Unintended downclocking on a GTX 260
Moderators: NeilBlanchard, Ralf Hutter, sthayashi, Lawrence Lee
good work
Great work there, I wouldn't have thought of the screen saver. Since all my systems are at home, I turned off the screen savers on all of them just to keep from losing the extra cycles, so I guess that explains why I haven't seen this. But nice job figuring out the problem.
-
- Friend of SPCR
- Posts: 381
- Joined: Fri Aug 01, 2003 12:13 pm
- Location: Rochester, MN
now I've got this
Hey, sorry to resurrect a thread, but now I seem to be having this problem under linux. I took a couple systems and swapped motherboards in them, so I basically wound up taking my system with a 295 and a 275 and moving them from my cosmos S case to my fortress f-02 case. And in the fortress case, they seem to be baking, temps go right into the high 80s for the 295 and stay there, and my ppd on the 295 gpu jobs drop in half. Not sure what I can do about it under ubuntu, today I'm trying it with the top panel of the fortress removed, maybe that will let more air flow out. I'll probably wind up having to move them back to the cosmos case. You know, I would not have predicted that the cosmos case kept the cards cooler than the fortress. But it sure is crappy to see the 295 running like a 260.
confirmed
So I swapped the cards between the systems last night, and in the cosmos case the 295 is chugging along at full speed, no overheating. Bizarre. I really thought that the fortress ft02 would be better at cooling the graphics cards. I'm going to have to post something about this over in the case forum. Huh.
-
- Friend of SPCR
- Posts: 381
- Joined: Fri Aug 01, 2003 12:13 pm
- Location: Rochester, MN
This thread isn't dead.
No problems on my GTX 260 in a while.
Changed one of my boxes to a Asus P5N-D (socket 775 with nvidia chipset, sli capable) and now have 2 9800GT EE cards with accelero coolers on them folding along with a Q9550 at stock. Nvidia drivers are 197.45.
Riva Tuner can't seem to force the second card (gpu 1) to stay at a constant performance level. The first card (gpu 0) will typically stay at the performance 3D clocks that I specify. I also specify it for the second card, but it will drop back into that really low performance mode but still fold away. I thought this stuff was behind me. (Get behind me, downclocking!) I've tried evga precision and it doesn't seem to act any differently, either. Anybody have any suggestions?
No problems on my GTX 260 in a while.
Changed one of my boxes to a Asus P5N-D (socket 775 with nvidia chipset, sli capable) and now have 2 9800GT EE cards with accelero coolers on them folding along with a Q9550 at stock. Nvidia drivers are 197.45.
Riva Tuner can't seem to force the second card (gpu 1) to stay at a constant performance level. The first card (gpu 0) will typically stay at the performance 3D clocks that I specify. I also specify it for the second card, but it will drop back into that really low performance mode but still fold away. I thought this stuff was behind me. (Get behind me, downclocking!) I've tried evga precision and it doesn't seem to act any differently, either. Anybody have any suggestions?
Hi KansaKilla, if you're only folding on the card and presumably 24/7 so don't care about the downclocked low power mode and just want performance for folding then maybe modify the BIOS so there are no "down clocks!"
I found it easy enough to mod GTX260 BIOS with Nibitor, NVFlash and a DOS bootable flash drive. (Flashing under Windows x64 is hit n miss)
Regards, Seb
I found it easy enough to mod GTX260 BIOS with Nibitor, NVFlash and a DOS bootable flash drive. (Flashing under Windows x64 is hit n miss)
Regards, Seb
-
- Friend of SPCR
- Posts: 381
- Joined: Fri Aug 01, 2003 12:13 pm
- Location: Rochester, MN
Hi KK, I don’t have a guide as such for BIOS editing.
USB Boot:
If you search for [url=http://www.google.co.uk/search?hl=en&q= ... =&gs_rfai=]â€
USB Boot:
If you search for [url=http://www.google.co.uk/search?hl=en&q= ... =&gs_rfai=]â€
-
- Friend of SPCR
- Posts: 381
- Joined: Fri Aug 01, 2003 12:13 pm
- Location: Rochester, MN