Problem #1258135 with my Sager NP9170 has cropped up recently.
I'm running these games off a Nvidia 680M with dual 128gb SSDs in RAID setup.
I'm having periodic framerate crashes that generally last only 5-10 seconds. This is happening in multiple games of varying graphics quality: Bioshock Infinite, FC3 Blood Dragon, Dragon Age: Origins, Mark of the Ninja (MARK OF THE FREAKING NINJA).
I'm running the 320 beta drivers, although the problem happens in the latest WHQL certifieds too.
When I'm playing these games, my framerate will be high and steady (or a steady 60 if I have vsync on), but at certain times it will slow down to an unplayable, choppy framerate, but ONLY for a few seconds. And it's not like I'm dealing with a problematic area with lots of NPCs, explosions, objects, etc. It just slows down for a few seconds, then reverts back to normal.
It's almost as if the slowdown is happening when the game loads a new area, but that can't be it because I've started to notice it'll happen areas I've been in for awhile.
I've guaranteed the following:
I'm NOT using Integrated Graphics. Power management is set for maximum performance. It happens with vsync on AND off. No abnormal processes are running in the background; no antivirus is running. I recently reformatted my hard drives and the problem persists.
Usage Logs
I ran MSI Afterburner while playing Dragon Age: Origins.
I notice my GPU usage during play hovers around the 70-80% range. But when I have these quick little framerate crashes, my GPU usage spikes to 95-99% for 8 to 10 seconds, then returns to normal.
My GPU temperature was at 75C when I started playing, and continually rose to finally plateau at around 88C, where it stays.
My core clock is at 719mhz and my memory clock is at 1800mhz. And during gameplay my memory usage is 511mb.
As I continue playing the problem gets worse.
Now my framerates are dipping from 60 to around 20 for periods up to 30 seconds.
My core clock during these framerate crashes is slowing down to as low as 135 mhz while the memory clock drops from 1800 mhz to around 300 mhz. GPU usage is hovering around 98% for much longer periods of time.
Temperature, however, is staying steady at around 89C.
-
Also, my fans seem to be whirring at a high rate constantly. Also they continue well after graphics intensive programs have been closed and my GPU temperature has settled down to 70C.
-
sounds like ur gpu is throttling. the 680m lowers its core and memory clocks once it hits 90c.
check for dust clogging in ur vents and heatsink fins, maybe also swap out ur thermal paste, see if that helps with the throttling.
another thing is to raise the back of ur machine with bottle caps, for example, that immensely helps with the airflow
Sent from my Galaxy Nexus using Tapatalk 2 -
-
how to stop the fans once youve stopped the demanding tasks.
open nvidia inspector. wait two seconds and then close it. fans should slow down.
its a glitch that a lot of us have. clevo are aware but still no fix. -
Prostar Computer Company Representative
-
off topic
could you send some summer weather our way please as i think its missing us again this year. one day of 23c last week and were back to storms and 8c at the moment.
back on topic -
Well I've taken up the suggestions in this thread and I have seen improvement. My temperatures are down and performance is up. So thank you everybody. I had no idea my GPU was just getting too hot.
-
It might need a repaste. The 680m is generally cool. I say generally because a card this powerful is bound to put out heat, no way around it.
Try giving it a good repaste if you never have and see if that helps you out. -
Prostar Computer Company Representative
-
off topic again
6c here today and been issing it down all day.
i will quite happily take 20c off you ProStar.
time for a trip to america me thinks.
Periodic framerate crashes in games (NP9170 with 680M)
Discussion in 'Sager and Clevo' started by mrro1, May 10, 2013.