Overclock.net banner

1 - 10 of 10 Posts

·
Registered
Joined
·
183 Posts
Discussion Starter · #1 ·
Hello,

my 1st core probably holds me back in overclocking my 2500K (L040B544).
It fails in Prime95. For 4.8 GHz I need 1.43V so that it does not fail.

Is the 1st core any special or is it just one of all the same 4 cores? If MB BIOS supported turning individual cores off (it does not at the moment), could I turn 1st core off?

Or is there any other way how to make it more stable?

At the moment I have 4.6 GHz at 1.36V and all is good, I do not have any urgent need to get it to 5 GHz, but it would be nice to know, if I should bump some other voltage up a bit to avoid need for increasing the core voltage etc.

My settings:

Asus P8P67 PRO:
LCC Ultra high
VRM 350 Hz
Phase extreme
Duty extreme
CPU current 130%
VCCSA 0,95V
VCCIO 1,1V
PLL 1,80V
PCH auto
CPU spread auto.
 

·
Premium Member
Joined
·
32,931 Posts
Set your CPU Current to 140%, and try lowering your PLL voltage. Might help out some.

My chip wanted 1.52v for 4.8GHz with 1.80v PLL. With 1.50v PLL, I get the clocks / volts in my sig stable (still can't bang out that magic 5.0 though).

There is a way to disable cores in the BIOS on that board, I forget exactly where its at though. Just keep tinkering with it, you'll get her stable.
 

·
Registered
Joined
·
1,352 Posts
Advanced -> CPU configuration -> Active processor cores.

I also had the same problem, if a worker failed it was always the first one. Only way I could get it to pass was to raise the Vcore.
 

·
Banned
Joined
·
7,501 Posts
Uhh, you need to be thinking more along the lines of: 1.45v - 1.48v for 4.8GHz stability. 1.43v isn't enough...

For 5GHz, count on 1.48v - 1.52v.
 

·
Registered
Joined
·
183 Posts
Discussion Starter · #6 ·
I hoped that 1st core serves some special function connected with some special voltage and I will just tweak this voltage a bit and everything will be OK.


Quote:


Originally Posted by pioneerisloud
View Post

Set your CPU Current to 140%, and try lowering your PLL voltage. Might help out some... With 1.50v PLL, I get the clocks / volts in my sig stable (still can't bang out that magic 5.0 though).

Isnt the drop to 1.5 too big? I have this voltage stock something around 1.775, lowering it seems a bit counter-intuitive.
 

·
Registered
Joined
·
1,352 Posts
From what I've read about the PLL voltage, some people can lower it quite a bit and still be stable. I tried mine between 1.5 to 1.85v and it made no difference.
 

·
Premium Member
Joined
·
32,931 Posts
Quote:


Originally Posted by StepanPepan
View Post

I hoped that 1st core serves some special function connected with some special voltage and I will just tweak this voltage a bit and everything will be OK.


Isnt the drop to 1.5 too big? I have this voltage stock something around 1.775, lowering it seems a bit counter-intuitive.

If you can still POST, then its fine. I can post down to 1.40v PLL. I'm not stable at ALL (Windows won't even boot) without 1.45v PLL or higher. So I set it to 1.50v PLL. And it was magic.


PLL voltage can make a HUGE impact in vcore that you need (and lower temps). It all depends on what YOUR chip wants though. If yours wants less PLL like mine, then that will help. If it doesn't care and only wants vcore...then that's what it wants.

Quote:


Originally Posted by blazed_1
View Post

From what I've read about the PLL voltage, some people can lower it quite a bit and still be stable. I tried mine between 1.5 to 1.85v and it made no difference.

Some people can lower it and get less vcore. Others it makes no difference at all. Clearly you're at the opposite end of the spectrum than me.
 

·
Registered
Joined
·
1,352 Posts
Quote:


Originally Posted by pioneerisloud

Some people can lower it and get less vcore. Others it makes no difference at all. Clearly you're at the opposite end of the spectrum than me.


Yeah, I was hoping my chip would be one of those that could use less vcore with lower PLL. Unfortunately all she wanted was more vcore.
Still have some playing around to do though once I get a loop set up, temps were just getting too high with current cooling.
 

·
Registered
Joined
·
183 Posts
Discussion Starter · #10 ·
PPL 1.7V, 4.7GHz at 1.40V was not stable. I am back on stock 1.8V at it seems stable so far.

I was at 5.2 GHz at 1.57V, only ran 1 thread load so that it did not everheat, I have no idea if it is really stable or not.

I guess I will settle on 4.6 or 4.7 GHz at 1.36 or 1.4V.
 
1 - 10 of 10 Posts
Top