Overclock.net banner

1 - 20 of 31 Posts

·
Registered
Joined
·
5,624 Posts
Discussion Starter #1
I'm trying for 4.4ghz and I have my cpu vcore at 1.4v and my ioh vcrore at 1.2v. Can they go higher safely? If so how high? Thanks!
 

·
Blue and Green Team!
Joined
·
10,736 Posts
You are not going to able to control the temperature of that CPU with your cooling my friend.

You have exceeded what is considered "an acceptable" limit for air cooling already with regards to voltage.
 

·
Registered
Joined
·
2,518 Posts
CPU cooling
Megahelams

lol............................................... ..

is this thread a joke? do you not monitor your temps when you overclock? go get some watercooling before even going beyond 1.3v on that hsf.
 

·
Registered
Joined
·
5,624 Posts
Discussion Starter #4
Quote:

Originally Posted by Cacophony View Post
CPU cooling
Megahelams

lol............................................... ..

is this thread a joke? do you not monitor your temps when you overclock? go get some watercooling before even going beyond 1.3v on that hsf.
While stress testing at 1.4v in prime, I get 87c max.
 

·
Registered
Joined
·
1,846 Posts
I had my i7 920 run at 4.2Ghz on air cooling with 1.32 Vcore.. That's about as high as I'm willing to go before temps start going 80-ish with my mugen-2. I also have a HAF X with lots of fans, and it's standing near an air conditioner.
So unless you got WC or phase change.. don't go any higher...
 

·
Registered
Joined
·
2,518 Posts
Quote:

Originally Posted by Willanhanyard View Post
While stress testing at 1.4v in prime, I get 87c max.
yea that is hot. don't go past 80C........your chip will start to degrade and bye bye cpu. trust me on this one buddy.
 

·
Blue and Green Team!
Joined
·
10,736 Posts
Quote:

Originally Posted by Willanhanyard View Post
While stress testing at 1.4v in prime, I get 87c max.
Now run LinX on MAX memory.
 

·
Registered
Joined
·
5,624 Posts
Discussion Starter #8
Quote:

Originally Posted by Cacophony View Post
yea that is hot. don't go past 80C........your chip will start to degrade and bye bye cpu. trust me on this one buddy.
I would think the 1.4v vcore would degrade it more than the temps.
 

·
Registered
Joined
·
1,489 Posts
Quote:

Originally Posted by Willanhanyard View Post
While stress testing at 1.4v in prime, I get 87c max.
Back off to 4.2Ghz, you won't see any appreciable difference in speed and your temps should drop to more tolerable levels ~80-85c at the most @ ~1.30-1.35v; with something like 21x200 this will also allow your memory to run almost exactly at their rated speed as well and you may (with a bump in vdimm and/or qpi run your memory at tighter timings).

Quote:

Originally Posted by Willanhanyard View Post
I would think the 1.4v vcore would degrade it more than the temps.
1.4v in and of itself isn't going to degrade it any more than usual. 1.55v is the max "Intel" approved voltage for the 45nm chips.
For 24/7 use, you generally want to stay > or = to 20c to tjmax (which is generally 80c or less; differs from chip to chip, core to core).
 

·
Premium Member
Joined
·
4,914 Posts
according to intel, the max vcore is 1.55v.

now, the megahalems is a very capapble cooler, and can handle 1.4v and keep it under 90C

what is your VTT at? if its not high enough it could be being made up for with lots of Vcore
 

·
Registered
Joined
·
2,518 Posts
run intel burn test and do about 20 runs with it....use 8threads. and max on memory.

after that go run prime 95 for 2 hours.
 

·
Registered
Joined
·
2,518 Posts
Quote:

Originally Posted by miahallen View Post
Not true.

gots to keep it under 80C...........more likely in intel burn test his chip won't do 87C more like 95C and if you don't think the cpu will start to degrade with that load, you're not thinking right.
 

·
Premium Member
Joined
·
4,914 Posts
Quote:

Originally Posted by Cacophony View Post
gots to keep it under 80C...........more likely in intel burn test his chip won't do 87C more like 95C and if you don't think the cpu will start to degrade with that load, you're not thinking right.
im sure he could run 90C for a year before it gave out. these things are pretty hard to kill.

I run 1.7v for over 8 hours at a time, twice this weekend, and multiple time before, and guess what? my chip is still clockin' like it always did!

Yes that was 1.7v, not a typo
 

·
Registered
Joined
·
1,465 Posts
I've benched my 920 at 4.6ghz several times with a thermalright true, its for sure hot. but if you keep it under a certain temp it wont throttle.

As long the CPU ain't throttling benching at high temps are fine. If the CPU starts to throttle at higher clock, you'll get a worse score.

The bloomfields seems pretty durable, but they are very hot agian so maybe 4.4-4.6 is the highest you can go for a running a benchmark at air/water. Thought I've done 4.8ghz at air in 3Dmark 05, but that was at winter time


Don't go above 1.3v at IHO, that should be running all rated speeds on several ram kits.
 

·
Registered
Joined
·
783 Posts
I don't mind seeing Bloomfield temps in the 90s during IBT cause nothing you throw at it day to day will ever get it that warm....and Intel says they're OK up to 100C....and I've run them at 102-103 and they've been fine


I'm not encouraging ppl to do this....I test a lot of chips and push things past the limit on purpose with full awareness of the potential risks
 

·
Registered
Joined
·
2,518 Posts
Quote:

Originally Posted by miahallen View Post
I don't mind seeing Bloomfield temps in the 90s during IBT cause nothing you throw at it day to day will ever get it that warm....and Intel says they're OK up to 100C....and I've run them at 102-103 and they've been fine


I'm not encouraging ppl to do this....I test a lot of chips and push things past the limit on purpose with full awareness of the potential risks

how many runs did you do in ibt?..........

oh i see your disclaimer lol, if op has disposable funds, than by all means.....as a 24/7 CPU folder, i like to keep it under 80C.
 

·
Registered
Joined
·
783 Posts
Yeah, I'd agree....a 24/7 folder under 80C is a good goal....is the OP building a 24/7 folder?

My point was stress testing in the 90s is not harmful and that would translate to a 24/7 folder in the 80s, or a daily user rarely exceeding 70C.....

Recently I've been building a new gaming rig "MAX11L" listed below, and my IBT runs were 20 loops (about 1 hour and 20 minutes). I tested with two CPUs....the first was an i5 670 Clarkdale at 4.8GHz and 1.45V and I was hitting 100C during the testing. That was done with the same chip that I did this with a few months ago:
http://valid.canardpc.com/show_oc.php?id=1094061

It's been over 100C numerous times in the past and will still game at nearly 5GHz all day long
 

·
Registered
Joined
·
1,465 Posts
Quote:

Originally Posted by miahallen View Post
I don't mind seeing Bloomfield temps in the 90s during IBT cause nothing you throw at it day to day will ever get it that warm....and Intel says they're OK up to 100C....and I've run them at 102-103 and they've been fine


I'm not encouraging ppl to do this....I test a lot of chips and push things past the limit on purpose with full awareness of the potential risks

Aye, but at those temps the cpu probably starts to throttle. So if the cpu is 4.6ghz stable in vantage and it hits very high temps, it will probably give less score than a 4.4ghz that could hold the temps better down. So you'll have to test how your cpu will react to higher temps like above 90...
 

·
Registered
Joined
·
783 Posts
Cacophony, everyone has a comfort zone with how hot they like to see their equipment...and everyone is entitled to their own opinion. But there are too many ppl running around spreading unfounded myths about it. If you want to advise ppl, my advice would be to do it in a statement of "in my opinion" type form. Unless you have some imperical data to backup your claims?


Again, here is the same CPU on air cooling....from the article I wrote about the build:



@ Kriztoffer....you can see in the capture above there is no throttling occuring (throttling is disabled in the BIOS).
 
1 - 20 of 31 Posts
Top