What are your idle and full load CPU temps?

[Tech_fanatic]

I know that idle temps depend on a lot of of variables like ambient temperature, cpu voltage offset, cpu cooler, case airflow, cpu thermal compound etc.
My Ryzen 9 7900x idles at 50°C at 21°C ambient with NZXT Kraken X63 mounted on it inside a Lian Li O11 dynamic EVO.

 

[Sithaer]

Well since I only have a 12100F its not hard to keep it cool even with a silent ~30$ tower cooler. 'ID-Cooling SE-224-XT ARGB V3 in my case'
Even tho my case doesn't have the best airflow for CPU coolers, only bottom intake and thats mainly blocked by my GPU so ye.

Idle is around 30-32 Celsius and during gaming its between 50-55 in most games, CPU heavy games like Cyberpunk can push it close to 60 over time.
Full load like Cinebench R23 multi looping is about the same. 'looped it for 10 mins now and it stayed at 60 max'

Currently its winter where I live but even during summer the temps barely went over 60 while gaming.

 

[xtreemchaos]

ive gone for the 2nd one, custom water loop cpu and gpu and 2x 480 rads.

 

[Shihab]

I voted third option, though the CPU does hit 90s under extreme loads. Room ambient is 33c right now, and it's a winter evening. Can easily hit mid 40s if I didn't have an air cooler on during summers.

I miss my h50. Never saw anything reaching over mid 60s, even after 48hours of continuous video encoding.

 

[ThrashZone]

Hi,
Option 2 for idle 30c to be exact and option 4 max load 85c to be exact
Other option numbers are pretty out of place or used because of really shitty coolers or contact idling at 40-70c is just crazy talk

So "I don't bother" on the poll because it's unrealistic

 

[INSTG8R]

I can’t vote, your differentials don’t match. 35C idle 75C max load under custom water 22C Ambient

 

[kilo]

Option 2.

Because my parts are so downclocked they use less power, therefore less heat. It was so hot this summer for so long by me, I put my 2070s to 60% power (125w avg). It got annoying sweating all over my mouse and keyboard. The fps per watt is good enough to justify keeping it that way indefinitely.

 

[freeagent]

Idles cool, loads hot

Between winter providing cool air, and the furnace to mess it all up... every day is different

 

[Bill_Bright]

So "I don't bother" on the poll because it's unrealistic

I can’t vote, your differentials don’t match. 35C idle 75C max load under custom water 22C Ambient

I agree. Idle and max together just doesn't make sense. IMO, having a poll for this doesn't make sense. Simply stating actual temps (rather than a range) makes more sense.

And freeagent is correct too - there are just too many variables influencing the temps. Even time of day for me since the sun shines through my (East facing) computer room windows in the morning, but not in the afternoon.

And of course, even with identical ambient and operational environments, every model CPU is different too.

So, I very much bother paying attention to my temps. But I am not going to bother participating in the poll since (1) no option fits my scenario and (2) it really would not be informative to anyone, but me.

 

[INSTG8R]

I agree. Idle and max together just doesn't make sense. IMO, having a poll for this doesn't make sense. Simply stating actual temps (rather than a range) makes more sense.

And freeagent is correct too - there are just too many variables influencing the temps. Even time of day for me since the sun shines through my (East facing) computer room windows in the morning, but not in the afternoon.

And of course, even with identical ambient and operational environments, every model CPU is different too.

So, I very much bother paying attention to my temps. But I am not going to bother participating in the poll since (1) no option fits my scenario and (2) it really would not be informative to anyone, but me.

Well said I just know my temps are “safe” but I don‘t get any direct sun, but the current season means I use my PC to heat the room it can easily get like 16-18C in there in winter being a small room with the door closed when I’m not in there. Being a small room my PC at least in previous builds will easily warm the room to a more comfortable 20-21 with a decent gaming session.

 

[Deleted member 185158]

I agree. Idle and max together just doesn't make sense. IMO, having a poll for this doesn't make sense. Simply stating actual temps (rather than a range) makes more sense.

And freeagent is correct too - there are just too many variables influencing the temps. Even time of day for me since the sun shines through my (East facing) computer room windows in the morning, but not in the afternoon.

And of course, even with identical ambient and operational environments, every model CPU is different too.

So, I very much bother paying attention to my temps. But I am not going to bother participating in the poll since (1) no option fits my scenario and (2) it really would not be informative to anyone, but me.

I can understand the lack of participation on this one.....

And I thought asking what your gear was worth is silly......

But my winter temps are much nicer than summer temps, that's for sure!

 

[Bill_Bright]

I use my PC to heat the room

LOL I hear you on this. My problem is the HVAC thermostat is just on the other side of a common interior (uninsulated) wall from the computer room. So in the winter, after a couple hours using the computer, heat from the computer room tells the thermostat the house is warm and cozy. But its freezing!

And in the summer, the heat from the computer room tells the thermostat the house is burning up - so it signals the AC to kick on, and stay on, freezing out the rest of the house!

My next house will have zone heating and cooling, with the computer room its own zone.

 

[Tech_fanatic]

Option 2.

Because my parts are so downclocked they use less power, therefore less heat. It was so hot this summer for so long by me, I put my 2070s to 60% power (125w avg). It got annoying sweating all over my mouse and keyboard. The fps per watt is good enough to justify keeping it that way indefinitely.

I too have 7900x undervolted to the limit just to keep it under idle as it would easily cross 52 degrees (at stock voltages) on idle on a hot summer day!

 

[INSTG8R]

LOL I hear you on this. My problem is the HVAC thermostat is just on the other side of a common interior (uninsulated) wall from the computer room. So in the winter, after a couple hours using the computer, heat from the computer room tells the thermostat the house is warm and cozy. But its freezing!

And in the summer, the heat from the computer room tells the thermostat the house is burning up - so it signals the AC to kick on, and stay on, freezing out the rest of the house!

My next house will have zone heating and cooling, with the computer room its own zone.

I have a heatpump in my living room/kitchen that is all I use the bedroom and PC room are opposite end down a hallway neither require heating The entire house has heated floors but I would never use them as they would just make my meter spin constantly...

 

[QuietBob]

A CPU temperature poll is an interesting idea, but given how many factors affect it, comparing different people's results is not much useful -- even with the same SKU.

For example, what kind of "full load" are we talking about here? Gaming load is pretty irrelevant, as it will rarely max out the CPU. With the dynamic nature of games, you won't ever get the same result unless testing the same specific scene under the same strict conditions. And even then the temperature may vary run to run.

OTOH, quoting CPU temperature in something like Linpack Xtreme is equally useless. Linpack may be helpful in finding the limits of your cooling solution, or verifying an overclock, but it does not resemble any kind of practical load. As such, it's irrelevant to 99.99% of users.

And of course CPU temps will vary with your ambients, type of cooling, case model and airflow, fan curve, the number of apps running in the background, power plan settings, whether the CPU is at stock or not, etc.

At the very least, we should specify the room temperature when the measurement was taken, the cooling solution used and CPU settings. For load temps, the software used to test them must be named.

FWIW, in a 25c room, my 5800X3D with -30 CO on all cores idles at 31-34c with a Deepcool Assassin III using stock fans. This was measured in HWinfo with a 1000ms polling interval, with Snapshot CPU Polling setting enabled. A third-party internet security app and MSI Afterburner polling every 1000ms were running in the background. Balanced power plan with power slider set to best energy savings was used.

In Cinebench R23 I top out at 76c with all fans running at maximum, under the same conditions.

 

[AusWolf]

I know that idle temps depend on a lot of of variables like ambient temperature, cpu voltage offset, cpu cooler, case airflow, cpu thermal compound etc.
My Ryzen 9 7900x idles at 50°C at 21°C ambient with NZXT Kraken X63 mounted on it inside a Lian Li O11 dynamic EVO.

Something must be wrong with your power saving settings. My 7700X idles at 35-36 °C with a be quiet! Silent Loop 2 280 mounted as intake in a Kolink Citadel Mesh.

 

[freeagent]

Mine?

Just kidding. I cracked a window while I cleaned the litter boxes

 

[DemonicRyzen666]

23.5c idle
max load in what ? Y-cruncher 84c-88c / prime95, small FFT's 82c, Lare FFT's 59c, Blend 65C/ coredamage 83c Intel burntest 72c /CinebenchR23 72c

 

[Tech_fanatic]

Something must be wrong with your power saving settings. My 7700X idles at 35-36 °C with a be quiet! Silent Loop 2 280 mounted as intake in a Kolink Citadel Mesh.

I have tried reseating the heatsink with arctic mx-5 paste but the temperatures during Summers are insane here ( 45 degrees + even in a hot October day). It's also the case with my laptop which frequently thermal throttles. But I will reseat the pump once again when I get the time.
My case airflow is fine btw.

 

[amit_talkin]

My Ryzen 3600 stays at around 45c idle and goes up to 83c on custom water loop and that with thermal grizzly conductonaut.

 

[Tech_fanatic]

A CPU temperature poll is an interesting idea, but given how many factors affect it, comparing different people's results is not much useful -- even with the same SKU.

For example, what kind of "full load" are we talking about here? Gaming load is pretty irrelevant, as it will rarely max out the CPU. With the dynamic nature of games, you won't ever get the same result unless testing the same specific scene under the same strict conditions. And even then the temperature may vary run to run.

OTOH, quoting CPU temperature in something like Linpack Xtreme is equally useless. Linpack may be helpful in finding the limits of your cooling solution, or verifying an overclock, but it does not resemble any kind of practical load. As such, it's irrelevant to 99.99% of users.

And of course CPU temps will vary with your ambients, type of cooling, case model and airflow, fan curve, the number of apps running in the background, power plan settings, whether the CPU is at stock or not, etc.

At the very least, we should specify the room temperature when the measurement was taken, the cooling solution used and CPU settings. For load temps, the software used to test them must be named.

FWIW, in a 25c room, my 5800X3D with -30 CO on all cores idles at 31-34c with a Deepcool Assassin III using stock fans. This was measured in HWinfo with a 1000ms polling interval, with Snapshot CPU Polling setting enabled. A third-party internet security app and MSI Afterburner polling every 1000ms were running in the background. Balanced power plan with power slider set to best energy savings was used.

In Cinebench R23 I top out at 76c with all fans running at maximum, under the same conditions.

Yes you are right and I knew that there are various factors in play here but that's not the point here. It's just to know what cpu temperatures people are ready to tolerate and if they like to stay under or go over the so called Tjmax (Max junction temp) of their silicon. I am personally obsessed with temp and power limits of these modem chips.
Thanks for sharing your opinion btw.

 

[AusWolf]

I have tried reseating the heatsink with arctic mx-5 paste but the temperatures during Summers are insane here ( 45 degrees + even in a hot October day). It's also the case with my laptop which frequently thermal throttles. But I will reseat the pump once again when I get the time.
My case airflow is fine btw.

Do you have C-states enabled in the BIOS? What are your windows power saving settings? What are your idle clocks and power usage?

 

[Synthwave]

Idle is 52-54 °C for an AMD Ryzen 7 2700 @ 3.5 GHz / 1,875 V | SMT on | SoC 0,9 V, but it never really gets above 65 °C, and haven't done excessive full load testing. It has a Noctua NH-D9L on it with HT-01 thermal paste.

A Noctua NF-A14 PWM fan in the front, a Noctua NF-S12B Redux-1200PWM fan in the back, and every fan is doing 300-450 rpm. The case is a Fractal Design Define Nano S (w/o glass side), so it's not geared towards awesome temps.

Mind you, most of the time I know the PC is running, beacuse I can see that the power LED is active.

 

[TheoneandonlyMrK]

Max 85°C most of the day, GPU at 45°c most of the day fully loaded but GPU down tuned with custom 2x360 rad.

 

[sLowEnd]

FWIW, my R5 3600 is currently loading at about 72°C doing F@H in a room that's about 20°C ambient. Cooling is a Thermalright Silver Arrow T8.
It could be running cooler, but I sleep in the same room as my computer, so the fan isn't set to go super fast - just around 1200RPM.

Idle temperatures aren't something I pay close attention to, but iirc it's usually less than 45°C around the aforementioned ambient temperature.