5820k temps under AIO?

I have a 5820k @ 4.2GHz using 1.25v. Other settings on auto.
I use an H100i with push / pull Akasa Pirahna fans on PWM in a Corsair 240 Air case.
It idles at around 38c and stress testing using Intel Burn test on very high settings hits 80c.
Temps seem a bit high to me?
I tried 4.4Ghz on 1.35v and temps hit 90c under stress testing.

Use Asus real bench to test with. Bit more realistic with load and therefore temps.
 
Joe so no regrets on the 5i20k ?

I originally wanted to go with 4790k (around 4-5 months ago now) but everyone told me to wait for Skylake, and since that was a pretty big disappointment I decided to spend the extra money I had saved on the 5820k. For a similar price I see no real reason to not take the extra 2 cores/4 threads for -300MHz on a 6700k.

I am happy, yes. I have a 6 core processor @ 4.5GHz, a chipset that is getting an upgrade in the near future and if need be, I can pop an 18 core Xeon in it - I think I made a good decision.
 
Joe, my 4.6 overclock would run Aida and Realbench no problems but wouldn't run OCCT for a minute before blue screening!! Run OCCT for 5 mins with 64bit Large Data Set and if it passes that too then it'll be fine :cool:

(since knocked mine down to 4.5 1.19v and it's solid, not bothered about chasing 24/7 numbers. Prefer a cooler rig)
 
I can't install .NET 3.5 apparently.

auhEP3J.png


I guess there's no stress testing for me. :eek:

Edit: Never mind, following this worked if anyone else runs into the same issue: http://winaero.com/blog/offline-install-of-net-framework-3-5-in-windows-10-using-dism/ :)
 
Last edited:
Hi, one more thing...

What should the gap between my highest and lowest core be? At the moment it's roughly 14-18c.
 
Okay I'll go and repaste it :p

Thanks!

Edit: The same core (core 5) is now 8-10 degrees different, every other core is roughly the same give or take a few degrees.

I guess I'll let that core be a special *********.

When idle it's around 3-4 lower than the rest.
 
Last edited:
I was using Arctic Silver 5 that I got with the paste cleaning kit a couple of years back, using the line method.
 
1.2v for 4.4ghz (125mhz x 35) seems to be about the usual for a stable 5820k running 1344k in-place ffts in prime, temps generally between 50°C to mid 60's.
 
I've decided to try and replace my H100i with a custom water loop for my 5820K in my Corsair Air 240 case.
I have 2 radiators to try, an XSPC RX240 and if I can't get that to fit, an EX240. This with an EK Supremacy Evo block and EK X-Res Revo res with D5 Vario pump. Res and pump to be placed in the back compartment in place of the SSD cage. I think this is going to be challenging to say the least....
 
I didn't think changing the CPU strap from 100 made any real difference? I'm new to OC-ing these Intel Haswell-E chips, and have simply set vcore to 1.1 and the multi to 41 as an initial overclock.
I used the Intel Extreme Tuning Utility (XTU) to stress test and it seems to do a good job, and is presumably safe as it's from Intel? Is AIDA better?
All is stable anyway with load temps in the high 50's cooled by my Noctua NH-D15S, so I reckon I'm good for 4.5GHz with luck.
 
Back
Top Bottom