delerious Posted January 8, 2017 Share Posted January 8, 2017 I've been running the AIDA64 System Stability Test (all tests checked) for a couple hours now. My Windows 10 is configured to put the screen to sleep after 5 minutes of inactivity. I have openhardwaremonitor open to show me the CPU and GPU temperatures. I have noticed that while the screen is not sleeping, the GPU temp stays around 65-67 C. But if I leave the computer for a while and then come back and wake up the screen, I see that the GPU temperature has gone up to around 74 C the whole time the screen was sleeping. Then it immediately goes back down to 65-67 C while the screen is not sleeping. If I leave the computer again, it will go back up to 74 C while the screen is asleep. Does Aida64 do different testing depending on whether or not the screen is asleep? Quote Link to comment Share on other sites More sharing options...
Fiery Posted January 9, 2017 Share Posted January 9, 2017 No, AIDA64 doesn't behave differently in different power management states. It executes the same stress testing code no matter the screen is powered or sleeping. Radeon video cards however work differently in those states, so it must be due to a video driver (Catalyst) pecularity. Regards, Fiery Quote Link to comment Share on other sites More sharing options...
delerious Posted January 9, 2017 Author Share Posted January 9, 2017 My GPU is a Nvidia K5100M. Quote Link to comment Share on other sites More sharing options...
Fiery Posted January 11, 2017 Share Posted January 11, 2017 On 2017. 01. 09. at 9:39 PM, delerious said: My GPU is a Nvidia K5100M. Maybe ForceWare uses similar power-saving tricks when the monitor is offline. I don't know, we haven't got any technical details on that yet. Quote Link to comment Share on other sites More sharing options...
delerious Posted January 12, 2017 Author Share Posted January 12, 2017 If it's a power saving trick, the GPU temperature should decrease rather than increase, no? Anyway, I stopped the test after 4 hours and no errors or crashes. I saw in another thread you told someone that 4 hours should be enough to say that your system is stable under heavy load. So hopefully I can say that even with this peculiarity. Quote Link to comment Share on other sites More sharing options...
Fiery Posted January 12, 2017 Share Posted January 12, 2017 2 hours ago, delerious said: If it's a power saving trick, the GPU temperature should decrease rather than increase, no? It depends on how the video driver handles that situation. Maybe it doesn't throttle the GPU itself, but stops the video card fan (or throttles it down) to make the system work quieter. Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.