cancel
Showing results for 
Search instead for 
Did you mean: 

CPU VCSSA Voltage

HK-47
Level 10
I have been having some issues with stability while gaming. I am running trifire 7970's 3x1 eyefinity and only way I can seem to get the system stable is to run VCSSA at 1.256-1.26V. Using a single 7970 card system is fine with a lot lower VCSSA voltage. Also the system only seems unstable when using eyefinity in crossfire/trifire. Single screen crossfire/trifire is fine.

When the system becomes unstable the image will freeze on the screen and I get a sound loop. Sometimes the sound loop does not happen.

Is this voltage safe and if so why is the extra VCSSA volatge needed with 3 cards in eyefinity mode?
-Desktop-
Asus Crosshair VIII Dark Hero / AMD 5800x / 32GB Corsair Dominator Platinum RBG (CMT32GX4M4C3600C18) @ 3800 /2x Sabrent 1TB Rocket NVMe 4.0 (Raid 0)/ 1x Samsung 860 Evo SSD / Corsair AX1200 PSU / ASUS Strix Helios Case / Corsair HydroX Custom Loop D5, CX7, XG7 420+280 Rad/ Asus Strix 3090 / Asus PG35VQ Ultrawide Monitor / Corsair Commander PRO 3x ql120, 4x ql140 fans / Edifier 1850DB Speakers + T5 Sub /
Asus Theta 7.1 Headset
13,165 Views
13 REPLIES 13

HiVizMan
Level 40
Do you have the extra power connected for the PCIe lanes? I have not had any issues when I have the additional power connected in.

What you could also do is run a full series of memtest86 and make 100% certain your ram is stable.
To help us help you - please provide as much information about your system and the problem as possible.

I have been running VCCSA at 1.25v since November without any ill effects. In my opinion that voltage is safe. Increasing VCCSA proves beneficial under a number of scenarios including overclocking the ram, running 32 or 64 GB of ram, compensating for a weaker IMC in the CPU, or overclocking the CPU itself to higher levels.

HiVizMan is going the better route here to see if the instability might be related to memory or power plane issues. It is good advise.

Barring any root cause identification, if you need to run your VCCSA at 1.25v to be stable, it's no big deal to do just that.

HK-47
Level 10
I have the extra cable plugged in and memtest86 no issues. A little after writing my post I did get a blue screen. It is the first one I have got! Normally the system locks up how I described in my original post. This time the game paused for a sec. Sound continued playing normal then screen went black followed by a blue screen stating a CPU core did not respond. How can I upload my minidump file?

This was after about 4 hrs of gaming prior to the tweaks I did to the system it would hard lock in about 5-15 min. So this is a improvement.
-Desktop-
Asus Crosshair VIII Dark Hero / AMD 5800x / 32GB Corsair Dominator Platinum RBG (CMT32GX4M4C3600C18) @ 3800 /2x Sabrent 1TB Rocket NVMe 4.0 (Raid 0)/ 1x Samsung 860 Evo SSD / Corsair AX1200 PSU / ASUS Strix Helios Case / Corsair HydroX Custom Loop D5, CX7, XG7 420+280 Rad/ Asus Strix 3090 / Asus PG35VQ Ultrawide Monitor / Corsair Commander PRO 3x ql120, 4x ql140 fans / Edifier 1850DB Speakers + T5 Sub /
Asus Theta 7.1 Headset

I'm running mine on VCCSA 1.25v too (it was the Auto choice of the board, so I just adopted it). Did some testing lowering the VCCSA voltage but it hurts the stability, so I left it on 1.25v. Haven't noticed any temp issues (same average temp regardless).

FYI I'm running my rig on 4.7G 24/7 with 32G RAM and all PCIe slots busy and some heavy duty cards in there, in addition to a bunch of HDDs so probably that's why the board raised the VCCSA to 1.25v (on run-time it's swinging between 1.225-1.275v). I think you'll be Ok.

HiVizMan
Level 40
What you have just described sounds like a possible driver issue. You can see the error in the Microsoft Event Viewer by the way.

And when you did the memtest86+ was your ram clocked to 2133MHz at the time?
To help us help you - please provide as much information about your system and the problem as possible.

HK-47
Level 10
Here is what I got from event viewer:

The computer has rebooted from a bugcheck. The bugcheck was: 0x00000101 (0x0000000000000011, 0x0000000000000000, 0xfffff88005fd7180, 0x0000000000000003). A dump was saved in: C:\Windows\MEMORY.DMP. Report Id: 082512-14476-01.


And this from BlueScreenView:


ntoskrnl.exe ntoskrnl.exe+d5efa fffff800`03250000 fffff800`03838000 0x005e8000 0x4fa390f3 5/4/2012 1:18:59 AM Microsoft® Windows® Operating System NT Kernel & System 6.1.7601.17835 (win7sp1_gdr.120503-2030) Microsoft Corporation C:\Windows\system32\ntoskrnl.exe


I am not sure if this is any help.
-Desktop-
Asus Crosshair VIII Dark Hero / AMD 5800x / 32GB Corsair Dominator Platinum RBG (CMT32GX4M4C3600C18) @ 3800 /2x Sabrent 1TB Rocket NVMe 4.0 (Raid 0)/ 1x Samsung 860 Evo SSD / Corsair AX1200 PSU / ASUS Strix Helios Case / Corsair HydroX Custom Loop D5, CX7, XG7 420+280 Rad/ Asus Strix 3090 / Asus PG35VQ Ultrawide Monitor / Corsair Commander PRO 3x ql120, 4x ql140 fans / Edifier 1850DB Speakers + T5 Sub /
Asus Theta 7.1 Headset

8_Pack
Level 12
101 means your Vcore is too low for the load your giving the CPU at a given frequency. Raise you Vcore slightly and retest.

HiVizMan
Level 40
Yeah Vcore is what that BSOD mostly indicates. Increase one notch or two and see if that resolves.
To help us help you - please provide as much information about your system and the problem as possible.

HK-47
Level 10
I am using offset mode at + .050.
CPU load Line calibration- Regular
4.8ghz
Vcore is 1.38 under load and 1.4v unloaded prior to it down clocking when idle. Idle it is 1.2ghz @ .88v does this sound ok?

Vcore prior to this was 1.38v and 1.36 under load.
-Desktop-
Asus Crosshair VIII Dark Hero / AMD 5800x / 32GB Corsair Dominator Platinum RBG (CMT32GX4M4C3600C18) @ 3800 /2x Sabrent 1TB Rocket NVMe 4.0 (Raid 0)/ 1x Samsung 860 Evo SSD / Corsair AX1200 PSU / ASUS Strix Helios Case / Corsair HydroX Custom Loop D5, CX7, XG7 420+280 Rad/ Asus Strix 3090 / Asus PG35VQ Ultrawide Monitor / Corsair Commander PRO 3x ql120, 4x ql140 fans / Edifier 1850DB Speakers + T5 Sub /
Asus Theta 7.1 Headset