9900k temps

[Articx420]

I idle at 25-32 c
Cinebench r20 temps max out to the 70-78 [Score of 4902] and Multi core enhancement is on. No other settings on other than xmp.  

 

I'm using the corsair 150i Pro with arctic silver.

However in Prime95 using the Small FFT s tests [Max power] I get 87-98 c on the cores.

I was wondering if this specific test is known for taking these chips to their thermal limits, as afterall its such a unrealistic load.
I'm just curious if this is normal or not.

[5x5]

It's normal, the 9900K is a space heater and with MCE pulls about 150-180 Watts under full load. 

[Turtle Rig]

41 minutes ago, 5x5 said:

It's normal, the 9900K is a space heater and with MCE pulls about 150-180 Watts under full load. 

He is close to TJ Max which is 100c but don't quote me on that.  Firstly do not use Prime95 as its a synthetic benchmark that taxes your CPU in a way that HT apps or games don't do.  Your temps are about 12c higher then running prime95 compared to Aida64 stability test.  Prime95 is junk and basically abuse on the CPU for no reason.

[5x5]

18 minutes ago, Turtle Rig said:

He is close to TJ Max which is 100c but don't quote me on that.  Firstly do not use Prime95 as its a synthetic benchmark that taxes your CPU in a way that HT apps or games don't do.  Your temps are about 12c higher then running prime95 compared to Aida64 stability test.  Prime95 is junk and basically abuse on the CPU for no reason.

You are right on the first part - 100*C is Tj Max.

 

You are wrong on the second one, though - unlike Furmark, Prime95 is the ultimate stress test for overclocking - it's not good for testing thermals on stock CPUs but crucial when testing overclocks - I've had unstable overclocks pass 6 hours Aida64 only to randomly BSOD in a game. Running Prime95 and the OC crashes in 20 mins or so.

[Turtle Rig]

9 minutes ago, 5x5 said:

You are right on the first part - 100*C is Tj Max.

 

You are wrong on the second one, though - unlike Furmark, Prime95 is the ultimate stress test for overclocking - it's not good for testing thermals on stock CPUs but crucial when testing overclocks - I've had unstable overclocks pass 6 hours Aida64 only to randomly BSOD in a game. Running Prime95 and the OC crashes in 20 mins or so.

Oh I believe you it is top of the line at taxing you are correct 100 percent.  My thing is it is a synthetic benchmark that basically abuses the CPU.  Aida64 stability test is about 12c lower in temps because it is not synthetic.  All Im saying is you will never tax your CPU the way Prime95 does it.  You want to know if system is stable run Aida stability test.  You want to torture your CPU with synthetic test then run Prime95, but I can tell you for example if you render a video its not gonna go in the 90's c  .. It will be about 80c.  We want real world, not synthetic, so whatever temp you get with Prime95 you can subtract  about 10c to 12c in the real world, using Premiere or PS or CAD or 3D modeling your temps will not hit 90's c.  Yes Prime95 is legit way to tax your CPU to death.  But means nothing compared to real world loads and temps and what not.  I agree with you 5x5, just trying to make a simple point.

[5x5]

1 minute ago, Turtle Rig said:

Oh I believe you it is top of the line at taxing you are correct 100 percent.  My thing is it is a synthetic benchmark that basically abuses the CPU.  Aida64 stability test is about 12c lower in temps because it is not synthetic.  All Im saying is you will never tax your CPU the way Prime95 does it.  You want to know if system is stable run Aida stability test.  You want to torture your CPU with synthetic test then run Prime95, but I can tell you for example if you render a video its not gonna go in the 90's c  .. It will be about 80c.  We want real world, not synthetic, so whatever temp you get with Prime95 you can subtract  about 10c to 12c in the real world, using Premiere or PS or CAD or 3D modeling your temps will not hit 90's c.  Yes Prime95 is legit way to tax your CPU to death.  But means nothing compared to real world loads and temps and what not.

Again, you aren't correct - Prime95 is not magic - it uses the same type of loads you find in many applications like games and video editing software - Aida64 is much simpler and thus less intensive. Prime 95 puts much heavier crossload exposing weak stability - that's why it's far and away vital when testing for stability. There are plenty of detailed guides explaining this in detail and why it's important. Feel free to google.

[Turtle Rig]

Just now, 5x5 said:

Again, you aren't correct - Prime95 is not magic - it uses the same type of loads you find in many applications like games and video editing software - Aida64 is much simpler and thus less intensive. Prime 95 puts much heavier crossload exposing weak stability - that's why it's far and away vital when testing for stability. There are plenty of detailed guides explaining this in detail and why it's important. Feel free to google.

I agree with you there my friend.  I guess I'm trying to make a small point ya know.

[5x5]

1 minute ago, Turtle Rig said:

I agree with you there my friend.  I guess I'm trying to make a small point ya know.

You're talking about thermals - and I understand your point - but Prime is not a thermal test, it's a stability test.

 

Also - the 9900K is hitting 90*C under a 360mm AIO - that's worse than the 9590 from old so I can safely describe it as a space heater and oven

[Turtle Rig]

7 minutes ago, 5x5 said:

You're talking about thermals - and I understand your point - but Prime is not a thermal test, it's a stability test.

 

Also - the 9900K is hitting 90*C under a 360mm AIO - that's worse than the 9590 from old so I can safely describe it as a space heater and oven

Your smart 5x5.  Thanks for not yelling at me or getting upset.  You and ton of users are cool peeps, but there are always a few that will bash, and you did not bash me.  That tells me your older man.