PC Pc Gaming/building thread

[dwwaino]

OK, home now. Funny, mine was sitting at 60'ish for a while even after booting up. After about 5 minutes it has come down to 45 and bouncing around a bit. It is a bit stuffy in here at the moment though so that might be due to ambient temp. That's sort of a quirk of AIOs and water though, they'll do great at limiting the peak load temp but won't do much to those CPU spikes.

It might not be what you want to hear but the h510 case is known to run a bit warm in some configurations. But that should affect the GPU more than the CPU, so let's just run through absolutely everything just to double check. Most obvious is the check the pump is mounted properly. I've experienced before that not having the block screwed down evenly and having even contact was enough to raise the temps. If you don't have any spare thermal paste on hand then I wouldn't go ripping it off just yet, maybe just try to re-fasten it. If you do take it off because you've got paste on hand then make sure to clean it all off properly and reapply. Don't run all of those monitoring programs at the same time either if you are. Not because you can crash (and you probably will, they don't like each other), but those are all processes hooking into your CPU at pretty low level which will activate more cores and drive the temps up. If you aren't using Ryzen Master for anything other than monitoring then I'd bugger it off, or just don't open it. Just iCue is fine, I wouldn't worry about CoreTemp or HWINFO or anything like that unless you're overclocking. It also won't hurt to update your BIOS just in case your board manufacturer has a better default profile that runs at lower stock voltages or something.

As long as you've got no fancy power boosting or overclocking settings going on in BIOS then the temps should be nothing firmware or software related. Since you said you were unsure with the fan settings then here is a full run through in my own opinion. Others will obviously have different ideas but this is generally how I roll.

Firstly, AIO setup. You might already know but I'll go through it anyway just for future reference or if anyone else wants to know. The pump cable can go to either the AIO header on your motherboard (if you have one) or to either of the CPU fan headers. The difference is that the AIO header is default to 100% speed but depending on make and model of motherboard they both should be PWM. The CPU header is designed for safety in that if the reading fails then you'll get a warning and the PC won't boot to windows. There is a myth around that a pump needs to run at 100% 24/7, but they just like to run at a constant speed. I run mine at 70% 24/7 which is about ~2200rpm on my z73. From there you can plug your fans into either the splitter from the pump or to the CPU headers. Personally I just plugged my pump into the AIO header and fans into the pump, and disabled CPU fan monitoring in BIOS so it would let me boot into Windows.

Fan profiles. On an AIO you don't need the fans reacting to every little spike. Not only is it unnecessary but the noise of constantly ramping up and down will do your head in. As long as you have a good water flow then your idle/silent profile you should just set to a maximum to where the fans are still inaudible. That's my preference anyway. I set my intakes (in my case they are the radiator fans as mine is mounted to the front of the case) to about 50% which is right now saying about ~1230rpm and my exhausts on the absolute minimum that CAM lets me which is 20%, about 600rpm. The reason is that it's the water in the loop that is carrying the heat away from the CPU and the air flow isn't as important. Right now my liquid is sitting at 30c and when I game it will take at least a few hours to hit 35c. As long as that water is staying cool and the pump is moving the water through at a decent rate then your max temps are going to stay pretty checked. But that's why it won't do much for those intermittent spikes. I run the intakes harder too just to try and do my best to achieve positive pressure. It's probably not perfect as I've got 4 exhaust vs 3 intakes but hopefully it's helping keep some of the dust out.

So in saying all that I've setup my case fan profiles actually for my GPU temp as the CPU doesn't really need it, and obviously when I'm playing games the CPU is going to warm up anyway. I monitored my GPU to find it's usual gaming temp just so I knew where to ramp it up so my fans don't just start buzzing about because I loaded a YouTube vid or something.

You'll be able to do something similar in iCue. Even under load I only run my exhausts at 50%. The case doesn't need to be a wind tunnel, just a nice steady air flow.



And as a comparison my intakes (which are on the radiator and connected to the pump splitter). I found anything above 80% and they start getting a little noisy.




Ideally your idle temps need to come down a bit. Even after writing this and my room air con has cooled everything down my 5900x has jumped up to around 50c a few times before coming back down to 43-45c. Clearly this humidity is still affecting the ambient temp a bit. You might want to check your peak load temp while you're at it as well. Just play a game for a little while or grab Cinebench from the Windows store and play with both multicore and single core tests. For me my max load temp is around 65c, yours might be higher but if it's above 75-80c then I'd definitely look at removing the block, reapplying the paste and re-fitting it.

 

[sataris]

OK, home now. Funny, mine was sitting at 60'ish for a while even after booting up. After about 5 minutes it has come down to 45 and bouncing around a bit. It is a bit stuffy in here at the moment though so that might be due to ambient temp. That's sort of a quirk of AIOs and water though, they'll do great at limiting the peak load temp but won't do much to those CPU spikes.

It might not be what you want to hear but the h510 case is known to run a bit warm in some configurations. But that should affect the GPU more than the CPU, so let's just run through absolutely everything just to double check. Most obvious is the check the pump is mounted properly. I've experienced before that not having the block screwed down evenly and having even contact was enough to raise the temps. If you don't have any spare thermal paste on hand then I wouldn't go ripping it off just yet, maybe just try to re-fasten it. If you do take it off because you've got paste on hand then make sure to clean it all off properly and reapply. Don't run all of those monitoring programs at the same time either if you are. Not because you can crash (and you probably will, they don't like each other), but those are all processes hooking into your CPU at pretty low level which will activate more cores and drive the temps up. If you aren't using Ryzen Master for anything other than monitoring then I'd bugger it off, or just don't open it. Just iCue is fine, I wouldn't worry about CoreTemp or HWINFO or anything like that unless you're overclocking. It also won't hurt to update your BIOS just in case your board manufacturer has a better default profile that runs at lower stock voltages or something.

As long as you've got no fancy power boosting or overclocking settings going on in BIOS then the temps should be nothing firmware or software related. Since you said you were unsure with the fan settings then here is a full run through in my own opinion. Others will obviously have different ideas but this is generally how I roll.

Firstly, AIO setup. You might already know but I'll go through it anyway just for future reference or if anyone else wants to know. The pump cable can go to either the AIO header on your motherboard (if you have one) or to either of the CPU fan headers. The difference is that the AIO header is default to 100% speed but depending on make and model of motherboard they both should be PWM. The CPU header is designed for safety in that if the reading fails then you'll get a warning and the PC won't boot to windows. There is a myth around that a pump needs to run at 100% 24/7, but they just like to run at a constant speed. I run mine at 70% 24/7 which is about ~2200rpm on my z73. From there you can plug your fans into either the splitter from the pump or to the CPU headers. Personally I just plugged my pump into the AIO header and fans into the pump, and disabled CPU fan monitoring in BIOS so it would let me boot into Windows.

Fan profiles. On an AIO you don't need the fans reacting to every little spike. Not only is it unnecessary but the noise of constantly ramping up and down will do your head in. As long as you have a good water flow then your idle/silent profile you should just set to a maximum to where the fans are still inaudible. That's my preference anyway. I set my intakes (in my case they are the radiator fans as mine is mounted to the front of the case) to about 50% which is right now saying about ~1230rpm and my exhausts on the absolute minimum that CAM lets me which is 20%, about 600rpm. The reason is that it's the water in the loop that is carrying the heat away from the CPU and the air flow isn't as important. Right now my liquid is sitting at 30c and when I game it will take at least a few hours to hit 35c. As long as that water is staying cool and the pump is moving the water through at a decent rate then your max temps are going to stay pretty checked. But that's why it won't do much for those intermittent spikes. I run the intakes harder too just to try and do my best to achieve positive pressure. It's probably not perfect as I've got 4 exhaust vs 3 intakes but hopefully it's helping keep some of the dust out.

So in saying all that I've setup my case fan profiles actually for my GPU temp as the CPU doesn't really need it, and obviously when I'm playing games the CPU is going to warm up anyway. I monitored my GPU to find it's usual gaming temp just so I knew where to ramp it up so my fans don't just start buzzing about because I loaded a YouTube vid or something.

You'll be able to do something similar in iCue. Even under load I only run my exhausts at 50%. The case doesn't need to be a wind tunnel, just a nice steady air flow.

View attachment 1029559

And as a comparison my intakes (which are on the radiator and connected to the pump splitter). I found anything above 80% and they start getting a little noisy.

View attachment 1029562


Ideally your idle temps need to come down a bit. Even after writing this and my room air con has cooled everything down my 5900x has jumped up to around 50c a few times before coming back down to 43-45c. Clearly this humidity is still affecting the ambient temp a bit. You might want to check your peak load temp while you're at it as well. Just play a game for a little while or grab Cinebench from the Windows store and play with both multicore and single core tests. For me my max load temp is around 65c, yours might be higher but if it's above 75-80c then I'd definitely look at removing the block, reapplying the paste and re-fitting it.

My pump is running at 30% for 2851RPM XD

 

[dwwaino]

My pump is running at 30% for 2851RPM XD

Holy s**t, what is the full speed of that thing? lmao

 

[sataris]

Holy sh*t, what is the full speed of that thing? lmao

Let me turn it to 100% PWM for shits and giggles.

 

[sataris]

You can almost see the point where it went 100% PWM.


Also... There is a tiny whirlpool in the tank right now.

 

[sataris]

View attachment 1029574

You can almost see the point where it went 100% PWM.


Also... There is a tiny whirlpool in the tank right now.

Running the pump at 100% PWM did 3 things

1) Reduced the Temp over ambient by about 10c.

2) Caused a whirlpool in the res.

3) Caused a tiny, tiny leak in the draining port (Nothing electronic near it)

 

[dwwaino]

Running the pump at 100% PWM did 3 things

1) Reduced the Temp over ambient by about 10c.

2) Caused a whirlpool in the res.

3) Caused a tiny, tiny leak in the draining port (Nothing electronic near it)

4) caused a ripple in space time

 

[Hate]

t might not be what you want to hear but the h510 case is known to run a bit warm in some configurations. But that should affect the GPU more than the CPU, so let's just run through absolutely everything just to double check.

I somewhat knew this going into buying it but thought I would hedge my bets as I wanted something rather small and thought that because I don't plan on doing anything heavy I could get away with it. Even if temps remain a problem, in hindsight I would probably go another way because of how tight a fit everything is.

Most obvious is the check the pump is mounted properly. I've experienced before that not having the block screwed down evenly and having even contact was enough to raise the temps. If you don't have any spare thermal paste on hand then I wouldn't go ripping it off just yet, maybe just try to re-fasten it. If you do take it off because you've got paste on hand then make sure to clean it all off properly and reapply. Don't run all of those monitoring programs at the same time either if you are. Not because you can crash (and you probably will, they don't like each other), but those are all processes hooking into your CPU at pretty low level which will activate more cores and drive the temps up. If you aren't using Ryzen Master for anything other than monitoring then I'd bugger it off, or just don't open it. Just iCue is fine, I wouldn't worry about CoreTemp or HWINFO or anything like that unless you're overclocking. It also won't hurt to update your BIOS just in case your board manufacturer has a better default profile that runs at lower stock voltages or something.

It seems like it is mounted OK. I had some trouble as I didn't have one of the hooks extended long enough but the videos I saw said it should bow (?) as it is applying even pressure. I purchased ArcticClean/Arctic Silver 5 because it was the first time I would have ever gone near one of these coolers and would say I was not confident. I assume it is just applied, wait, wipe on both CPU. BIOS is one behind the AGESA patch D released on 8th/12 - whatever that is for. Have removed all other monitoring software now. Where exactly in iCue can I view the CPU temp?

irstly, AIO setup. You might already know but I'll go through it anyway just for future reference or if anyone else wants to know. The pump cable can go to either the AIO header on your motherboard (if you have one) or to either of the CPU fan headers. The difference is that the AIO header is default to 100% speed but depending on make and model of motherboard they both should be PWM. The CPU header is designed for safety in that if the reading fails then you'll get a warning and the PC won't boot to windows. There is a myth around that a pump needs to run at 100% 24/7, but they just like to run at a constant speed. I run mine at 70% 24/7 which is about ~2200rpm on my z73. From there you can plug your fans into either the splitter from the pump or to the CPU headers. Personally I just plugged my pump into the AIO header and fans into the pump, and disabled CPU fan monitoring in BIOS so it would let me boot into Windows.

I am running an x570 ASRock Steel Legend. I just put the 3 pin from the AIO pump into CPU_FAN2/WP FROM CPU_FAN1 (should have done that already I guess?). Fans LEDs are all connected to lighting node core The front radiator fans are plugged into the splitter from the pump. The rear exhausts are on a PWM splitter to CHA_FAN3/WP. I put all settings back to Auto/Standard and it bobbled around 60+ on start up, sat back down to high-mid 50s and I played a game for about 15 minutes and it was high 60s. I have no clue why it went up in the 70s at idle before.

This is where I start to get confused and it all goes over my head, haha. BIOS says the fans on the CPU_FAN2/WP header are running anywhere from 0 to 5k rpm, while the CHA3 says under 1k rpm. iCue has the 2 fans on the h100i fairly low around 600 rpm. CPU2 is set to Auto (DC and PWM selectable) and Standard (Custom, Silent, Standard, Performance - I guess custom is where I would set the stuff to do with the curve graph I have seen). CHA fans are Auto/Standard too now and set to "Monitor MB" which there is no option for on the other CPU fans. Where and what exactly is controlling all these fans? And what should I be setting these to?

Fan profiles. On an AIO you don't need the fans reacting to every little spike. Not only is it unnecessary but the noise of constantly ramping up and down will do your head in. As long as you have a good water flow then your idle/silent profile you should just set to a maximum to where the fans are still inaudible. That's my preference anyway. I set my intakes (in my case they are the radiator fans as mine is mounted to the front of the case) to about 50% which is right now saying about ~1230rpm and my exhausts on the absolute minimum that CAM lets me which is 20%, about 600rpm. The reason is that it's the water in the loop that is carrying the heat away from the CPU and the air flow isn't as important. Right now my liquid is sitting at 30c and when I game it will take at least a few hours to hit 35c. As long as that water is staying cool and the pump is moving the water through at a decent rate then your max temps are going to stay pretty checked. But that's why it won't do much for those intermittent spikes. I run the intakes harder too just to try and do my best to achieve positive pressure. It's probably not perfect as I've got 4 exhaust vs 3 intakes but hopefully it's helping keep some of the dust out.

The little temp icon on the pump head in iCue, I am assuming is the liquid temp? Pls excuse my ignorance on all this - I do appreciate the response immensely. I have not built for near 6 years, and it was BigFooty who also helped me then too.

Chucking the AC in the home seems to have done something.

 

[Hate]

Back at 70s on idle.

 

[dwwaino]

I somewhat knew this going into buying it but thought I would hedge my bets as I wanted something rather small and thought that because I don't plan on doing anything heavy I could get away with it. Even if temps remain a problem, in hindsight I would probably go another way because of how tight a fit everything is.


It seems like it is mounted OK. I had some trouble as I didn't have one of the hooks extended long enough but the videos I saw said it should bow (?) as it is applying even pressure. I purchased ArcticClean/Arctic Silver 5 because it was the first time I would have ever gone near one of these coolers and would say I was not confident. I assume it is just applied, wait, wipe on both CPU. BIOS is one behind the AGESA patch D released on 8th/12 - whatever that is for. Have removed all other monitoring software now. Where exactly in iCue can I view the CPU temp?


I am running an x570 ASRock Steel Legend. I just put the 3 pin from the AIO pump into CPU_FAN2/WP FROM CPU_FAN1 (should have done that already I guess?). Fans LEDs are all connected to lighting node core The front radiator fans are plugged into the splitter from the pump. The rear exhausts are on a PWM splitter to CHA_FAN3/WP. I put all settings back to Auto/Standard and it bobbled around 60+ on start up, sat back down to high-mid 50s and I played a game for about 15 minutes and it was high 60s. I have no clue why it went up in the 70s at idle before.

This is where I start to get confused and it all goes over my head, haha. BIOS says the fans on the CPU_FAN2/WP header are running anywhere from 0 to 5k rpm, while the CHA3 says under 1k rpm. iCue has the 2 fans on the h100i fairly low around 600 rpm. CPU2 is set to Auto (DC and PWM selectable) and Standard (Custom, Silent, Standard, Performance - I guess custom is where I would set the stuff to do with the curve graph I have seen). CHA fans are Auto/Standard too now and set to "Monitor MB" which there is no option for on the other CPU fans. Where and what exactly is controlling all these fans? And what should I be setting these to?


The little temp icon on the pump head in iCue, I am assuming is the liquid temp? Pls excuse my ignorance on all this - I do appreciate the response immensely. I have not built for near 6 years, and it was BigFooty who also helped me then too.

Chucking the AC in the home seems to have done something.

Just out of curiosity, you've taken the plastic cover off the plate of the pump? I know it's a stupid question, but they're clear and sometimes a really snug fit and it's surprising the amount of people I've seen accidentally leave them on because they honestly didn't notice it. I doubt you've done that though because usually the temps are closer to 100 when people do that lol. They have pre-applied paste too which should be fine, but if you're going to reapply it just to tick it off the list then that Arctic Clean stuff is good. I have some of that somewhere myself. Just make sure it's super clean and dry before reapplying the paste.

It doesn't matter which CPU header the pump is plugged into, they both do the same thing. Leaving the fans on PWM in BIOS is fine because then software like iCue can control them. If you're going to set the fan speeds in iCue I wouldn't worry too much about what BIOS profile is set on them because iCue is just going to override them. If the CPU fans are still spinning at 600rpm then that would likely be a preset cooling profile. Sounds like a silent profile. You should be able to create a new profile and change the curve for those inside the software.

It hasn't been long since I used iCue but I've already forgotten where some of the stuff is off the top of my head. I think there is a Dashboard tab or something? From memory you can set up a custom cooling display and you can add CPU package temp to that.

What game were you playing for those temps? I was just playing Destiny for about an hour and didn't go above 55c, but when I play Cyberpunk I'm usually about 65c. So it's game determined. If you had a serious issue I'd be expecting temps closer to 80s. Not sure why the idle temp is so high though. Do you have HWINFO? If not grab it here. This is to check CPU voltage and core speeds. I'm wondering if your cores are all synced for somereason?

HWiNFO - Free System Information, Monitoring and Diagnostics

Free Hardware Analysis, Monitoring and Reporting. In-depth Hardware Information, Real-Time System Monitoring, Reporting & more

www.hwinfo.com

Then grab Cinebench from Windows store and we'll go from there.

Of course reapply the thermal paste and reseat the pump first, just in case that was the problem.

 

[Hate]

Just out of curiosity, you've taken the plastic cover off the plate of the pump? I know it's a stupid question, but they're clear and sometimes a really snug fit and it's surprising the amount of people I've seen accidentally leave them on because they honestly didn't notice it. I doubt you've done that though because usually the temps are closer to 100 when people do that lol. They have pre-applied paste too which should be fine, but if you're going to reapply it just to tick it off the list then that Arctic Clean stuff is good. I have some of that somewhere myself. Just make sure it's super clean and dry before reapplying the paste.

It doesn't matter which CPU header the pump is plugged into, they both do the same thing. Leaving the fans on PWM in BIOS is fine because then software like iCue can control them. If you're going to set the fan speeds in iCue I wouldn't worry too much about what BIOS profile is set on them because iCue is just going to override them. If the CPU fans are still spinning at 600rpm then that would likely be a preset cooling profile. Sounds like a silent profile. You should be able to create a new profile and change the curve for those inside the software.

It hasn't been long since I used iCue but I've already forgotten where some of the stuff is off the top of my head. I think there is a Dashboard tab or something? From memory you can set up a custom cooling display and you can add CPU package temp to that.

What game were you playing for those temps? I was just playing Destiny for about an hour and didn't go above 55c, but when I play Cyberpunk I'm usually about 65c. So it's game determined. If you had a serious issue I'd be expecting temps closer to 80s. Not sure why the idle temp is so high though. Do you have HWINFO? If not grab it here. This is to check CPU voltage and core speeds. I'm wondering if your cores are all synced for somereason?

HWiNFO - Free System Information, Monitoring and Diagnostics

Free Hardware Analysis, Monitoring and Reporting. In-depth Hardware Information, Real-Time System Monitoring, Reporting & more

www.hwinfo.com

Then grab Cinebench from Windows store and we'll go from there.

Of course reapply the thermal paste and reseat the pump first, just in case that was the problem.

I removed a lot of plastic (one of my ram sticks had a plastic triangle stuffed in the side...), so can’t recall exactly but I would say I probably removed it as I did explore the cooler for a good while.

I was playing A Plague Tale on 1080p Ultra.

I’ll try reinstall it all tonight or tomorrow and then play around with the settings/downloads you have suggested. Again, appreciate the guidance.

 

[Hate]

At first the pump was a bit lose - it seemed to lower the temps when tightened but then back to idling very high, so I spent the last few hours taking it off and cleaning the CPU and block. I went back to see how it is installed to see if I missed something. Pea sized thermal paste, tightening the thumb screws evenly until snug. Not sure how I could do it any cleaner.

First boot I saw the temp as low as 36c, and then it was idling at 40-45c. Restarted a few times and it seemed ok. I ran Cinebench (r23) and got 19,720 MC and 1,562 SC with temps around 68-72C. Let it sit idle for a while after the SC (temp dropped after MC) run and temps would not drop below 70c.

iCue does not allow for me to display CPU temp package, just a list full of load %.

Installed HWiNFO but not sure what to look for. Idling mid 40s at the moment, will let it go until I head to sleep and check then and again in the AM.

 

[dwwaino]

In HWinfo you just click on sensors and sift through the piles of data. There is definitely a way for iCue to display CPU temp, I just can't remember exactly where to find it without connecting and booting my old system back up. I had a custom dashboard that I setup in it that I used to run on my second monitor (when I used one) that displayed my fan speeds, CPU and GPU temps as well as GPU usage.

Those temps sound much better. When you say the temps wouldn't drop below 70c after some multicore passes in Cinebench did you mean during the runs? Or after? Because that's normal during a multicore load. Especially a synthetic load like Cinebench which will absolutely flog the cores, that's normal. I was hitting 70c myself during a longer Cyberpunk session before. In the past I only noticed it around 65c, but it uses more cores than some other games I play so it's going to get a bit warmer and not boost as high. During Destiny it will rarely see it above 60c, usually around 55c because the game pretty much only runs off a single core.

If it's idling at 30s-40s then jumps to 50s for a while it could be just Windows or something doing something in the background.

Unless I've misread your post then I think you're all good now, so HWInfo might not be required. Maybe the block just wasn't making good, even contact? It happens to everyone. As much as it's a pain to stuff around with it's good of that's what it was and you've now fixed it. I don't know how it has happened considering there is a rigid mounting bracket, but my block is ever so slightly rotated and I really don't want to go fiddling with it when it is all running so well lol.

I was curious if you had some sort of all core boost enabled that you weren't sure about (blame the motherboard) as that would cause more heat even at idle. What I was going to suggest was running single core Cinebench passes and looking at HWinfo to see if only one or two of all the cores boosted. You can still do this if you want just to have a look at how it works under the hood. Under single core my first core peaks at around 4974mhz which I was really happy with. Then running multicore passes you'll see all the cores engage, the clock speeds won't boost as much and the temps will shoot up more. Idling in the 40s seems pretty normal though from what I've seen from other owners, and gaming from 55-70c is perfectly fine.

The 5900x is an awesome CPU. First time I've been excited about a CPU in a long time. Probably since the 2500k.

 

[Hate]

In HWinfo you just click on sensors and sift through the piles of data. There is definitely a way for iCue to display CPU temp, I just can't remember exactly where to find it without connecting and booting my old system back up. I had a custom dashboard that I setup in it that I used to run on my second monitor (when I used one) that displayed my fan speeds, CPU and GPU temps as well as GPU usage.

Those temps sound much better. When you say the temps wouldn't drop below 70c after some multicore passes in Cinebench did you mean during the runs? Or after? Because that's normal during a multicore load. Especially a synthetic load like Cinebench which will absolutely flog the cores, that's normal. I was hitting 70c myself during a longer Cyberpunk session before because it uses more of my 5900x cores than Destiny does. Easiest way to see is that during Cyberpunk my clock speeds bounce between 4.5-4.8ghz, where as Destiny I usually get the full 4.9ghz because it pretty much only runs off the single core. Less cores working means less heat which is why it usually only sits around 55c when I play that.

If it's idling at 30s-40s then jumps to 50s for a while it could be just Windows or something doing something in the background.

Unless I've misread your post then I think you're all good now, so HWInfo might not be required. Maybe the block just wasn't making good, even contact? It happens to everyone. As much as it's a pain to stuff around with it's good of that's what it was and you've now fixed it. I don't know how it has happened considering there is a rigid mounting bracket, but my block is ever so slightly rotated and I really don't want to go fiddling with it when it is all running so well lol.

I was curious if you had some sort of all core boost enabled that you weren't sure about (blame the motherboard) as that would cause more heat even at idle. What I was going to suggest was running single core Cinebench passes and looking at HWinfo to see if only one or two of all the cores boosted. You can still do this if you want just to have a look at how it works under the hood. Under single core my first core peaks at around 4974mhz which I was really happy with. Then running multicore passes you'll see all the cores engage, the clock speeds won't boost as much and the temps will shoot up more. The 5900x is an awesome CPU and I could see little reason to use any power boosting or overclocking features given its stock performance. Idling in the 40s seems pretty normal though from what I've seen, and gaming from 55-70c is perfectly fine.

Still somewhat concered. Hopefully this explains better than my previous post.

What I mean in regards to Cinebench is that I ran a multi core pass which ran at about 68-72c throughout. After it stopped the temp dropped back to ‘normal’.

So I ran a single core pass, same temp, but after it had stopped the temp stayed at 70c for a long while after (until I restarted).

I can add dashboard items in iCue under system information tab, but it’s only recognising the mobo, gpu, and ram. A few others seem to have the same issue but no solution - no CPU package, just a whole long list of ‘AMD K19’. Could just be my board?

 

[dwwaino]

If other programs can pick up the CPU then it's nothing to do with your board as to why iCue can't. Strange that it can't, I'll try and look into it over the weekend if you don't work it out. Worst case and I get a chance I'll setup my old system and have a look.

Odd that the temp didn't come down after single core runs but did for multicore. I'd have expected that more for multicore. That it came down after you restarted maybe there is some quirk with the monitoring software. What were you using as you can see CPU temp on iCue? The max temps are perfect so I don't know why it wasn't dropping. Just try some regular use tonight, games etc and just watch it.

It probably won't help but it definitely won't hurt, but also update your chipset driver if you haven't already. Just go to AMD website or search for AMD chipset drivers and get the latest x570 drivers.

 

[sataris]

iCue is bloated and buggy regardless.

Use HWInfo and see what that says

 

[dwwaino]

iCue is bloated and buggy regardless.

Use HWInfo and see what that says

Not as bad as it used to be. I used to hunt down the older versions of Cue and Corsair Link but sometime last year I got some new Corsair gear and had to get the latest iCue. It wasn't too bad, it's just probably the least friendly with other monitoring software out of all of those sorts of apps.

 

[sataris]

Not as bad as it used to be. I used to hunt down the older versions of Cue and Corsair Link but sometime last year I got some new Corsair gear and had to get the latest iCue. It wasn't too bad, it's just probably the least friendly with other monitoring software out of all of those sorts of apps.

It's basically malware.

 

[Hate]

If other programs can pick up the CPU then it's nothing to do with your board as to why iCue can't. Strange that it can't, I'll try and look into it over the weekend if you don't work it out. Worst case and I get a chance I'll setup my old system and have a look.

Odd that the temp didn't come down after single core runs but did for multicore. I'd have expected that more for multicore. That it came down after you restarted maybe there is some quirk with the monitoring software. What were you using as you can see CPU temp on iCue? The max temps are perfect so I don't know why it wasn't dropping. Just try some regular use tonight, games etc and just watch it.

It probably won't help but it definitely won't hurt, but also update your chipset driver if you haven't already. Just go to AMD website or search for AMD chipset drivers and get the latest x570 drivers.

I have had quite the eventful time keeping an eye on things today.

Searched around for iCue not displaying CPU temps. As you have mentioned with software conflicting, I deleted all monitoring software and I read to kill the iCue processes/services and restart. I did so, while I can not view it as its own package tab/block on the dashboard like I have seen in other images, and I may have been blind here, but I believe the Temp 3 under the motherboard tab on the dashboard is the CPU temp.

I was reading about power plan settings last night and noticed I had changed mine to high performance, so put it back to Balanced (and far right on the slider) (Windows) although some seem to suggest that there should/could/might be an AMD Balanced profile. I still had it idle high after running Cinebench at one point, and saw in HWiNFO that two or so cores were being full used. It could be a coincidence, but I killed the G.Skill RAM software (annoying if I can't use it as it reset to rainbow RGB lol) and it went back to normal. I read their forum and there are a few posts about it using high CPU levels, not sure if exactly related or not. I've ran Cinebench since with it being enabled and disabled and have not seen high idles, so I am really unsure still. I do still think that the block was not mounted correctly the first time but is now.

Got 3DMark on sale so ran that a few times and got excellent with the temperature sitting around the low 60s, aside from the end when it would ramp up to 70s, but went back down. Also installed chipset from AMD, which I am sure I had done but downloaded and installed again in any event.

It seems ok, but I don't know. I'll let it idle for a while now and keep an eye on it via HWiNFO and if it is ok, I might just cable manage and call it a day and pray I never see such high idles again.

 

[drd23]

HWInfo has a logging feature fwiw. It spits out an excel spreadsheet with a lot of numbers for every sensor it has, but it can be useful for looking at temps over a period of time instead of just occasionally pulling it up and looking at the numbers

 

[dwwaino]

I have had quite the eventful time keeping an eye on things today.

Searched around for iCue not displaying CPU temps. As you have mentioned with software conflicting, I deleted all monitoring software and I read to kill the iCue processes/services and restart. I did so, while I can not view it as its own package tab/block on the dashboard like I have seen in other images, and I may have been blind here, but I believe the Temp 3 under the motherboard tab on the dashboard is the CPU temp.

I was reading about power plan settings last night and noticed I had changed mine to high performance, so put it back to Balanced (and far right on the slider) (Windows) although some seem to suggest that there should/could/might be an AMD Balanced profile. I still had it idle high after running Cinebench at one point, and saw in HWiNFO that two or so cores were being full used. It could be a coincidence, but I killed the G.Skill RAM software (annoying if I can't use it as it reset to rainbow RGB lol) and it went back to normal. I read their forum and there are a few posts about it using high CPU levels, not sure if exactly related or not. I've ran Cinebench since with it being enabled and disabled and have not seen high idles, so I am really unsure still. I do still think that the block was not mounted correctly the first time but is now.

Got 3DMark on sale so ran that a few times and got excellent with the temperature sitting around the low 60s, aside from the end when it would ramp up to 70s, but went back down. Also installed chipset from AMD, which I am sure I had done but downloaded and installed again in any event.

It seems ok, but I don't know. I'll let it idle for a while now and keep an eye on it via HWiNFO and if it is ok, I might just cable manage and call it a day and pray I never see such high idles again.

For Ryzen 5000 there is no longer a need for the AMD power profile so it was removed. Normal balanced is fine. High performance will definitely lift temps if that was enabled. Also don't use that G.Skill software, it's known to brick memory. G.Skill uses Aura Sync so use something like Asus Aura Creator instead. They have a full package called Armory Crate but it's really bloaty and if you just want to set and forget your RAM's RGB then Aura Creator is a better option.

Sounds like everything else is starting to work out which is good. After all you want to enjoy your PC, not constantly troubleshoot it lol.

 

[Hate]

HWInfo has a logging feature fwiw. It spits out an excel spreadsheet with a lot of numbers for every sensor it has, but it can be useful for looking at temps over a period of time instead of just occasionally pulling it up and looking at the numbers

Oh, I appreciate this tip - thanks a heap. I have just started logging (hopefully correctly) and will see what it shows later.

For Ryzen 5000 there is no longer a need for the AMD power profile so it was removed. Normal balanced is fine. High performance will definitely lift temps if that was enabled. Also don't use that G.Skill software, it's known to brick memory. G.Skill uses Aura Sync so use something like Asus Aura Creator instead. They have a full package called Armory Crate but it's really bloaty and if you just want to set and forget your RAM's RGB then Aura Creator is a better option.

Sounds like everything else is starting to work out which is good. After all you want to enjoy your PC, not constantly troubleshoot it lol.

I just want to set and forget the LEDs so I will give your suggestion a shot.

I hope so too. Been hanging out to play games. Any game. At reasonable quality (which will be 1080p on a 144hz monitor).

 

[dwwaino]

Oh, I appreciate this tip - thanks a heap. I have just started logging (hopefully correctly) and will see what it shows later.


I just want to set and forget the LEDs so I will give your suggestion a shot.

I hope so too. Been hanging out to play games. Any game. At reasonable quality (which will be 1080p on a 144hz monitor).

What's your graphics card out of interest? Probably not something you will want to stuff around with now, but have you tuned up your RAM as well since Ryzen is pretty sensitive to memory speeds?

The 5900x is an awesome CPU too, I think you'll really like it. I haven't been excited over a CPU since the 2500k. My last system had a 8700k, I had some issues with it over the time and while it did the job for gaming it just wasn't a very exciting part.

Out of the box the 5900x has been a delight. At stock settings it completely wipes the floor with the 8700k and can actually hit the advertised single core boost clock. I never saw the 4.7ghz turbo on the 8700k, but I regularly see 4.9ghz boost on the 5900x. I've actually seen it nearly hit 5ghz pretty regularly so that's even better. Not to mention all the cores to play with. Can't wait to get back into Unreal and Unity and see how it goes building lighting and compiling code.

 

[Hate]

What's your graphics card out of interest? Probably not something you will want to stuff around with now, but have you tuned up your RAM as well since Ryzen is pretty sensitive to memory speeds?

The 5900x is an awesome CPU too, I think you'll really like it. I haven't been excited over a CPU since the 2500k. My last system had a 8700k, I had some issues with it over the time and while it did the job for gaming it just wasn't a very exciting part.

Out of the box the 5900x has been a delight. At stock settings it completely wipes the floor with the 8700k and can actually hit the advertised single core boost clock. I never saw the 4.7ghz turbo on the 8700k, but I regularly see 4.9ghz boost on the 5900x. I've actually seen it nearly hit 5ghz pretty regularly so that's even better. Not to mention all the cores to play with. Can't wait to get back into Unreal and Unity and see how it goes building lighting and compiling code.

I was after a 3080 but figured I don’t ~ really ~ need it and could save myself a few bucks. I could have probably gone a 5600X too but I had the chance to buy, to my surprise given stock issues, a 5900x on the provision it was purchased with an x570. I always wanted a white build and then had the chance to snare the 3070 VISION OC. If I can find one in the new year I might upgrade to a 3080.

Only thing I changed was the XMP profile on the RAM so it runs 3600.

 

[dwwaino]

I was after a 3080 but figured I don’t ~ really ~ need it and could save myself a few bucks. I could have probably gone a 5600X too but I had the chance to buy, to my surprise given stock issues, a 5900x on the provision it was purchased with an x570. I always wanted a white build and then had the chance to snare the 3070 VISION OC. If I can find one in the new year I might upgrade to a 3080.

Only thing I changed was the XMP profile on the RAM so it runs 3600.

Those Gigabyte Vision cards are really nice, I've seen some nice white builds with them. They're one of the most common I've seen sniped as well. If you look up bapcsalesaustralia on Reddit it's a really good Australian building group and there are threads dedicated to availability of parts and stuff. Seems like the Gigabyte are the most commonly grabbed 3080 cards.

3600mhz still seems to be the sweet spot for Ryzen 5000 so you're set. If you find yourself back in the BIOS one day have a look for the FCLK setting. It will be set to auto but just change it to 1800mhz (which is half your memory speed). I don't know if there is a difference between auto or manually setting it, I didn't notice one, but I just changed it anyway just to be 100% the fabric and memory were 1:1.