgears, and it gave me
> > > > consistant scores if slightly below 2000 fps. With it on, however,
> > > > glxgears gives me scores consistently slightly above 2100 fps. What
> > > > gives?
> > >
> > > tough to say.  It could be that the default clocks as the bios sets
> > > them up are not quite optimal and when dynamicclocks is enabled, the
> > > high end gets a little higher.  One thing to note, if you have a
> > > desktop chip, dynamicclocks does nothing as the code is only executed
> > > on mobility chips.
> >
> > Oh. Boy.
> >
> > I'm on a desktop using a desktop chip,using a Radeon 9100, lspci says:
> >
> > 0000:01:00.0 VGA compatible controller: ATI Technologies Inc Radeon R200
> > QM [Radeon 9100]
> >
> > So wtf is going on? I've done the test several times, with it on, ~2100,
> > with it off, ~2000. (Not that I'm actually complaining, its just that if
> > it gives more people speed boosts, its worth looking at why)
>
> it doesn't do anything though.  none of the dynamicclocks code is even
> executed on desktop cards.  I suspect there is some difference in your
> operating evironment.
>
> Alex

You should be suspecting glxgears as a bad benchmark.

reboot the whole machine, start x with it on:
diablo at infinity:~ $ glxgears
8768 frames in 5.0 seconds = 1753.600 FPS
10145 frames in 5.0 seconds = 2029.000 FPS
10383 frames in 5.0 seconds = 2076.600 FPS
10397 frames in 5.0 seconds = 2079.400 FPS
10345 frames in 5.0 seconds = 2069.000 FPS
10399 frames in 5.0 seconds = 2079.800 FPS
10516 frames in 5.0 seconds = 2103.200 FPS
10516 frames in 5.0 seconds = 2103.200 FPS
10592 frames in 5.0 seconds = 2118.400 FPS

reboot the whole machine, start x with it off:
diablo at infinity:~ $ glxgears
8472 frames in 5.0 seconds = 1694.400 FPS
10575 frames in 5.0 seconds = 2115.000 FPS
10577 frames in 5.0 seconds = 2115.400 FPS
10571 frames in 5.0 seconds = 2114.200 FPS
10571 frames in 5.0 seconds = 2114.200 FPS
10346 frames in 5.0 seconds = 2069.200 FPS
10322 frames in 5.0 seconds = 2064.400 FPS
10556 frames in 5.0 seconds = 2111.200 FPS

So I think this was a false alarm. Tomorrow I'll probably get entirely 
different values. :(

-- 
Patrick "Diablo-D3" McFarland || pmcfarland at downeast.net
"Computer games don't affect kids; I mean if Pac-Man affected us as kids, we'd 
all be running around in darkened rooms, munching magic pills and listening to
repetitive electronic music." -- Kristian Wilson, Nintendo, Inc, 1989
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
URL: <http://lists.x.org/archives/xorg/attachments/20050724/5a166ca1/attachment.pgp>