[SOLVED] R7 3800x vs R5 2600 temps

Page 2 - Seeking answers? Join the Tom's Hardware community: where nearly two million members share solutions and discuss the latest tech.
So today I got my 3800x as an upgrade from the 2600.

My temps are much higher idle and under load. Is this normal for the 3800x compared to the 2600? My 2600 idled at 38c-48c (it would spike randomly which is normal behavior for ryzen). The 3800x idles at 45c-70c. Yes the same type of "spike" behavior even idle but when it spikes sometimes it has reached up to 70c, but mostly it just spikes to around 55c-61c, however I seen it reach 70c idle.

I current have the clock at 4.4ghz which is .1ghz lower than its max boost speed, and voltage set to 1.42 (It was reaching 1.47 at 4.3ghz on auto mode under load).

My memory is on XMP 3000mhz cl15.

My cooler isnt the greatest, its the Cooler Master Hyper T2, but Ive heard its within a few degrees on performance compared to the 212 evo.

So is this normal or do these temps seem high? I even had to adjust my fan curve so it didnt keep ramping on idle.

MOBO: MSI Tomahawk B450
CPU: Ryzen 7 3800x (4.4ghz 1.42v)
Cooler: Cooler Master Hyper T2
GPU: Gigabyte RTX 2060 Super OC 3x Windforce White (8pin only version)
MEMORY: HyperX RGB 2x8gb DDR4 3000mhz
Storage: Samsung 860 Evo 250gb / 500gb Seagate / 1tb Samsung 860 Qvo
PSU: Corsair cx550m
WIN: Windows 10 Pro
Case Cooling: Antec Prizm 3x 120mm rgb case fans with controller and 2 RGB strips / 1 default 120mm case fan
Case: Cougar MX330
 
Solution
Not really. That is pretty much a Prism only thing. There are coolers with RGB fans, but honestly, they suck. Same as with the conversation we have daily around here regarding case fans, which are the same as fans on heatsinks, you can get good fans or RGB fans. There are really no good RGB fans, not for use on radiators or heatsinks anyhow. They all either lack performance, or lack static pressure or lack airflow. Seems to be getting slowly better but for the most part it's hard for them to include a good fan motor AND the electronics required to make the RGB factor work, on the same fan. Just not enough room I guess.

As far as the coolers go, the Prism is probably definitely the better cooler than your T2. The T2 is a two heatpipe...
Not really. That is pretty much a Prism only thing. There are coolers with RGB fans, but honestly, they suck. Same as with the conversation we have daily around here regarding case fans, which are the same as fans on heatsinks, you can get good fans or RGB fans. There are really no good RGB fans, not for use on radiators or heatsinks anyhow. They all either lack performance, or lack static pressure or lack airflow. Seems to be getting slowly better but for the most part it's hard for them to include a good fan motor AND the electronics required to make the RGB factor work, on the same fan. Just not enough room I guess.

As far as the coolers go, the Prism is probably definitely the better cooler than your T2. The T2 is a two heatpipe heatsink, which is extremely poor for an aftermarket cooler.

The Prism is a four heat pipe cooler, sort of. It really is a modified or "split" two heatpipe cooler but technically it does have four heat pipes so there IS more overall surface area and coverage than there is with a standard two pipe version.

If you don't like the Ninja, because of it's size, then the next best option I'd recommend would be this:

https://pcpartpicker.com/product/8GBrxr/scythe-mugen-5-rev-b-5117-cfm-cpu-cooler-scmg-5100


The smaller the cooler, the (Obviously) worse the performance is going to be, to some degree at least. There are definitely many other coolers, some better, some cheaper, that would be good options too but unfortunately the height limitation of your case eliminates the majority of them because most good coolers are more than 155mm tall.
 
Solution
Not really. That is pretty much a Prism only thing. There are coolers with RGB fans, but honestly, they suck. Same as with the conversation we have daily around here regarding case fans, which are the same as fans on heatsinks, you can get good fans or RGB fans. There are really no good RGB fans, not for use on radiators or heatsinks anyhow. They all either lack performance, or lack static pressure or lack airflow. Seems to be getting slowly better but for the most part it's hard for them to include a good fan motor AND the electronics required to make the RGB factor work, on the same fan. Just not enough room I guess.

As far as the coolers go, the Prism is probably definitely the better cooler than your T2. The T2 is a two heatpipe heatsink, which is extremely poor for an aftermarket cooler.

The Prism is a four heat pipe cooler, sort of. It really is a modified or "split" two heatpipe cooler but technically it does have four heat pipes so there IS more overall surface area and coverage than there is with a standard two pipe version.

If you don't like the Ninja, because of it's size, then the next best option I'd recommend would be this:

https://pcpartpicker.com/product/8GBrxr/scythe-mugen-5-rev-b-5117-cfm-cpu-cooler-scmg-5100


The smaller the cooler, the (Obviously) worse the performance is going to be, to some degree at least. There are definitely many other coolers, some better, some cheaper, that would be good options too but unfortunately the height limitation of your case eliminates the majority of them because most good coolers are more than 155mm tall.

Thats unfortunate because the prism is pretty lol.

BUT ty for all the help man! I will probably get the previous one you posted.

Last question, will I have to take my mobo out and have to install a new backplate with the new cooler?:
 
You will have to install a new backplate, but you SHOULD NOT have to take the motherboard out so long as the case in use has a cutout on the motherboard tray to access the back of the motherboard, which that MX330 (And most case models sold from the last ten to fifteen years) does.
 
  • Like
Reactions: Jason H.
Cool man. Glad to help.

Hey dark, so I got the Mugen 5 Rev B in today and got it installed (the directions said to use the backplate already on my mobo for AM4 as the backplate they provide is for intel only).

So while doing the same burn test from furmark I was peaking at about 78c. But this is weird as I did the same test last night on the Prism and was maxing about 84c, so not a big differance.

Also I tried a new test, Prime95 and used the Small FFT, 16 threads option and the temps were at 102c within 1 minute so I stopped the test as the temp was climbing pretty steadily.

Is this normal or did I run the wrong test on Prime? Does 102c on this cooler with prime small fft with 16 threads sound too high because I thought I would be able to have MUCH better cooling with this cooler.

IMPORTANT
I have my clock at 4.3ghz at 1.375v, so IDK if the temps because of the OC or what but its just an average oc for this chip. Also, at idle, the voltage stays at 1.37 but under load drops all the way to 1.33, not sure how to load line calibrate but worth noting.

Any advice would be great, thanks!
 
Something is not right. No way that cooler gets less than ten degrees better than the stock cooler.

Yes, apparently the stock backplate is used but you will need to use different brackets AND there are AM4 specific standoffs that have to be used. Did you make sure to use the AMD/AM4 standoffs and not the ones for Intel? They are likely different heights. That could make a big difference in cooler performance. I've seen the wrong standoffs get used on coolers before and cause similar issues.

This video shows both Intel and AMD/AM4 mounting. You can skip forward to about 7:50 in the video to see the AM4 mounting. He briefly discusses that the standoffs are different. He's kind of a dweeb, but it offers a decent insight to the AM4 mounting on this cooler. I'm sure there are others as well.

View: https://www.youtube.com/watch?v=qZoDaZmjcbU


Otherwise, something else must be going on. Maybe you failed to remove the protective plastic covering the bottom of the heatsink? Too much paste? Too little paste?

Standoffs not fully seated into the backplate, brackets not fully seated into the standoffs or heatsink not fully tightened down onto the brackets?

As far as Prime95 is concerned, you MUST disable AVX and AVX2, in the popup options, when you start Prime95. Be sure to select the "Small FFT" option. Not "Smallest FFT". Disabling the AVX and AVX2 options are imperative though.
 
Something is not right. No way that cooler gets less than ten degrees better than the stock cooler.

Yes, apparently the stock backplate is used but you will need to use different brackets AND there are AM4 specific standoffs that have to be used. Did you make sure to use the AMD/AM4 standoffs and not the ones for Intel? They are likely different heights. That could make a big difference in cooler performance. I've seen the wrong standoffs get used on coolers before and cause similar issues.

This video shows both Intel and AMD/AM4 mounting. You can skip forward to about 7:50 in the video to see the AM4 mounting. He briefly discusses that the standoffs are different. He's kind of a dweeb, but it offers a decent insight to the AM4 mounting on this cooler. I'm sure there are others as well.

View: https://www.youtube.com/watch?v=qZoDaZmjcbU


Otherwise, something else must be going on. Maybe you failed to remove the protective plastic covering the bottom of the heatsink? Too much paste? Too little paste?

Standoffs not fully seated into the backplate, brackets not fully seated into the standoffs or heatsink not fully tightened down onto the brackets?

As far as Prime95 is concerned, you MUST disable AVX and AVX2, in the popup options, when you start Prime95. Be sure to select the "Small FFT" option. Not "Smallest FFT". Disabling the AVX and AVX2 options are imperative though.

Yes I made sure. I used the standoffs in the same bag the AMD brackets came in, which includes the 3m sticky square thing but even in his video he didnt use it, and I had no use for it either.

I installed it properly followed the instructions and had watched a video prior. All connections/screws are secured tightly to backplate/brackets/heatsink to brackets. Paste is perfect amount.

I mean its POSSIBLE when I had ran the Prism test that it got to 88c and Im seeing 78c now however I even disabled my manual OC of 4.3ghz at 1.375v, and turned PBO back to AUTO as well and it ran at 4.15ghz at 1.3v and the temps only dropped by MAYBE 1c.

I mean its not a surprise as HWMonitor shows its only using 5w less, and 0.03v less than my manual OC.

But the voltages at idle with stock were reaching up to 1.495v on single cores and multiple cores and the clocks was boosting up to 4.55ghz. Which was odd.

But honestly all in all, does the 78-80c range sound right for this cooler on this cpu? I will try to do that thing you said for Prime and see if it makes a differance as well and will report back temp.
 
Something is not right. No way that cooler gets less than ten degrees better than the stock cooler.

Yes, apparently the stock backplate is used but you will need to use different brackets AND there are AM4 specific standoffs that have to be used. Did you make sure to use the AMD/AM4 standoffs and not the ones for Intel? They are likely different heights. That could make a big difference in cooler performance. I've seen the wrong standoffs get used on coolers before and cause similar issues.

This video shows both Intel and AMD/AM4 mounting. You can skip forward to about 7:50 in the video to see the AM4 mounting. He briefly discusses that the standoffs are different. He's kind of a dweeb, but it offers a decent insight to the AM4 mounting on this cooler. I'm sure there are others as well.

View: https://www.youtube.com/watch?v=qZoDaZmjcbU


Otherwise, something else must be going on. Maybe you failed to remove the protective plastic covering the bottom of the heatsink? Too much paste? Too little paste?

Standoffs not fully seated into the backplate, brackets not fully seated into the standoffs or heatsink not fully tightened down onto the brackets?

As far as Prime95 is concerned, you MUST disable AVX and AVX2, in the popup options, when you start Prime95. Be sure to select the "Small FFT" option. Not "Smallest FFT". Disabling the AVX and AVX2 options are imperative though.

OK so with Prime95 small fft disabled avx and avx2, with stock 3800x settings, it was leveling off at about 83c after 10 minutes. Could have reached 85-87c after another 30-45 minutes. This was at 4.15ghz with 1.495v at idle, and 1.1-1.3v during the test, with PBO enabled.

The same test with the manual OC of 4.3ghz 1.375v at idle, and 1.308-1.31v during the test, PBO disabled, the temp was was 92c after 10 minutes, could have reached 95c within 20-30 minutes.

Is this good for the cooler? And whats up with the idle voltage at stock settings?

Also The results of the Cinebench with the manual oc was a Multi core score of 5003cb (83c) and single core of 492cb (56c).
 
Last edited:
You're using HWmonitor? Yeah, don't do that. Use Ryzen master or HWinfo. Or even Core Temp. HWmonitor is garbage in my opinion. Way too many instances where I've seen inaccuracies or outright failure to read the right sensor data for a given sensor or to even READ a given sensor at all, while other utilities read and report the same sensor or chipset data perfectly fine.

Here is my spiel on that, which has become a commonplace posting here for me.


Monitoring software

HWmonitor, Open hardware monitor, Realtemp, Speccy, Speedfan, Windows utilities, CPU-Z, NZXT CAM and most of the bundled motherboard utilities are often not the best choice as they are not always accurate. Some are actually grossly inaccurate, especially with certain chipsets or specific sensors that for whatever reason they tend to not like or work well with. I've found HWinfo or CoreTemp to be the MOST accurate with the broadest range of chipsets and sensors. They are also almost religiously kept up to date.

CoreTemp is great for just CPU thermals including core temps or distance to TJmax on older AMD platforms.

HWinfo is great for pretty much EVERYTHING, including CPU thermals, core loads, core temps, package temps, GPU sensors, HDD and SSD sensors, motherboard chipset and VRM sensor, all of it. When starting HWinfo after installation, always check the box next to "sensors only" and de-select the box next to "summary".


Run HWinfo and look at system voltages and other sensor readings.

Monitoring temperatures, core speeds, voltages, clock ratios and other reported sensor data can often help to pick out an issue right off the bat. HWinfo is a good way to get that data and in my experience tends to be more accurate than some of the other utilities available. CPU-Z, GPU-Z and Core Temp all have their uses but HWinfo tends to have it all laid out in a more convenient fashion so you can usually see what one sensor is reporting while looking at another instead of having to flip through various tabs that have specific groupings, plus, it is extremely rare for HWinfo to not report the correct sensor values under the correct sensor listings, or misreport other information. Utilities like HWmonitor, Openhardware monitor and Speccy, tend to COMMONLY misreport sensor data, or not report it at all.

After installation, run the utility and when asked, choose "sensors only". The other window options have some use but in most cases everything you need will be located in the sensors window. If you're taking screenshots to post for troubleshooting, it will most likely require taking three screenshots and scrolling down the sensors window between screenshots in order to capture them all.

It is most helpful if you can take a series of HWinfo screenshots at idle, after a cold boot to the desktop. Open HWinfo and wait for all of the Windows startup processes to complete. Usually about four or five minutes should be plenty. Take screenshots of all the HWinfo sensors.

Next, run something demanding like Prime95 (With AVX and AVX2 disabled) or Heaven benchmark. Take another set of screenshots while either of those is running so we can see what the hardware is doing while under a load.


*Download HWinfo



For temperature monitoring only, I feel Core Temp is the most accurate and also offers a quick visual reference for core speed, load and CPU voltage:


*Download Core Temp




Ryzen master for Zen or newer AMD CPUs, or Overdrive for older Pre-Ryzen platforms (AM3/AM3+/FM2/FM2+)

For monitoring on AMD Ryzen and Threadripper platforms including Zen or newer architectures, it is recommended that you use Ryzen master if for no other reason than because any updates or changes to monitoring requirements are more likely to be implemented sooner, and properly, than with other monitoring utilities. Core Temp and HWinfo are still good, with this platform, but when changes to CPU micro code or other BIOS modifications occur, or there are driver or power plan changes, it sometimes takes a while before those get implemented by 3rd party utilities, while Ryzen master, being a direct AMD product, generally gets updated immediately. Since it is also specific to the hardware in question, it can be more accurately and specifically developed without any requirement for inclusion of other architectures which won't be compatible in any case. You wouldn't use a hammer to drive a wood screw in (At least I hope not) and this is very much the same, being the right tool for the job at hand.

As far as the older AMD FX AM3+ platforms including Bulldozer and Piledriver families go, there are only two real options here. You can use Core Temp, but you will need to click on the Options menu, click Settings, click Advanced and put a check mark next to the setting that says "Show Distance to TJmax in temperature fields" and then save settings and exit the options menu system. This may or may not work for every FX platform, so using AMD Overdrive is the specific, again, right tool for the job, and recommended monitoring solution for this architecture. Since these FX platforms use "Thermal margins" rather than an actual "core/package" temp type thermal monitoring implementation, monitoring as you would with older or newer AMD platforms, or any Intel platform, won't work properly.

For more information about this, please visit here for an in depth explanation of AMD thermal margin monitoring.

Understanding AMD thermal margins for Pre-Ryzen processors





*Download Ryzen Master




*Download AMD Overdrive



Also, posting screenshots, when requested, is helpful so WE can see what is going on as well and you can learn how to do that here:

How to post images on Tom's hardware forums




And for what it's worth, I'd start out with PBO "DISABLED" and leave the PB enabled. Precision boost overdrive is not the same as precision boost. Precision boost is the default boost behavior while Precision boost overdrive is an automatic overclocking mechanism. It has been recommended to me by users with much more experience than myself on Ryzen to do the following.

Disable PBO but leave Precision boost enabled, in the BIOS.

Enable Cool N' Quiet
Enable Core CPPC
Enable CPPC preferred cores
Enable advanced/global C-states

All of which are BIOS settings.

Using HWinfo (Sensors only option. Uncheck the option for "Summary") or Ryzen master, run your tests again. Some utilities, including some versions of HWmonitor, report an inaccurate ten degree offset on Ryzen chipsets. I doubt that is the case at this late date but anything is possible and like I said, I don't much trust the accuracy of HWmonitor anyhow.
 
You're using HWmonitor? Yeah, don't do that. Use Ryzen master or HWinfo. Or even Core Temp. HWmonitor is garbage in my opinion. Way too many instances where I've seen inaccuracies or outright failure to read the right sensor data for a given sensor or to even READ a given sensor at all, while other utilities read and report the same sensor or chipset data perfectly fine.

Here is my spiel on that, which has become a commonplace posting here for me.


Monitoring software

HWmonitor, Open hardware monitor, Realtemp, Speccy, Speedfan, Windows utilities, CPU-Z, NZXT CAM and most of the bundled motherboard utilities are often not the best choice as they are not always accurate. Some are actually grossly inaccurate, especially with certain chipsets or specific sensors that for whatever reason they tend to not like or work well with. I've found HWinfo or CoreTemp to be the MOST accurate with the broadest range of chipsets and sensors. They are also almost religiously kept up to date.

CoreTemp is great for just CPU thermals including core temps or distance to TJmax on older AMD platforms.

HWinfo is great for pretty much EVERYTHING, including CPU thermals, core loads, core temps, package temps, GPU sensors, HDD and SSD sensors, motherboard chipset and VRM sensor, all of it. When starting HWinfo after installation, always check the box next to "sensors only" and de-select the box next to "summary".


Run HWinfo and look at system voltages and other sensor readings.

Monitoring temperatures, core speeds, voltages, clock ratios and other reported sensor data can often help to pick out an issue right off the bat. HWinfo is a good way to get that data and in my experience tends to be more accurate than some of the other utilities available. CPU-Z, GPU-Z and Core Temp all have their uses but HWinfo tends to have it all laid out in a more convenient fashion so you can usually see what one sensor is reporting while looking at another instead of having to flip through various tabs that have specific groupings, plus, it is extremely rare for HWinfo to not report the correct sensor values under the correct sensor listings, or misreport other information. Utilities like HWmonitor, Openhardware monitor and Speccy, tend to COMMONLY misreport sensor data, or not report it at all.

After installation, run the utility and when asked, choose "sensors only". The other window options have some use but in most cases everything you need will be located in the sensors window. If you're taking screenshots to post for troubleshooting, it will most likely require taking three screenshots and scrolling down the sensors window between screenshots in order to capture them all.

It is most helpful if you can take a series of HWinfo screenshots at idle, after a cold boot to the desktop. Open HWinfo and wait for all of the Windows startup processes to complete. Usually about four or five minutes should be plenty. Take screenshots of all the HWinfo sensors.

Next, run something demanding like Prime95 (With AVX and AVX2 disabled) or Heaven benchmark. Take another set of screenshots while either of those is running so we can see what the hardware is doing while under a load.


*Download HWinfo



For temperature monitoring only, I feel Core Temp is the most accurate and also offers a quick visual reference for core speed, load and CPU voltage:


*Download Core Temp




Ryzen master for Zen or newer AMD CPUs, or Overdrive for older Pre-Ryzen platforms (AM3/AM3+/FM2/FM2+)

For monitoring on AMD Ryzen and Threadripper platforms including Zen or newer architectures, it is recommended that you use Ryzen master if for no other reason than because any updates or changes to monitoring requirements are more likely to be implemented sooner, and properly, than with other monitoring utilities. Core Temp and HWinfo are still good, with this platform, but when changes to CPU micro code or other BIOS modifications occur, or there are driver or power plan changes, it sometimes takes a while before those get implemented by 3rd party utilities, while Ryzen master, being a direct AMD product, generally gets updated immediately. Since it is also specific to the hardware in question, it can be more accurately and specifically developed without any requirement for inclusion of other architectures which won't be compatible in any case. You wouldn't use a hammer to drive a wood screw in (At least I hope not) and this is very much the same, being the right tool for the job at hand.

As far as the older AMD FX AM3+ platforms including Bulldozer and Piledriver families go, there are only two real options here. You can use Core Temp, but you will need to click on the Options menu, click Settings, click Advanced and put a check mark next to the setting that says "Show Distance to TJmax in temperature fields" and then save settings and exit the options menu system. This may or may not work for every FX platform, so using AMD Overdrive is the specific, again, right tool for the job, and recommended monitoring solution for this architecture. Since these FX platforms use "Thermal margins" rather than an actual "core/package" temp type thermal monitoring implementation, monitoring as you would with older or newer AMD platforms, or any Intel platform, won't work properly.

For more information about this, please visit here for an in depth explanation of AMD thermal margin monitoring.

Understanding AMD thermal margins for Pre-Ryzen processors





*Download Ryzen Master




*Download AMD Overdrive



Also, posting screenshots, when requested, is helpful so WE can see what is going on as well and you can learn how to do that here:

How to post images on Tom's hardware forums




And for what it's worth, I'd start out with PBO "DISABLED" and leave the PB enabled. Precision boost overdrive is not the same as precision boost. Precision boost is the default boost behavior while Precision boost overdrive is an automatic overclocking mechanism. It has been recommended to me by users with much more experience than myself on Ryzen to do the following.

Disable PBO but leave Precision boost enabled, in the BIOS.

Enable Cool N' Quiet
Enable Core CPPC
Enable CPPC preferred cores
Enable advanced/global C-states

All of which are BIOS settings.

Using HWinfo (Sensors only option. Uncheck the option for "Summary") or Ryzen master, run your tests again. Some utilities, including some versions of HWmonitor, report an inaccurate ten degree offset on Ryzen chipsets. I doubt that is the case at this late date but anything is possible and like I said, I don't much trust the accuracy of HWmonitor anyhow.

RM and HWInfo both show the same temps as HWMonitor in my testing. However when viewing WATTS RM shows lower wattage being used than hwmonitor and hwinfo. Both hw programs show 120w usage while RM shows a 98w usage, which doesnt make sense because with the 4.3ghz manual oc it should be at over 105w in RM, meaning RM SHOULD be reporting the same wattage as HWInfo or HWMonitor.

Also before I do all that, which will have to be when I get off work, I did some more things last nite and I was able to get the voltage down to 1.325v at 4.3ghz at idle, and 1.26v under load.

The new temps are 86c with prime95 77c with furmark cpu burner and 81c with cinebench.

Do those temps seem more acceptable?

Im just wondering why this cooler, without lowering the voltage, only dropped temps by like 5-7c. I figured it would be much better than the prism.

So with the lowered viltage and lower temps, Im assuming the prism wouldnt have been so bad to begin with as while having the same 1.37v 4.3ghz oc on both coolers only showwed a 5-7c decrease.

Nevertheless, now Im seeing the 10-11c drop with the lowered voltage that I thought I would see without lowering the voltage.
 
Anything over 80°C is unacceptable, if you want to see consistent and full boost profiles, according to every review I've seen.

Of course, you are overclocking, which is completely throwing everything off. You need to NOT overclock and you need to DISABLE PBO leaving only Precision boost enabled, and THEN do your thermal testing to determine a baseline from which to operate. Because if you do not have a very acceptable, MUCH lower temperature, with the stock configuration WITHOUT PBO enabled, then there IS a problem somewhere. No way to tell that when you are trying to run PBO or manually overclock. Those things should come later after a baseline with the new cooler has been fully established.

Also, I think you are seriously toeing the line in terms of an all core manual overclock. Most reviews of overclocking on the 3700x and 3800x show 4.3Ghz as the MAXIMUM, not just some mediocre overclock, and in most cases not achievable with the majority of hardware. When it comes time to overclock I would advise that you start at like 4.1Ghz and begin verifying thermal compliance from there, working your way up while also testing stability along the way, for each step. So, 4.1Ghz all core, test thermal compliance, then test stability.

For stability I'd recommend using Realbench on the stress test option. 1 hour minimum between steps. 8 hours for final testing once you are satisfied you've taken the overclock as far as the silicon and hardware are going to allow.

Prime95 thermal compliance testing should not take more than 15 minutes. After 15 minutes you are unlikely to see any significant increase in thermals no matter how long you run it, assuming you didn't start in the winter and run it until summer when the ambient is 40 degrees different. LOL.
 
  • Like
Reactions: Jason H.
Anything over 80°C is unacceptable, if you want to see consistent and full boost profiles, according to every review I've seen.

Of course, you are overclocking, which is completely throwing everything off. You need to NOT overclock and you need to DISABLE PBO leaving only Precision boost enabled, and THEN do your thermal testing to determine a baseline from which to operate. Because if you do not have a very acceptable, MUCH lower temperature, with the stock configuration WITHOUT PBO enabled, then there IS a problem somewhere. No way to tell that when you are trying to run PBO or manually overclock. Those things should come later after a baseline with the new cooler has been fully established.

Also, I think you are seriously toeing the line in terms of an all core manual overclock. Most reviews of overclocking on the 3700x and 3800x show 4.3Ghz as the MAXIMUM, not just some mediocre overclock, and in most cases not achievable with the majority of hardware. When it comes time to overclock I would advise that you start at like 4.1Ghz and begin verifying thermal compliance from there, working your way up while also testing stability along the way, for each step. So, 4.1Ghz all core, test thermal compliance, then test stability.

For stability I'd recommend using Realbench on the stress test option. 1 hour minimum between steps. 8 hours for final testing once you are satisfied you've taken the overclock as far as the silicon and hardware are going to allow.

Prime95 thermal compliance testing should not take more than 15 minutes. After 15 minutes you are unlikely to see any significant increase in thermals no matter how long you run it, assuming you didn't start in the winter and run it until summer when the ambient is 40 degrees different. LOL.

Ok got off work and did everything you said. Disabled PBO, enabled cool n quiet, enabled (all the other stuff you said) and left clocks and voltages at auto.

Results
  • Furmark 15 minute CPU burn test hit 76c at 4.15 ghz with 1.29-1.3v
  • Cinenbench (2 back to back runs) hit 80c with a 4814cb
  • Prime95 15 minute Small FFT hit 82c, but only at 4.07ghz at 1.31-1.33v
So now we have the baseline. Any good? Seems its only a 1c drop on Furmark, a 2c drop on Cinebench, and a 4c drop on Prime.

Also, the voltage is at 1.489v at idle now.

EDIT: Also something worth noting is that in HWInfo there are 2 different cpu temps, 1 on the chip itself, CPU CCD1, then a sensor on my mobo just called CPU. The one on my mobo is 5-10c higher than the one in Ryzen Master, and the CCD1 temp, it also jumps around more sparatically than the other 2 sensors.

Also while gaming at stock settings the clocks boost to 4.275-4.3ghz anyways so the OC doesnt make too much sense I guess in those regards but Id still like to have an all core OC as I feel Ive lost about 10-15fps. And the temps are max 75c but mostly stay at 60-68c. The 75c was only once.
 
Last edited:
So now I think we need to determine whether or not this is a problem caused by your fan profile for the CPU cooler in the BIOS?

What and how is that set? Have you done anything with that at all?

It should be set up with a custom curve, I'd like to see it at full speed around the time the CPU hits maybe 70-75°C, with whatever sort of curve you prefer at lower temps. Also, make sure the CPU_FAN header is set to CPU as the source and that it is set to PWM rather than DC.

There has to be some reason you are not getting better temps. Are you testing this with the side panel on or off? Try it with it off and see if things improve.
 
So now I think we need to determine whether or not this is a problem caused by your fan profile for the CPU cooler in the BIOS?

What and how is that set? Have you done anything with that at all?

It should be set up with a custom curve, I'd like to see it at full speed around the time the CPU hits maybe 70-75°C, with whatever sort of curve you prefer at lower temps. Also, make sure the CPU_FAN header is set to CPU as the source and that it is set to PWM rather than DC.

There has to be some reason you are not getting better temps. Are you testing this with the side panel on or off? Try it with it off and see if things improve.

Custom curve is something like 40c/60% 55c/70% 65c/80% 75c/100% for my CPU cooler. And for my case fans, to make it simple, they hit 85% at 75c. They are loud so I dont like them going to 100% I have 1 fan that is 100% and thats the fan that came with the case. Im using it on the top as a secondary exhaust. Its not loud at max and has low rps. Crap fan but figured why not have it up top. so 4 case fans total. 2 intake, 2 exhaust. All 120mm.

Furmark CPU burner with side panel off is 74.5c which is 1.5c lower than with side panel on, but tbh thats not a surprise in my eyes.

Also, I uninstalled the cooler fully and reinstalled and even put a tad bit more paste, same temps.

EDIT: I just found a video of a guy using the same cooler on a R7 1700x and his max temp using the CPU-Z stress test was 76c. My temp on the 3800x using the same exact test was 80c, which is expected seeing as its a better CPU with higher TDP Im assuming. So would it be sage to assume this is just normal for this cooler? Heres his video
View: https://www.youtube.com/watch?v=0oKZ5yVo76Q
 
Last edited:
If the stock cooler is able to keep the CPU under 80°C, which it IS supposed to be able to do, even at 100% usage in the stock configuration, then any good aftermarket cooler like this one should be able to do that and it seems as though it's not able to. I suspect that either there is a misconfiguration of fan orientations somewhere or you got a rotten CPU sample.

Did you try running those tests with the side panel off? That is the only way to eliminate poor case cooling as the culprit. If your temperatures are even moderately improved with the side panel off then there is an issue with the case cooling configuration or performance.

Out of curiosity, have you made ANY changes to the load line calibration settings at all? Have you even looked at making changes there or has that been left to Auto throughout this process?
 
  • Like
Reactions: M3rKn

M3rKn

Respectable
Nov 13, 2019
315
70
1,890
Well the Prism didnt come with standoffs or brackets, I was using whatever was already installed on the MOBO and never had an issue until this new CPU so Im fairly confident that the brackets are fine. Also my T2 didnt come with AMD brackets, it only came with Intel brackets so the entire time I was just using what was already on the MOBO. Also I bought the cooler about 1 and a half years ago so AM4 was already out.

HOWEVER I will say, its much harder getting the Prism on as the mechanism on the cooler that latches onto the bracket doesnt seem to line up very well, like the brackets are just a millimeter or so too low, but never the less I did get it installed and it cools a lot better than the T2 and seems to apply a lot more pressure. Im thinking the fan on my T2 is dying, it doesnt make noise but it still is pretty much on par with the Prism and has such higher temps which leads me to believe something is wrong with the cooler itself as it should at LEAST keep temps around the same temps as the Prism.

Also when removing the Prism cooler, it was so hard to pull off that it pulled the CPU right out of the socket. I almost had a heart attack (1 day old paste). But there was 0 damage to the CPU or the socket which Im thankful for but now Im scared to pull the damn thing off again lol. Any advice on how to pull it off without pulling out the cpu?

The temps are not too high, and as you are aware these temps are only reached for brief moments at at time. If the temp spikes bother you there is a simple solution. I am assuming you are using the AMD Ryzen Balanced power plan. Due to the dynamic nature of Ryzen, the spikes are not going away, its how the cpu is designed. But you can change the power plan. I have found this is more effective than trying to under-clock the CPU. When I am gaming I use Ryzen Balanced. When I am doing anything other than gaming I use Windows (balanced) power plan and I set the Power mode: Best energy savings. You should have lower temps, and more stable boosting. The CPU will still boost when it needs to, but it isn't going to boost every time a you click your mouse or some random process in the background begins.

The Prism is easier to install with the mobo out of the case in my experience, but it can be installed just fine inside the case with patience. FYI give the cooler a twist when you are removing it, because even if the paste is only a day old it can tend to stick to your CPU enough to pull it out. This won't damage your CPU, but if it were to fall back down into your case afterwards a pin could get bent or broken in the process.
 

M3rKn

Respectable
Nov 13, 2019
315
70
1,890
1IDBoa6.png

This is Windows Balanced: Best energy savings

It basically makes the CPU run at base clock speeds and below. I have zero issues running multiple tabs and programs with this setting. The system is virtually silent.
If I switch to power saver mode the system is dead silent and I can still stream youtube or netflix without any issues.

LaQrtUj.png

This is AMD Ryzen Balanced

As you can see temps are higher, and will fluctuate between 45-55c every few seconds, it will reach 70c sometimes, but that's because it is constantly trying to boost to max performance. I only reserve this for gaming. My system will boost and hit 70c no problem, but once in game I will have steady temps of about 55-60c and they system no longer fluctuates. When I am done gaming I switch the power settings.

give the stock cooler a chance before your right if off
 
If the stock cooler is able to keep the CPU under 80°C, which it IS supposed to be able to do, even at 100% usage in the stock configuration, then any good aftermarket cooler like this one should be able to do that and it seems as though it's not able to. I suspect that either there is a misconfiguration of fan orientations somewhere or you got a rotten CPU sample.

Did you try running those tests with the side panel off? That is the only way to eliminate poor case cooling as the culprit. If your temperatures are even moderately improved with the side panel off then there is an issue with the case cooling configuration or performance.

Out of curiosity, have you made ANY changes to the load line calibration settings at all? Have you even looked at making changes there or has that been left to Auto throughout this process?

I stated in my previous reply, I tested with the side panel off and there was a 1.5c decrease. Which is normal in my opinion. The internal temp of my PC is 30-32c. All my fans are in the correct orientation.

The stock cooler doesnt keep the temps below 80 with prime95.

I watched MANY videos lastnite and although I did see some people maxing out at 80c with the stock cooler, most were maxing at abt 85 which is on par with my stock cooler.

And the video I linked also shows the Mugen 5 on the 1700x, and his max was 75c, and on my 3800x its 82c, which is normal in my opinion.

And I have not touched LLC, but figured with stock settings I wouldnt need to.
 
1IDBoa6.png

This is Windows Balanced: Best energy savings

It basically makes the CPU run at base clock speeds and below. I have zero issues running multiple tabs and programs with this setting. The system is virtually silent.
If I switch to power saver mode the system is dead silent and I can still stream youtube or netflix without any issues.

LaQrtUj.png

This is AMD Ryzen Balanced

As you can see temps are higher, and will fluctuate between 45-55c every few seconds, it will reach 70c sometimes, but that's because it is constantly trying to boost to max performance. I only reserve this for gaming. My system will boost and hit 70c no problem, but once in game I will have steady temps of about 55-60c and they system no longer fluctuates. When I am done gaming I switch the power settings.

give the stock cooler a chance before your right if off

My windows is up to date, there is no AMD Ryzen Balanced plan. Also yes, while gaming my temps are about 55-65c at stock, PBO disabled, but with the Sycthe Mugen 5, not stock cooler. (Im downloading the chipset from AMDs website now for the Power Plan)

but also, thats not really my main concern. My main concern is if these are normal temps, with this cooler, on the 3800x. The Sycthe Mugen 5 rev.b
 
Last edited:
there is no AMD Ryzen Balanced plan

Did you manually install the most recent AMD chipset drivers, downloaded directly from the AMD website, or are you running the natively supplied Microsoft chipset drivers?

If you have the latest AMD B450 chipset drivers installed then you SHOULD see the Ryzen balanced power plan in the control panel power options applet. That alone, if you don't might account for some of the power discrepancies itself.
 
Did you manually install the most recent AMD chipset drivers, downloaded directly from the AMD website, or are you running the natively supplied Microsoft chipset drivers?

If you have the latest AMD B450 chipset drivers installed then you SHOULD see the Ryzen balanced power plan in the control panel power options applet. That alone, if you don't might account for some of the power discrepancies itself.

Did you read my reply before that one? ABout the side panel?

And yea I have the AMD Chipset drivers now from AMD themself. Same temps. Im seeing 80-82c at stock settings PBO disabled (boosts to 4.05 on Prime95). And 84-86c with the 4.3ghz all core OC.

The Prism is 92-95c with the all core OC.
 
You can probably live with the 84-86 degree spikes, but if it were my CPU I wouldn't want to see anything higher than 80°C consistently. Since you probably won't see those same temps most of the time you might be ok with it, but if you run ANYTHING that makes frequent use of AVX or AVX2 instructions, and there ARE plenty of games and applications that do, then you will want to keep a close eye on temps and you may end up wanting to drop your manual OC down to 4.2Ghz if you wish to see your piece of silicon last a nice long time.

While it's true that you are unlikely to cause any immediate damage at temps between 80-90°C, consistent temperatures in that range will almost certainly contribute to premature degradation through electromigration and VT shift over the course of months or years.