What is the typical cause of thes task results? I'm getting 4-6 a day

---

What is the typical cause of thes task results? I'm getting 4-6 a day

By Running 295.73 drivers, once the monitor connected to the DVI port goes to sleep, the cuda device becomes unavailable, dropping the Cuda app into CPU fallback mode, eventually ending with a Maximum Time Exceeded error,
eithier run a 290.53 or earlier driver, or set the Monitor not to go to sleep, and physically turn it off,

Claggy

Your problem is caused by your use of the nvidia 295.73 driver which has been well documented in this forum.

If you want to continue using this driver then set your OS to never let the monitor go to sleep (you will lose the video card when the monitor goes to sleep) or revert back to an earlier driver.

Cheers.

---

Hmm! Ok. We'll then, i'm getting MORE than 10 a day.
Does turning off the monitor have the same impact? or is it just when windows does it?

---

Help fix the economy - FAIR TAX!
Pledge, to preserve your liberty and mine!

Hmm! Ok. We'll then, i'm getting MORE than 10 a day.
Does turning off the monitor have the same impact? or is it just when windows does it?

Only when Windows does it.

The power button on the front on the monitor has no detrimental effect on the tasks being processed - at least, none that I've ever seen reported here.

thanks.
fyi - it does not seem to have adversely affected the EINSTEIN jobs that I've been running.

---

Help fix the economy - FAIR TAX!
Pledge, to preserve your liberty and mine!

Thanks to Claggy & others. I switched to 290.53

I ran into this with a EVGA GT 440, with the 266.33 drivers when I originally used the DVI output.

Killing the display sleep doesn't bother me.

Now I have the output through the HDMI, but still leave the power feature off.

(my display is a Visio E3D420VX 3D HDTV :Þ)

---

Has anyone heard of an NVIDIA "FIX" ???

Has NVIDIA promised a fix to this problem ?

---

Has NVIDIA promised a fix to this problem ?

Yes. According to this message (by a project developer at Einstein):

Update: this bug is considered as release critical (show-stopper) for the next NVIDIA driver release that's due in 2-4 weeks. Thus a fix will be available by that time.

Mind you, that was posted over 7 weeks ago, so the timetable has slipped a bit...

AFAIK 301.24 drivers don't have that problem. They are beta though.

---

I'm not the Pope. I don't speak Ex Cathedra!

AFAIK 301.24 drivers don't have that problem. They are beta though.

301.32 dev driver is OK too.

---

"Living by the wisdom of computer science doesn't sound so bad after all. And unlike most advice, it's backed up by proofs." -- Algorithms to live by: The computer science of human decisions.

Did a "Custom - (Clean Install) " Of Nvidia 310.24 Beta Driver.
I'll post any problems &/or if the GPU bug has been fixed

---

No "Compute Errors" since the 301.24 Beta install !!!
Thanks to Jason_Gee for the heads-up.

Just did a Clean Install of 301.42.
Will post if any negatives show up

No errors overnight !!!!

---

ZERO errors!!! Nvidia has solved the SETI-GPU problem !!!

---

301.42 (WHQL not Beta) continues to run with no "Sleeping GPU" errors !

If you aren't using an NVIDIA card, what are the other causes of these errors? I'm currently using my CPU only and over night someone kept pushing the power button on my machine to try and get it to come on (it was on, its just very quiet and the monitor was on another source) so there was a lot of powering on/off, does this cause errors? I ask because I woke up today and my screen is filled with computation errors, it seems to be running fine now but there's like 40-50 that have errors.

They haven't been reported yet but my system is usually very stable so I'm wondering if there's anyway that they can be re-run to try them again?

---