• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

GPU-Z Intel Graphics Glitch

xLightningx

New Member
Joined
Mar 14, 2015
Messages
5 (0.00/day)
Ok so I got into a game in which I'm using the GeForce card. GTX 780M to be specific. But I have this strange issue with the GPU-Z program.

According to the image below, the Intel card heats up to about 95 degrees if I get into the game and then check temperature. Then it drops to about 75. Is this normal?

jLyTSSy.png


I decided to check frame rate when gaming and it appeared I was losing some frames then it all went back to normal. Any reason why is this happening?

It doesn't do that when idle. And I'm definitely using the GeForce card for playing.
 
The HD 4600 is integrated into the CPU. Check the temperature of the CPU while you're playing and report back.
 
If I relaunch GPU-Z or any other program that does checking, CPUs heat up to about 90 degrees. Then stay into the low-high 70s. What is causing that?

It just doesn't seem realistic that programs which check for my CPU or GPU temperatures make them heat up 20 degrees higher for the moment. The fan gets louder as well during that time.
 
Last edited:
When a program initially polls a chip/card/sensors, it generally causes it to go to full clocks for a short period of time. Temperatures go up during this. Make sure you have your vents unobstructed and clear of debris.
 
It just doesn't seem realistic that programs which check for my CPU or GPU temperatures make them heat up 20 degrees higher for the moment. The fan gets louder as well during that time.
The program start will briefly put your GPU in a higher performance state, but yes, I agree that it shouldn't make it go to 91°C.

What CPU and cooler do you use ? Could you try disabling your antivirus once, to see if maybe that causes it? Also check CPU load in Windows Task Manager during startup.
 
When a program initially polls a chip/card/sensors, it generally causes it to go to full clocks for a short period of time. Temperatures go up during this. Make sure you have your vents unobstructed and clear of debris.
Thermal compound could have dried out causing a bad thermal connection between the heatpipe and the CPU or it could have been an uneven application of thermal compound. Even more so if the GPU and CPU share the same cooler. If it's jumping that fast under minimal load, I would imagine replacing it could be a solution.
 
I am on a laptop obviously (MSI brand) but I did buy a cooler. I was playing Black Ops at the time and the temperature seemed ok for my Nvidia card. It's just that the CPU pulled some crazy heat. I read for my 4700MQ CPU thermal junction is 100 Celsius.
 
I am on a laptop obviously (MSI brand) but I did buy a cooler. I was playing Black Ops at the time and the temperature seemed ok for my Nvidia card. It's just that the CPU pulled some crazy heat. I read for my 4700MQ CPU thermal junction is 100 Celsius.
It's probably starting to throttle. If the laptop is the the MSI GT70, the CPU and GPU have different heat sinks but use the same fan. I highly recommend replacing the thermal compound on the CPU side of the cooler as it sounds like it's not making very good thermal contact. The under-side of the laptop must be getting pretty toasty where the CPU is.
 
do you have a max fan speed button like on my MSI laptop? is it about the same cpu temp if you put only cpu load (like cinebench) on it? are you capping your framerate to refresh?
 
 
Last edited:
Yes, I do have a GT70, it's the Dragon Edition one. And I do cap frame rate at 60, I don't need more frames than that. The laptop is still under warranty, I'm not really sure if it needs any fixes applied. I've had it for over a year with no real problems other than the blue screen error with LAN drivers but I've fixed that one.
 
One year old, means overheating is a possible issue due to decay in cooling performance for various reason, paste losing its efectiveness, fan/radiator being clogged up. If you don't want to mess with it during warranty period just use throttlestop and disable turbo boost on your cpu. You will loose performance but there wont be throttling to lower than stock speed which still should be enough in most cases. When not gaming you can reenable turbo boost. I did this thing on my Dell for 3 years, until i finally repasted it and temps dropped 10* C and no more thermal shutdown or anything like that.
 
just chiming in I redid the thermal paste on a customers MSI laptop last year and the plastic strip the paste comes on was still on the cooler. If thats any indication of their QC. That said with such performance in a small space it wouldnt surprise me if the paste was dry either. Its not like any manufacturer does a great job at pre-application anyway.
 
If you don't want to mess with it during warranty period just use throttlestop and disable turbo boost on your cpu.
I heard I can also do that in Windows itself by making a power settings profile with maximum performance of 99% for the CPU. It says 'Maximum processor state'.
 
I heard I can also do that in Windows itself by making a power settings profile with maximum performance of 99% for the CPU. It says 'Maximum processor state'.

I also heard that, I wanted to disable turbo boost but this did not work for me for some reason.
I have installed Intel Extreme Tuning Utility now and have set the multiplier manually for all cores, this settings will stay this way if you turn off/on your system.
The settings are gone if you must "Restart" your system for lets say windows updates.
 
Back
Top