OC i5 3570k @ 4.1 GHz - thoughts?

neograndizer

Distinguished
Sep 10, 2010
179
0
18,710
Okay, I'm just wondering if my OC is fine as it is. I know with my current setup I cannot realistically push 4.2 GHz without a good spike in temps and vcore and not getting any real benefit from it. I settled with 4.1 GHz as 4.0 GHz yielded almost the same results in temps and voltage. Both required 1v < 1.1v vcore and temps around 65-73'C stress testing. I am using offset - 0.025v. 8GB RAM is 8-8-8-24 2T @ 1.45v.

My results from stress testing:

P95 blend still running close to 9 hours
IB41xOC20130212p95blend8hrs.jpg


P95 blend stopped at 9 hours
IB41xOC20130212p95blend9hrs.jpg


IBT 10 runs Max settings
IB41xOC20130212IBTMax10runs.jpg


Memtest86+ 10 passes with no errors.

I don't run my comp 24/7, so I think this is fairly good for stability (not rock solid stable, but stable enough).

Thoughts?
 

Adroid

Distinguished
Temps are mediocre, not outrageous for ivy. I agree they might be a touch high - but if your using a low end heatsink and/or high ambient temperature, that would explain it.
 

neograndizer

Distinguished
Sep 10, 2010
179
0
18,710
Thanks the replies.

I don't have a high-end cooler. It's a Scythe Mugen 2 Rev.B equipped with Noctua NT-H1 TIM and 2 x NF-F12 PWM fans in push/pull config. Much quieter than my previous Scythe Slipstream Kaze Jyuni 1900RPM fan. Unfortunately, it looks like getting a quieter machine results in slightly elevated temps. My one Scythe fan in push going full 1900RPM does almost the same as what the two Noctua fans are doing.

My ambient temps are around 26'C, but it is in a semi-enclosed space (in a open cabinet under my desk). Using a Raven RV02-E (limited edition white) case. Even with an open cabinet, I can only guess that it doesn't provide good air ciruculation as say an open room.

Tried to push vcore down to 0.9v, to hopefully help with temps, but proved unstable. With my current setup, I think this is the best I can push out of the system in terms of a reasonable OC.

Guess when I went IB, I should have considered getting a better cooler. LOL.
 

Mt Power

Distinguished
Jul 17, 2012
250
1
18,810
yeah that seems high! you may want to investigate your hsf install. while the ombient looks ok, the 73c for such low voltage is high.

as a point of reference, i have 4.7 ghz, 1.296v, and i get 88c on 8hrs prime. im using a 212 evo. you should be able to do better.

again check the install on the hsf.
 
I would say if you want to overclock further than get a better cooler, if you don't then what you have now is ok as your load temps aren't dangerous, also Prime95/Intel Burn test stresses the CPU more than almost any real world application so in day to day use I would expect your CPU temps to be significantly lower.
 

neograndizer

Distinguished
Sep 10, 2010
179
0
18,710
Thanks for the responses. Nothing wrong with my hsf install. I reseated it three times during the swapping of parts, etc. All showed the same results. Unless I botched the install three times in a row. :p

I don't plan going any more higher than this. I think I hit the 'happy' medium. I think I ended up with a hotter chip out of the lot. Maybe bad paste under the lid. Not planning to delid it to find out. LOL.
 

Adroid

Distinguished
Try taking it out of the cabinet - that sounds like a bad idea. Otherwise you may consider cutting a hole in the cabinet to provide suffecient airflow.

You might take it out of the cabinet just to check if it affects your temps.
 

neograndizer

Distinguished
Sep 10, 2010
179
0
18,710
It's an open cabinet. The front and back of the cabinet is open (not enclosed). Also, the Raven RV02-E has airflow going from bottom to top. It does not have front to back airflow like regular cases. I have about a 5" clearance from the top vents of the case to the ceiling of the cabinet. I think I could have gotten it a celcius or two lower with another Scythe 1900RPM fan, but then it would be vacuum cleaner noisy. Currently it's whisper quiet and I prefer it that way. :)

Honestly, I think it's more of the chip here. Each chip will not perform the same and mine seems to run hotter than a regular chip.