Reducing power consumption from my video card?

Ecological issues around computing. This is an experimental forum.

Moderators: Ralf Hutter, Lawrence Lee

Post Reply
padmewan
Posts: 106
Joined: Fri Mar 12, 2004 7:56 am
Location: Cambridge, MA
Contact:

Reducing power consumption from my video card?

Post by padmewan » Mon Oct 16, 2006 7:23 pm

I've got an AGP Radeon 9800 Pro, which was pretty power-hungry for its time but a welterweight by today's standards. How much power does this baby draw when I'm just sitting around using Word or Firefox? And does cranking it down the clock/memory with ATITool reduce power consumption in any meaningful way?

jaganath
Posts: 5085
Joined: Tue Sep 20, 2005 6:55 am
Location: UK

Post by jaganath » Tue Oct 17, 2006 2:10 am

How much power does this baby draw when I'm just sitting around using Word or Firefox?
~30W.

Xbitlabs 9800Pro measurements
does cranking it down the clock/memory with ATITool reduce power consumption in any meaningful way?
Heat decreases linearly with frequency, but quadratically with voltage; so you should see some reduction (~5W); remember, every little counts! :wink:

padmewan
Posts: 106
Joined: Fri Mar 12, 2004 7:56 am
Location: Cambridge, MA
Contact:

Post by padmewan » Wed Oct 25, 2006 10:23 am

Either ATITool or a new SATA drive blew my Windows install and I've been out of commission for a week! Horrible, it's like having a sick child or something. Must have spent 20+ hours figuring out how to reinstall (turns out I ALSO had faulty Windows disks)
jaganath wrote:Heat decreases linearly with frequency, but quadratically with voltage; so you should see some reduction (~5W); remember, every little counts! :wink:
Are you pulling my leg???

btw, that is a great resource you're pointing to at X-bit labs. The updated version is here: http://www.xbitlabs.com/articles/video/ ... noise.html

Great to know ppl are out there testing this stuff, whatever the reason (power or noise savings).

merlin
Friend of SPCR
Posts: 717
Joined: Mon Jun 13, 2005 6:48 am
Location: San Francisco, CA

Post by merlin » Wed Oct 25, 2006 11:35 am

padmewan wrote:Either ATITool or a new SATA drive blew my Windows install and I've been out of commission for a week! Horrible, it's like having a sick child or something. Must have spent 20+ hours figuring out how to reinstall (turns out I ALSO had faulty Windows disks)
jaganath wrote:Heat decreases linearly with frequency, but quadratically with voltage; so you should see some reduction (~5W); remember, every little counts! :wink:
Are you pulling my leg???

btw, that is a great resource you're pointing to at X-bit labs. The updated version is here: http://www.xbitlabs.com/articles/video/ ... noise.html

Great to know ppl are out there testing this stuff, whatever the reason (power or noise savings).
Why would he be pulling any legs? There will be reductions with both frequency decreases and voltage decreases. A voltage decrease has a FAR greater effect though. If you want to shrink the voltage, you'll have to do some soldering and also figure out a stable mhz.

padmewan
Posts: 106
Joined: Fri Mar 12, 2004 7:56 am
Location: Cambridge, MA
Contact:

Post by padmewan » Wed Oct 25, 2006 2:48 pm

merlin wrote:Why would he be pulling any legs?
Well, the winky for starters ;) There's that famous joke page about how overclocking will allow your TV card to pull in HD from around the globe, so, anyway...
merlin wrote:There will be reductions with both frequency decreases and voltage decreases. A voltage decrease has a FAR greater effect though. If you want to shrink the voltage, you'll have to do some soldering and also figure out a stable mhz.
Yeah, sadly my mobo allows me to increase my voltage but not decrease it. It's also set to "Auto" overclock by default, so I think that every once in a while when the GPU freezes up for whatever reason and I reboot, the BIOS claims that "Overclocking failed." Anyway, that's not the point, the point is that my mobo allows OVERclocking but AFAIK not underclocking. And now I've stopped trusting ATITool :(

Post Reply