Question New GPU = Hotter CPU??

[Steven409]

So I just upgraded my GPU from a 5700 (non-XT) to a 3070FE and after installing it and all drivers, I noticed my CPU idle temps and load temps increased dramatically. While they are all still within acceptable range, the sudden increase has me concerned something else may be going on.

My CPU is a 3700X with stock cooler, mobo is a B550, and the GPU is the only thing that has changed in my system. I did have to upgrade the version of HWINFO I was using since I could not get the old one to see the 3070 and the new version had the setting for "Power Reporting Deviation" which got me doing some research. No other settings were messed with other than changing the CPU fan curve in BIOS while trying to drive the temps down.

With the 5700 installed, my CPU temps would be the following:
Idle = 36C
Load = 65-68C
Stress Test (CPU-Z) = 80C (after 60 seconds)
Multi-Tread Score = 5480 (CPU-Z Reference = 5433)
Single-Tread Score = 512 (CPU-Z Reference = 511)


With the 3070 installed, my CPU temps are the following:
Idle = 43C
Load = 78C
Stress Test (CPU-Z) = 92C (after only 10 seconds...I immediately stopped it)
Multi-Tread Score = 5321 (CPU-Z Reference = 5433)
Single-Tread Score = 468 (CPU-Z Reference = 511)

Prior to this version of HWINFO, I never had the "Power Reporting Deviation" setting, but understand it only means something when under full load. These are the number being reported:
Idle = 92.8%
Stress Test = 49.3%
CPU PPT 53.171W
CPU PPT Limit 60.4%

Something else I've noticed is that the spikes have become much more frequent and more noticeable than before. Just opening Chrome and typing this sees spikes to 69C and average spikes in the upper 50C/low 60C when it would never get above 55C during normal tasks. Now the spikes are going from 44C to 58C to 42C to 69C and they seem to do it constantly...and they're lingering at those higher temps for several seconds whereas before it would be a quick spike then back to normal.

Ambient room temp is 25-27C, which is a bit hotter than normal, but it's been hotter in the past and this was never an issue. I put a floor fan blowing directly on the case and the idle temps have dropped to an average of 38C from 43C...still doesn't help the increased spikes when using Chrome or other applications, though.



Could changing a GPU affect CPU performance this much? My 5700 was a blower style card, so perhaps the 3070FE is creating more heat in the case (this much, though)?
Does nVidia software use more CPU than AMD? I have not tried putting the 5700 back in, but I suppose I can when I have some extra time just to see what happens.

Any other thoughts of what might be causing it?
Thermal paste issues?
Failing cooler?
Faulty sensor?
Windows "Power Options" issues (I did notice a change in these with a Windows update recently...just don't remember what I had it on)?

I'm hoping it's just something simple and I'm overthinking it.


EDIT: I tried to attach a screenshot that shows HWINFO, Ryzen Master, and Cinebench R23 all running, but it would not upload...here's a link, instead. Note that Cinebench was only running for 24 seconds before these temps were reached. Before, it would be running for minutes before ever getting close to 90C. Unfortunately, I do not remember my stats from before.
https://photos.app.goo.gl/T5keW9Bs4AV2PbiYA

 

[Lutfij]

Welcome to the forums, newcomer!

Part of the heat that's being dumped by the GPU does get sucked into the stock cooler. Side note, did you use DDU to rid your platform of your prior GPU's drivers before you removed the AMD card from the system? BIOS version for your motherboard at this moment of time? FYI, B550 is the chipset off of which the board is based off of, it's not the make and/or model of the motherboard. The other factor is the make and model of your chassis and the orientation of the fans in said case.

 

[Karadjgne]

You set it free. Your old card was small enough that the cpu didn't have to work very hard to saturate the gpu ability. Game settings are not all graphical, there's a whole bunch that are most definitely cpu bound. As you move up in ability of the gpu, the detail levels also move up, so the cpu has to deal with more Ai, more physX computations, more lighting, more object collisions, just more.

More work means higher heat. Some view that as a bottleneck, it really isn't. Your old gpu didn't do DLSS or Raytracing, wouldn't do 4k DSR, wouldn't do all the lighting and cloud extras, wouldn't do all the extra details at longer viewing distances, your new one will, so that's extra work the cpu has to deal with that it didn't have to before.

And that applies to everything, even Windows. There are very few pure tests, Prime95 isnt even pure cpu, it does use ram also. Cpu-z also deals with the nvidia drivers vrs amd drivers, the extra loads it's capable of, 8bit vrs 10bit with HDR etc.

TThere's also little mention of other changes. Like the drivers. Mixing nvidia drivers and Amd drivers in the registry can create conflicts leading to higher cpu workloads, and higher heat, so overall loss of performance. Was DDU used prior to the switch?

Ryzen work best when AMD Balanced or Microsoft Balanced power plan is used. If those were changed or adjusted after a Windows update, it's pretty much guaranteed that that was a Major Windows update, which happens periodically. With those updates, you have to go back through your settings and reapply any you had prior. Stuff like disable Xbox DVR and Game Helper, even some startup items get re-initialized.

Adding more cpu workload to deal with. More heat.

 

[DRagor]

My 5700 was a blower style card, so perhaps the 3070FE is creating more heat in the case (this much, though)?

Yep, even this much. If you happen to have PCIe riser you can connect GPU through riser and put it outside case and see how much this changes CPU temps.

 

[Steven409]

Thanks for all the quick replies...I appreciate that!

Here is a bit more info about my setup:
Mobo = "ASrock B550AM Gaming" (yes, this is a CyberPowerPC that I got at a ridiculously low price at close out a few years back...could not pass it up...don't give me that look! LOL)
BIOS Mode = UEFI
BIOS Version = American Megatrends Inc. P1.10, 8/21/2019
Windows Version = 10.0.19044 Build 19044
PSU = 600watt Gold

My case has screened vents on the top and bottom, two intake fans on the front, two exhaust in the rear (including PSU).

Yes, I did use DDU when switching cards. Set it to remove AMD drivers for GPU.

As for the new GPU pushing the CPU harder, would that still be the case even in a CPU stress test in which the GPU stays at idle? What about the extremely low PRD of 47.3%? I have no idea what this setting was before since I had been using an old version of HWINFO that did not have this reading, but the way I understand it is that anything less than 85% under load is cause for concern and the lower the percentage, the harder it works which results in more heat.

I think I may just be more paranoid than I need to be. In gaming, the spikes are up to 80-82C, but they are only for a split second or two. The average seems to stay mid-70s. I've always aimed for <75C and I'm now at or above that. It's just that PRD setting and the fact it gets to +90C in less that 10 seconds.


EDIT: Here are some BIOS settings...
Voltage Mode = Stable Mode
CPU Load-Line Calibration = Level 5 / Auto
CPU VDDCR_SOC Load-Line Calibration = Level 5 / Auto
VPPM = 2.500V / Auto
2.50V Voltage = 2.500V / Auto
DRAM Voltage = 1.35V / 1.350
+1.8 Voltage = 1.800V / Auto
VDDP = 0.900V / Auto
Chipset 1.05V Voltage = 1.050V / Auto

AMD Precision Boost Overdrive = Auto (with "0" in all sub-fields such as PPT Limit, TDC Limit, etc)
ECO Mode = Disabled

Fan Curves =
40C/40%
60C/40%
70C/90%
75C/100%
Critical = 89C

 

[Steven409]

Just an update in case someone else finds their way here looking for a solution...

My temps kept getting worse so I decided to replace the CPU cooler completely. In removing the old cooler, I noticed the thermal paste was extremely poor and almost non-existent in the center of the IHS which would explain the dramatic jumps in temps I was getting. It was about time to reapply the paste anyway, but even more so considering the crappy application job it had upon being built. Figured I might as well replace the cooler with a better one while I had it off.

With the new cooler (CoolerMaster Hyper 212 Black Edition) and new paste (Arctic MX-4) applied, I'm now averaging 33C at idle in a 28C room. Max temps in gaming have yet to go over 65C and using Cinebench, it averaged only 77C after 5 minutes of running and never got above 80C (it was previously hitting 92C after just 10 seconds). This new cooler is also so, so much quieter!! Makes me wish I would have done this sooner.