Question Does high CPU temperature = shorter lifespan ?

[SleuthX]

About to install a 13700k with a triple fan 7 heat piper cooler(Cougar forza 135) and I'm getting some info that my CPU will be averaging 75c-80c if not more during gaming so I need to know from a real expert, I'm getting this PC and keeping it for a good 7-10 years if it allows so how's that looking for the overall lifespan of my CPU?

 

[Lutfij]

Please list the specs to your build like so:
CPU:
CPU cooler:
Motherboard:
Ram:
SSD/HDD:
GPU:
PSU:
Chassis:
number of fans in your build and their orientation. Ambient room air temps?

 

[SleuthX]

Please list the specs to your build like so:
CPU:
CPU cooler:
Motherboard:
Ram:
SSD/HDD:
GPU:
PSU:
Chassis:
number of fans in your build and their orientation. Ambient room air temps?

Lol how is "SSD/HDD & Motherboard" related to CPU temps? I'm gonna assume it's a CTRL+C and go on. CPU cooler is above already, case is a Bitfenix Nova Mesh TG with 1 exhaust, 3 front and 2 top. Ambient room temp is 30c-36c in the summer, 27c-30c in the spring and under 24c in the winter.

 

[Misgar]

If you buy a really cheap PSU with a 3-year warranty (or less), it probably won't last 7 years, let alone 10. That's why we need to know the spec of your main components.

Get a Tier A High-End PSU from this list for longevity.
https://cultists.network/140/psu-tier-list/

80C during gaming is nothing to worry about. I'm running my 7950X at 90C+ for 36 hours continuously during video rendering sessions.

 

[TerryLaze]

I'm getting some info that my CPU will be averaging 75c-80c if not more during gaming

That would mean that it would be overclocked, most mobos overclock by default and you have to change it manually if you don't want it.
These temps, 75-80, aren't bad by themselves, shut off temp for intel is 130 so there is a lot of headroom.
The issue is with the Volts that the mobo will be pushing through the CPU, that will be the thing that could possibly degrade your CPU.

 

[geofelt]

Intel processors monitor their own temperatures.
If a core gets hot enough to cause damage, it will slow down or even shut off.
That temperature is 100c.
The situation is even less ominous for a gamer where only a few processing threads will even approach 100c.
There is not just one temperature, there are many, one for each thread.
A few threads hitting 100c. is not a big deal.
High voltages can damage a cpu, but if the voltage is high enough to cause a 100c. temperature, the chip itself will throttle.
One Intel engineer said "if you do not reach 100c. you are leaving performance on the table."

The forza 135 looks to be a decent cooler.
Feed it with a couple of good front intake fans and you will be fine.

 

[Karadjgne]

Cpus have changed, evolved, but the thinking about them for many ppl hasn't.

Back in the day, there weren't any consumer grade cpus with high core counts, you got 2 or 4 cores and that was basically it. So turbo or boost was relatively simple to deal with, you set a voltage and a speed and temps followed suit.

Today's cpus and how the boost works are different. There's not really a set turbo, there's a cpu figured turbo. So if you get 5.2GHz and 75°, the cpu will push it, 5.3GHz or 5.4GHz on some cores, just because it has decided that it's not running over 90°, so therefore it can.

So getting info that your cpu should be averaging 75°-80° is useless. The best thing you can do is provide @ 1.5x to 2x power use worth of cooler capacity. Since the 13700k when pushed to it's limits can hit @ 250w, you should be looking at 360mm or 420mm AIO's unless as said "you'll be leaving performance on the table" because temps will limit boost.

Now you can most definitely run with a NH-D15 or similar 250w+ aircooler and get the same range of temps for less money, but the difference between the two cooling systems will be 100-300 (ish) MHz on multiple cores as the cpu self temp limitations put the brakes on higher turbo boost speeds when using multiple or high core counts.

 

[TerryLaze]

So getting info that your cpu should be averaging 75°-80° is useless. The best thing you can do is provide @ 1.5x to 2x power use worth of cooler capacity. Since the 13700k when pushed to it's limits can hit @ 250w, you should be looking at 360mm or 420mm AIO's unless as said "you'll be leaving performance on the table" because temps will limit boost.

Now you can most definitely run with a NH-D15 or similar 250w+ aircooler and get the same range of temps for less money, but the difference between the two cooling systems will be 100-300 (ish) MHz on multiple cores as the cpu self temp limitations put the brakes on higher turbo boost speeds when using multiple or high core counts.

While this is valid for people that want to overclock as high as it goes, we should also look at how much, or how little rather, this difference is.
Especially for gaming you can go very low on power and not lose enough frames for you to even notice anything.
This is for the 13900k but the same principles apply.

View: https://www.youtube.com/watch?v=H4Bm0Wr6OEQ

 

[SleuthX]

While this is valid for people that want to overclock as high as it goes, we should also look at how much, or how little rather, this difference is.
Especially for gaming you can go very low on power and not lose enough frames for you to even notice anything.
This is for the 13900k but the same principles apply.

View: https://www.youtube.com/watch?v=H4Bm0Wr6OEQ

Reading this has been useful. So the only thing I should worry about is the voltage, huh? are there official guides for undervolting the 13700k without losing any performance?

 

[TerryLaze]

Reading this has been useful. So the only thing I should worry about is the voltage, huh? are there official guides for undervolting the 13700k without losing any performance?

You don't have to go crazy, you don't need to undervolt.
Just look at your bios settings when the system is ready and, if you want to, share the settings here to hear opinions on if there is something to change.
In general the only thing to be mindful of is, if the mobo overclocks too hard to step it back a little.
Since everything is automated if you limit the power everything else will drop accordingly. If default of your mobo is unlimited power then you can turn that back to official power draw and go from there.
If temps are fine and vcore is fine you can go higher if not you can go lower on power.

 

[geofelt]

The danger of fiddling with voltages is that you may not understand the implications and cause damage.
While defaults may not be bad, at least they are not likely to cause damage.

Current processors are so darn powerful that getting the last ounce out of them is not likely to be worth the effort.

If you are a hobbyist/experimenter and not a user, go ahead and fiddle away.

 

[Karadjgne]

While this is valid for people that want to overclock as high as it goes, we should also look at how much, or how little rather, this difference is.
Especially for gaming you can go very low on power and not lose enough frames for you to even notice anything.
This is for the 13900k but the same principles apply.

View: https://www.youtube.com/watch?v=H4Bm0Wr6OEQ

Yes and no. It affects multi core apps a lot more than it affects games which use a relatively tame amount of P-cores. So if doing stuff like rendering or compiling which can easily use up every core, having multiple cores at (for instance the E-cores) at a considerably higher frequency because of cooling capacity, is going to be a time benefit. There are also many games, like Skyrim or Fall/Out that use an insanely high amount of papyrus scripted mods, enbs etc that can punish P-cores with AVX/AVX2 use further degrading boosts if there isn't cooling capacity to handle those loads.

You cannot over-cool a cpu. Ever. Not with standard cooling. There does come a point of diminishing returns and sheer pointlessness to some cooling, but having too much capacity can never hurt, has no drawbacks or downsides. Not unless you consider having pumps and fans that have zero need to ever run at full speed/volume a drawback.

In simple terms, the more active cooling capacity you have, the closer to passive cooling you get.

 

[SleuthX]

The danger of fiddling with voltages is that you may not understand the implications and cause damage.
While defaults may not be bad, at least they are not likely to cause damage.

Current processors are so darn powerful that getting the last ounce out of them is not likely to be worth the effort.

If you are a hobbyist/experimenter and not a user, go ahead and fiddle away.

I said undervolt, not overvolt

 

[geofelt]

I said undervolt, not overvolt

I know. but making a mistake can happen when you fiddle with voltages.

 

[Karadjgne]

I said undervolt, not overvolt

Cpus require Power. When you drop voltages, it's entirely possible, even probable, that amperage use goes up, raising temps in the VRM's which can lead to instability. There are also voltages that rely on the voltage range of other settings, so if you lower one voltage slightly, it can automatically lower other voltages without your knowledge or input.

The problem there is that there's generally a range of acceptable voltages. For instance if you lower one voltage from 1.1v to 1.0v, there can be another voltage that requires it be at 0.1v below, so automatically switches to 0.9v. But. It's range of operation is 0.8v-1.1v nominally, so even if you put the first voltage back to 1.1v, the second auto voltage remains at 0.9v as that's an acceptable voltage setting. And then you can't figure out why the cpu now won't boost to the same GHz as before.

Undervolting isn't as simple as just dropping vcore, it's just as complex as overvolting, undervolting is essentially overclocking, attaining a particular speed for a lower voltage in order to reduce temps, regardless of what that actual speed is.