PC Processors

[mickyfin]

I have paired a new Ryzen 9 5950x with an Asus Crosshair viii hero, but I am suffering really high temperatures. at idle the processor is running at 75C.   Under very light load goes up to 80C

I have tried lot of things and just for clarity I am very experienced in building machines (over 25 years). 

I used a Corsair 100i platinum cooler and after several attempts to cool ( 4 reseats using 3 different CPU thermal pastes) I came initially to the conclusion it was the cooler at fault. 

Wanting to use I bit the bullet and bought a replacement cooler a Kraken X73, I installed and net result was  1C cooler i.e. 74C.  So not the cooler but something else!

I did some searching online to discover others have been having a similar issue multiple motherboards, I found some suggestions to help which were to switch my board into an eco mode (which did nothing) the next suggestion was to disable boosting (which kind of cripples the chip) , this I did and now see temperatures in the region of 45C. I don't know where the problem lies exactly, but if several manufactures are seeing then it kind of points to an AMD issue with the Chip or something they have supplied to the board manufacturers. After days of building and then re-applying the coolers I do feel a little cheated. I tried raising with ASUS but because I registered my motherboard for cash back it is saying the serial number is already registered so cannot raise a support case. I coming on here hoping that someone can give some advice and maybe someone from AMD can help.   

[Ero_Sennin]

@CalinB 

Hi buddy,

Thank you for sharing these information. Let's hope we are getting a solution as soon as possible. Fingers crossed #team5950x

[CalinB]

What kind of temps are you getting now with msi? I’m getting 76c at stock settings full load, 90c PBO on, curve optimiser-14, 4450mhz all core, 1.3v. The case doesn’t have any fans at the moment, just the Aio bear in mind.

[Ero_Sennin]

@CalinB 

Hey buddy. I had tagged you in my post after I got the motherboard. Check here please . Also today I was streaming, had chrome with 10 tabs open, had the music software running along with the stream, had discord running and I got max 77c.

Everything is in stock settings. I only change the XMP to profile 1 so RAM ll get its max factory speed and I have set all PC fans to maximum speed.

But the environment here it's about 23-24 degree. Don't know if it gets warmer how much the CPU could reach.

On idle of course I have also much lower temps compared with my previous motherboard. As well as in BIOS. BIOS also got about 27-34c. sometimes. I also have a photo in idle with stream running.

 

 

Check the photos bellow kindly:

 

Hope with future BIOS and AGESA updates to be even better. Maybe with more cpu stress and more environment heat, temps could go again higher.

However it's much better compared to my previous Asus motherboard.

[giboune]

@Ero_Sennin 

Hello,

You said "Everything is in stock settings. "... PBO is enable with stock setting with MSI mobo ?

PBO is in default setting on Asus but not with all brands.

[Ero_Sennin]

@giboune 

It's in Auto. Don't know what does that mean for the MSI bios.

[Ero_Sennin]

@giboune 

@CalinB 

And everyone. You may check out this fresh news here. It may help us in some cases. I am just reading in time the article I mentioned.

[JoltCola]

PBO "auto" means it's disabled. Should apply to every motherboard.

[Ero_Sennin]

@JoltCola 

Yes that's what I read in first place but just for my previous mobo Asus. And what's the reason to enable it while I can see already that CPU having auto boost etc. its own with the Auto settings? If PBO ll be enabled it means temps ll go even higher right? But from what I see now is that CPU reaches the 4.9GHz max boost when it uses it. And again, this is with the Auto settings in BIOS.

[JoltCola]

Temps won't go higher on single-core, PBO mainly helps multi-core in my experience, and that primarily applies to 5900X and 5950X which are much more power-limited by default.

Curve optimizer will help with temps.

[Ero_Sennin]

@JoltCola 

But unfortunately curve optimizer works easier and better only in the mobo dark hero because it has the Dynamic OC Switcher option. So at the moment I am not sure if I should use these. Maybe when my mobo ll get the agesa combo 1.2.0.0, will try the CTR app I posted above from Guru3D.

Thanks however.

[JoltCola]

I don't have a dark hero and CO works great-- but yes you do want AGESA 1.2.0.0.

[Ero_Sennin]

@JoltCola 

I will wait for the BIOS update and then I may try too. Did you notice difference in temps and power consumption too?

[giboune]

@vice65 @Ero_Sennin 

I reinstalled Windows and now I get 47082pts on CPU Mark !!! (37900 before)

I'm happy...I'm happy like this... good temp with TPU1 or 2, and 1 fan front, 1 on Ventirad and 1 back...

My computer is pretty quiet, temps not so bad... and I read that PBO is not good for gaming !!!

NICE !!!

[JoltCola]

Yes, CO leads to lower temps and/or higher clocks, depending on PBO limits set. Definitely worth messing around with it.

[Ero_Sennin]

@giboune 

Hi buddy. Good to hear. So you had probably installed an app that was absorbing some resources and caused that? However if I were you, I would try to run the benchmark in Auto BIOS settings too and see how the temps and the result would go. Of course with XMP 1.

@JoltCola 

I may try it when I got the BIOS update. If it's fine by you I may also tag you if I need help.

[spykezap]

After two months of frustration i have finally figured it out.

Last Friday i gave up completely, i emailed the shop i bought the kit from and demanded to return all of it and just swap it for a Intel system. My favorite game was not running and the temps just annoyed me while not getting the performance i wanted.

Saturday i had calmed down lol, and figured i'd give the 5950x one last chance. I picked up a Gigabyte X570 Aorus Master and spent the weekend dissassembling my watercooling, redoing all the pipes and installing the new board.

I used the same windows install and booted it all up. Same issue, although temps a bit better.

This morning i noticed all the Asus crap software was still installed. I uninstalled what i could as i like my Windows install tidy. However after every boot i still got asus pop-ups. I investigated further and saw services still running, and one program just refused to uninstall from add/remove programs.

Right, pissed off i googled for removal tools. Only found one that didn't really work. So i started deleting services and cleaning up the registry.

Eventually i said f*** it, and just reinstalled windows completely. I took to heart one comment i had on a reddit post about this issue, saying the Asus (motherboard manufacturer software in general) are crap. So i let windows updates install the drivers it could, then went to AMDs site and ran the auto update tool they have.

My system is finally under control. Idle temps are down to low 30's and StormWorks runs great again! 

Now i can't tell for certain whether the Asus crap caused this but i have a strong suspicion this is it.
For those of you who remember, i have a flatmate with a Asus board and the 5950x, he has similar issues. I'll have him do a reinstall, not download anything from Asus's site and do a check. 

I can't be bothered taking all the watercooling apart again to test this on my Crosshair VIII Hero, I'll stick with the gigabyte board as it has better VRMs anyways.

So, for anyone with this issue, if you have installed anything from Asus's website, try reinstall windows and do not install anything from Asus. Let windows update install your initial drivers, then run this tool from AMD to install latest chipset.

Feel like i finally have the system i payed for which is a really good feeling.

[Ero_Sennin]

Hi buddy @spykezap 

Glad to hear you are having a stable results now. However I knew already about this. Someone had posted something here before if I am not wrong and I was doing the same for my X570 E Gaming Asus.

I had used the Revo Uninstaller program to remove all Asus stuff. But only this that installed by the drivers etc. Because I never used the Aura or Armour Crate.

However in my case things were sti the same. So I had to return my motherboard and while I am waiting a replacement or a refund I bought a new one.

Now I am waiting the new BIOS from MSI to try the undervolting guide that looks promising.

[spykezap]

Hey buddy @Ero_Sennin,

Did you ever try to reinstall windows?
Cuz i did a deep clean of Windows and removed every shred of Asus. But it did not help. Only after reinstalling windows (on the same board) did the system eventually behave.

[Ero_Sennin]

@spykezap 

Yes I did. 2 times. But results were the same.

[MakeshiftApe]

So I'm using an ASUS ROG STRIX B550-E motherboard and a 5950X (with Dark Rock 4 air cooler) and appear to be having a similar issue. I have a theory for what might be causing it, which I've written at the bottom of my post (skip to the bold text if you just want to read my theory). 

The weird thing is it seems like it started recently (or well, worsened).

When I first set up my build (20th January), I found I was getting ~45C idle, 60 to low 70s in games, and temps would occasionally shoot up as high as the high 70s, up to a max of 82C briefly when opening up apps, or when Windows Defender was scanning, etc. 

As of yesterday I noticed the machine seemed noisier than usual (one of my first observations about my new build is how it was near silent). I checked my temps (tried separate times with both hwmonitor and hwinfo64 just in case one of them was giving wrong readings), and noticed I was getting a constant ~80C while playing GTA IV.

Now that game was never the most well optimised so I chalked it up to it being that, and went on with my day. However I noticed lots of other things, including something as simple as opening Chrome or Firefox, would spike the temp to high 70s/low 80s. It seemed like my temperatures were sitting in the low 60s just browsing the web, when I'm sure they were never that high before.

The other day I did get a 1440p 165Hz monitor, and it's possible that the temps have just been higher because of that, and I never noticed the extra fan noise until yesterday - but not very likely since I'm very sensitive to fan noise, especially when the idle noise for my PC is usually almost completely silent, and it's super obvious when the fans start increasing. Still, I mention the new monitor just in case that's part of it.

Anyway, I tried various things. First I located the "Performance Enhancer" setting that was mentioned disabling earlier in this thread, but it was already set to disabled, not auto/enabled. 

Next I tried undervolting, using the Optimum Tech PBO2 undervolting video as a guide. I set a negative offset of -10 just to be cautious, and didn't add anything to the clock speeds. It seemed like my idle temps were maybe a degree better, but didn't seem to do anything for my load temperatures, which still hit 80C when I booted up Hitman 2 (1440p, 120FPS, highest settings, HDR enabled). 

I also tried updating the BIOS. I was using the 1801 beta BIOS from January, and I updated to the new 1804 stable BIOS released this month. Early testing results are mixed. I booted up Hitman 2 and only hit a max of 75C, with temps mostly in the high 60s, rather than high 70s like before.. BUT, I just tried a Windows Defender scan and the temps hit a high of 85C (that's a new record). 

---

Looking at hwmonitor right now while Windows Defender is running, I think I have a theory about what might be causing this issue.

My CPU package temp is 80C, but my CPUTIN, i.e. the motherboard CPU temperature sensor, is only 51C.

Why is this relevant? Well if you look in ASUS AI Suite, the CPU temperature reading there is CPUTIN, not CPU Package temp. So I wonder if ASUS is somehow adjusting voltages/clocks based on CPUTIN rather than package temperatures, and if said sensor readings are too low, causing it to boost voltage + clocks more aggressively than it should at these temperatures.

I seem to recall earlier in the thread someone sharing an email from an AMD rep that mentioned it could be a sensor related issue. So maybe this is it? 

Maybe some of you who are also experiencing this issue could try and check those temperatures and see if you also have low CPUTIN readings. 

[astrosquirrel]

@CalinB 

Why wait? return it now. You clearly have expectations from the product which will never materialize. Get yourself something that makes you happy (however difficult that is). 

[trek]

---

@Ero_Sennin wrote:

Sorry but it's possible. I have completely removed all the films from mobo and cooler back. When I used the PBO, CPB and I used the performance enhancer to Default or Level 1, when I run the CPU mark, the temps went 100-105c and PC rebooted.

Unfortunately I can't do it and show you again because I already sent the mobo to the shop for warranty stuff etc.

It seems the only key motherboard from Asus for the Ryzen 5950x as much as I read, is the Dark Hero because it has the Dynamic OC Switcher.

---

PBO is overclocking and if you enable it power limits are maximized (PPT, TDC, EDC). I am shocked that you are complaining whole time about temperatures while overclocking without power limits.

Dynamic OC switcher is nice feature but some do not use it even if tehey have Dark Hero. If you set PBO correctly, do curve optimizer and set vcore offset negative in Asus motherboard, then it will yield better performance than manual overclocking if you have good cooling (custom water loop).

[astrosquirrel]

Here's a dump of a portion of time of typical usage from my system.

https://drive.google.com/file/d/1AD3abcoTHQKaEnFy83zuwbHQ0mS6NuuC/view?usp=sharing

It starts with me watching a youtube video, Chrome remote desktop session up opening steam and browsing the store. 

Then I close those programs and let it idle (except for the logging of course). No background processes killed.

Then I do an R23 Cinebench multicore

Then a Single core

Then close programs and open an excel book

The file graphs mainly the CPU Die, Socket and Package temps . You can easily go in the graph and plot the other data points including fan speeds. Graph's at the bottom of the spreadsheet

As you can see, idling at 45 (due to logging, goes to 40-42 normally), with 8 out of 11 fans completely off and the 3 remaining in silent operation.

In this run, the CPU got to a max of about 73 degrees in the single core R23 run. 

For comparisons, here's my config:

5950x 

ROG Hero VIII WiFi

32GB GSkill Trident 3600MHz 16, 19, 19, 19, 39

Corsair H150i PRO XT AIO (Balanced Pump at 2300 RPM)

Aorus RTX 3070 Master

Asus Essence STX 

WD SN750 1TB NVME

Corsair Obsidian 900D case

3 120mm front intake fans

1 140mm rear intake fans

3 120mm fans on the Radiator exhausting

3 fans on the graphic card

1 fan never switches on on the PSU

Next in my list is running undervolting with PBOII....just haven't got the time for the moment for that. 

[Ero_Sennin]

@trek 

Should I mention again that I had try the guide from a great YouTuber that undervolted the 5950x and I did the same but I had not the dark hero like him and my temps went the same? And I used everything he mentioned about PBO and Curve Optimizer.

And how you talk about OC when I don't even touch the stock auto frequency of the CPU? :/

[trek]

---

@Ero_Sennin wrote:

And how you talk about OC when I don't even touch the stock auto frequency of the CPU? :/

---

Just enabling PBO implicates massive overclocking. The best is first to test system in stock settings (load optimized defaults in bios) and just set memory XMP profile. Then if everything works correctly enable PBO and do curve optimizer. If temperature is high one must lower limits. It is simple.

If one requires enthusiast temperatures with PBO, she must do very good custom water cooling otherwise welcome to 90C territories.

[Ero_Sennin]

@trek 

When I started my PC for a first time the only thing I did was to change ram to profile for its 3600MHz speed that is its max speed out of the box.

I didn't change anything else. Then I had again in idle 60-70c and in load (not full) more than 80c. Then I tried to find solutions and watching guides and reading articles. But nothing worked.

Only the TPU setting 1 ot 2 that reduced frequency to 4 and 4.1GHz without auto boost etc. and again I had max load of 79-80 that wasn't satisfied considering I have very good air flow and 360m CoolerMaster liquid cooler AIO as I mentioned before.

[trek]

---

@Ero_Sennin wrote:

Then I had again in idle 60-70c and in load (not full) more than 80c. Then I tried to find solutions and watching guides and reading articles. But nothing worked.

---

Then it was working as expected, you could not find any guides because your system worked perfectly as it should. Temperature "issue" is non existent and virtual. It is because some people have expectations not aligned with reality, they want to bend reality to their wishes, but in such a case they would have to throw in custom high performance water cooling - and some people are indeed are using it and they have lower temperatures.

AMD clearly said that 90C is ok for 5950x in full load, 80C in partial load is also ok. Zen 3 is 7nm which means higher heat density so it is more difficult to cool it. AIO performace is not as great as many thinks, custom water loops is answer for teperature sensitive enthusiasts.

[Chibi28]

Hopefully managed to add a video. This was at stock bios defaults. Any idea why temp went down when stress test was applied?

[astrosquirrel]

---

@Chibi28 wrote:

---

Not visible yet but will give it a gander once it's up. :smileyhappy:

[astrosquirrel]

---

@Chibi28 wrote:

Hopefully managed to add a video. This was at stock bios defaults. Any idea why temp went down when stress test was applied?

---

I myself can't see it unfortunately. It won't play and if I go into details it tells me it's pending approval.

[Chibi28]

Same here. Waiting approval. Hopefully it means it's waiting for mod approval. 

[astrosquirrel]

---

@trek wrote:

---

@Ero_Sennin wrote:

Then I had again in idle 60-70c and in load (not full) more than 80c. Then I tried to find solutions and watching guides and reading articles. But nothing worked.

---

Then it was working as expected, you could not find any guides because your system worked perfectly as it should. Temperature "issue" is non existent and virtual. It is because some people have expectations not aligned with reality, they want to bend reality to their wishes, but in such a case they would have to throw in custom high performance water cooling - and some people are indeed are using it and they have lower temperatures.

AMD clearly said that 90C is ok for 5950x in full load, 80C in partial load is also ok. Zen 3 is 7nm which means higher heat density so it is more difficult to cool it. AIO performace is not as great as many thinks, custom water loops is answer for teperature sensitive enthusiasts.

---

I agree on the feedback you're giving to the child. His system was working properly but shhhh he's going to flip and call us AMD employees 

On the other hand, I've seen some people here with watercooled loops an temps much higher than mine. So far we haven't been able to work it out why. 

An AIO is an alright step up from an air cooler but will never replace a custom loop, so I'm a bit at a stump as to why there's particularly a user in this thread that can't get his temps under control with that. 

Has anyone (other than the child) tried undervolting with PBOII and the new BIOS?

[Chibi28]

---

@trek wrote:

Streefights, you bought high performance workstation processor 5950x and you are complaining about performance and related power consumption (which is btw smaller than Intel cpus and performance is much better), which is illogical.

As  5950x produces much less heat than Intel competing products which have worse performance your conclusions are invalid.

Nevertheless I have solution for you, you can set with 5950x desired throttling temperature you want in BIOS, it is called Plaform Thermal Throttle Limit and it will do what you need and want while keeping performance, set this limit to 70C if you want to have less than or equal temp to 70C for example, keep all other settings on Auto

This is advantage of Zen 3 processors, you can run them how you want. There is also new curve optimizer which allows undervolting and increasing performance - used with Plaform Thermal Throttle Limit I have better performance than stock with much lower temperatures.

I think you're missing the point. Many people that buy a CPU don't want to fiddle around with settings once they buy it. They just want to install it and use it

If you want to tinker with settings to get extra performance thats an extra, but the CPU should work and not have these issues that everyone is seeing.

---

[Ero_Sennin]

Hi @Chibi28 

I have seen this option but if I set it for example to 70 degrees then it means we are getting less performance because CPU will not use higher frequencies to not increase temps? If yes, then what's the point to get a 950€ CPU and not use it at its max capabilities? Foul for AMD I guess.

Also the curve optimizer came with the AGESA version 1.1.8.0 and while I tried it with negative effect for all cores and magnitude to 12, still no difference in power nor then temps.

Something's wrong here and maybe it's that we need a finally proper BIOS and AGESA update for the motherboards. Because as you can see many people with different mobos experiencing this.

If I wanted just to lower the temps in sacrifice of performance is the only easy thing. I can enabled CPPC Cores, Global C State, Efficiency Mode and use the TPU setting to 1 or 2.

That means no more auto boost, no more higher frequency and max temp about 70 which is safe for the CPU. But performance ll not be the same.

[trek]

---

@Ero_Sennin wrote:

I have seen this option but if I set it for example to 70 degrees then it means we are getting less performance because CPU will not use higher frequencies to not increase temps? If yes, then what's the point to get a 950€ CPU and not use it at its max capabilities? Foul for AMD I guess.

It is you who insist on having low temperatures while AMD said that 90C is ok, you have option to set CPU however you want and you complain :-/

I have set PBO throttle to 80C and I have better performance with curve optimizer set to -10 all core.

Setting curve optimizer alone will not drop temps perhaps, it will increase performance so you can lower thermal limit.

I understand why people are worried, when I had Intel and looked in datasheet max package temp was something like 74C and anything over that meant CPU damage in long term, but AMD said 90C is ok so it is ok.

Basically people here require fix on something that is not broken.

[Ero_Sennin]

Are you sure is just me? Did you read every single post here since start? Probably not. And even if AMD was mentioning this, do we have to let a CPU running the most hours of the day at 90 degrees? Who's telling us that this is safe for future usage in time?

Also the only motherboard that seems to have the solution for this CPU is the Dark Hero III due to its special option Dynamic OC Switcher that if it combined with the Curve Optimizer it makes amazing results.

I have posted a video in previous post from a YouTuber that tried this. If you ask me when I run GTA V ultra settings with RTX 3090 and I am using always the Auto settings in BIOS I didn't try any overclock because I don't need it, I can see some soft performance reduce when the temps when 89-90.

Is this the "90-95 degrees for this CPU is fine" that AMD claims? Then why the YouTubers in their reviews back when this CPU released, had not these temperatures? This is another thing mentioned by people in this thread here.

Did they had the perfect CPU units for reviews? Just saying.

[trek]

---

@Ero_Sennin wrote:

Is this the "90-95 degrees for this CPU is fine" that AMD claims? Then why the YouTubers in their reviews back when this CPU released, had not these temperatures? This is another thing mentioned by people in this thread here.

---

AMD does not claim that 90 -95C is fine, if you take a look at advertised CPU specifications then e.g. 5900x has max temp 90C as described by AMD.

Your temperatures are ok, if you have 89-90C while gaming with 3090 then it is ok, if you want better you must add cooling - I suppose better case cooling, custom water loops for bot CPU and GPU etc.

Also I think that people should start separate discussion threads because every "issue" is different.

In your case I suppose cause of higher than expected CPU temperature while gaming is that you probably have bad case cooling, you can test what happens if you remove side panels etc.

[Ero_Sennin]

I had air cooling and I changed recently to CoolerMaster MasterLiquid 360R liquid cooling that is suggested by CoolerMaster and AMD for the 5950x. And I used an extra thermal paste not the package one. I used the CoolerMaster MasterGel Maker.

Also people here with Kraken Z73 liquid cooling have the same situations. I have posted AMD support answers to me back in some posts here while the employee were saying that these temps are fine.

But if we get performance drop I can't see any fine thing here. I have tried also guides from YouTubers and the only thing I got is even more temperature.

In one test it reached even 100 degree for a first time imagine. I am also using 7 120 CoolerMaster sickerflow fans.

I mean air flow is good and I can sense it. Anyway. As you may understand, this situation made a lot of us coming here in the forums to find a solution at all.

I just hope it's about time until we got the right BIOS. Otherwise I may request a CPU replacement or Motherboard while my PC it's just new.

[trek]

You can check this with thermal camera, e.g. while playing Cyberpunk I have following burning 3080 brick in my PC case (the most yellow color is toward 80 - 90C - color range is 40C blue to 80C white), heatsink is black due to emmisivity, but under heatsink you can see "burning" motherboard

If you have 3090 you need massive PC case cooling or 3090 turbo fan version. Remove all side panels for test, you might have bad airflow with hot spots.

[Chibi28]

---

@trek wrote:

You can check this with thermal camera, e.g. while playing Cyberpunk I have following burning 3080 brick in my PC case (the most yellow color is toward 80 - 90C - color range is 40C blue to 80C white), heatsink is black due to emmisivity, but under heatsink you can see "burning" motherboard

If you have 3090 you need massive PC case cooling or 3090 turbo fan version. Remove all side panels for test, you might have bad airflow with hot spots.

---

High temps with gaming is not the initial problem this thread was dealing with. It was a newly installed 5950X running hot (70C+) just loading into windows