Vcore problem on Asrock P67 Extreme 4 rev 3B

opti2k4

Distinguished
Oct 5, 2011
10
0
18,510
Hello,

i noticed there is a problem with Vcore voltage on this MB. In bios i setup up voltage manually for my 2600K to 1.165V. Then i enter to windows, everything is good and stable, fine temperatures, fine voltage. Then i put my computer to sleep and after it wakes up voltage of CPU it 1.36V ! If i restart it, voltage is again on 1.36V! Only solution that works is that i enter bios, put voltage to 1.160V save and exit. After the restart voltage is 1.160V so i need to change it to 1.165V. During 1st restartat if i select 1.165V instead of 1.160V system wouldn't change my voltage and it would remain on 1.36V. So with 1st restart you MUST change Vcore to different value so system would accept the change.

That is really irritaing! Anyone else experiencing same problem?
 

opti2k4

Distinguished
Oct 5, 2011
10
0
18,510
Didn't help :(

Speedstep is turned on and power savings option is disabled. When computer wakes up from sleep i noticed Vcore is going und 1V and CPU Clock goes under 1600MHz.

If i boot computer and i let it be without any load, CPU clock goes to 1600MHz and never under. Vcore remains the same (1.165/1.176) but never under.


Possible solution could be in turning speedstep and doing tests.
 

This is NORMAL, leave it. Once under Load the CPU Ratio will increase. Use CPU-z and Prime95 (running), you'll see the full CPU speed. Keeping the CPU at full speed unnecessarily wastes electricity and CPU high temps.


The vCore from the 1st post to your 2nd post seems to be corrected.
 

opti2k4

Distinguished
Oct 5, 2011
10
0
18,510
I am using Aida64 and that is not normal behavior. When voltage is set on given value (1.165V) computer never downclocks itself under 1600MHz which is OK by me and i don't mind doing that since it saves power and CPU vcore never goes under 1.165V.

But when computer wakes up i see Vcore going under 1V. This is what bugs me. It shouldn't do that since load line calibration is reduced to minimal value, power savings are not turned on and i've set up vcore manually and not on auto.

Btw i tried disableing Speedstep but i didn't help. I have 1.165 shown in bios, turned off speedstep, powered off the computer. After i powered it on, again Vcore is on 1.36V. I mean FFS!! What is going on with this shitty MB?!

 
The Default Sandy Bridge 'CPU Multiplier' is indeed (16) - Zero Doubts. {1600MHz is 1.60GHz}

If you really want it to be running at full speed 7/24/365 then:
C1E -> Disabled
SpeedStep -> Disabled

IMO - bad idea, but knock yourself out.

This is normal:

capturevq.png
 

opti2k4

Distinguished
Oct 5, 2011
10
0
18,510
pic1od.jpg


pic2iuq.jpg


I reseted all settings to EUFI defaults and those are Vcore values that i have. It's funny that i have stable 4GHz running on lower Vcore then it is currently set automatically when CPU is under load on stock frequency.
 

opti2k4

Distinguished
Oct 5, 2011
10
0
18,510
And i tested the sleep mode but Vcore didn't go to 1.36V after i woke the computer, it remained on these values.

Btw i am overclocking with use of manual vcore setting since i don't know how to use offset :)
 
If you're going to OC up tp 4.5GHz then less is more; start with only:
BCLK -> 100MHz
vCore -> {lowest stable} ; depends on Phases and CPU
CPU Ratio -> depends on vCore and temps

The ASRock are nice, but 8 Phase is typically going to raise your vCore above a 12~20+ Phase MOBO.

The image I posted is NOT mine, below is mine:
CPUZ-GPUZ-J.jpg
 

opti2k4

Distinguished
Oct 5, 2011
10
0
18,510
I am stable at 1.165V on 4GHz. That is why i am frustrated about this problem. I dont want MB to put 1.36V on my cpu since it heats up pretty much on that voltage and i dont know where is the problem.
 

opti2k4

Distinguished
Oct 5, 2011
10
0
18,510
Ha another mistery :)

Fixed Vcore setting of 1.165 V -> 1.168V detected -> everythig runs stable on 4GHz
Offset value of -0.175V -> 1.168V detected -> i can't even boot windows (blue screen or auto reset).

Wtf?
 
That's from a vDroop, use the following:

"I am stable at 1.165V on 4GHz. That is why i am frustrated about this problem. I dont want MB to put 1.36V on my cpu since it heats up pretty much on that voltage and i dont know where is the problem.

It's not that I mind, too much, OC'ing the CPU it's that those posts can get lengthy.

Here's our OC'ing section -> http://www.tomshardware.com/forum/forum-29-347.html

Ryan loves ASRock and is an expert with the ASRock - http://www.tomshardware.com/forum/profile-5043.htm
 

opti2k4

Distinguished
Oct 5, 2011
10
0
18,510
Well the funny thing is that it happens also on stock clock so i would say this has nothing to do with overclocking.