Overclock.net › Forums › AMD › AMD Motherboards › ROG Crosshair VI overclocking thread
New Posts  All Forums:Forum Nav:

ROG Crosshair VI overclocking thread - Page 851

post #8501 of 16839
Still getting a few F9 reboots during, well, reboots, but much more consistent than before. Most of the time it boots first time.

4.0Ghz is no go at 1.42v LLC3 using IBT AVX, getting WHEA errors
post #8502 of 16839
Quote:
Originally Posted by TheK View Post

@elmor

could you provide a 1002 bios with just new agesa and nothing else more???

very unstable new bioses here in ram frown.gif
I've lost 3200 14 14 14 14 30 stable @1.375
I've lost 3200 14 12 12 12 24 stable @1.405
I've lost 3200 12 11 11 11 28 stable @1.485
I've lost 3240+ any cas and voltage

Yep, all of the new BIOS setups have killed my 3200 R1700 except 083, which takes it from 1T to 2T. So I'm likely going back to 1002 tonight.
post #8503 of 16839
Quote:
Originally Posted by gupsterg View Post

I read ~1.380V +/- 5mV on ProbeIt VCORE point, R7 1700, +162mV LLC [Auto].





I see MAX as 1.356V in screenie above, so if I add 0.011V = 1.367V as Pantsoftime highlighted in this post, seems right to me.
At stock during x264 I read a max of 1.2417 V at ProbeIt vs. 1.2111 V at socket vs. 1.194 V SVI2 (hit 1.20 V some time later).

Adding 0.011 V to SVI2 closes the gap between SVI2 and socket, but ProbeIt still is far too high. And that's at stock LLC0 (Auto), it gets worse at higher LLC settings, except that I measured LLC0 and LLC1 to be the very same (!) at ProbeIt. Yep, useless. rolleyes.gif
post #8504 of 16839
Quote:
Originally Posted by Frikencio View Post

Disabled Sense Mi Skew.

Now my temp readings are +20ºC
These temps are correct. If you own an X processor you will notice jumps of +10 C and +20 C happening often, this is the dynamic offset at work.
post #8505 of 16839
Quote:
Originally Posted by Timur Born View Post

These temps are correct. If you own an X processor you will notice jumps of +10 C and +20 C happening often, this is the dynamic offset at work.

Is this just speculation or do you know this?

Anything official from AMD?

Seems a bit odd.
post #8506 of 16839
DDR4-3845 12-12-12-22 1T "stable" for aida64 and Cinebenches.



bios 0082
Edited by infraredbg - 4/11/17 at 1:08pm
post #8507 of 16839
Quote:
Originally Posted by infraredbg View Post

DDR4-3845 12-12-12-22 1T "stable" for aida64 and Cinebenches.


Wow... do you have any thoughts you could share on how you got that dialed in? What BIOS are you on? Did you notice the 'memory hole' as you were trying different frequencies? How far do you have to back off to be stable for regular use?
Computar Macheine
(10 items)
 
  
CPUMotherboardGraphicsRAM
i7-4930K @ 4.7GHz ASUS X79-Deluxe EVGA GTX1080Ti FE (w/EK waterblock) G-Skill Ripjaws Z 4 x 4GB 2133MHz 
CoolingKeyboardPowerCase
Custom Water (1 x EK 240mm, 1 x XSPC 140mm) Razer Blackwidow Ultimate Thermaltake Toughpower XT 775W Cooler Master HAF X 
Mouse
G700 
  hide details  
Reply
Computar Macheine
(10 items)
 
  
CPUMotherboardGraphicsRAM
i7-4930K @ 4.7GHz ASUS X79-Deluxe EVGA GTX1080Ti FE (w/EK waterblock) G-Skill Ripjaws Z 4 x 4GB 2133MHz 
CoolingKeyboardPowerCase
Custom Water (1 x EK 240mm, 1 x XSPC 140mm) Razer Blackwidow Ultimate Thermaltake Toughpower XT 775W Cooler Master HAF X 
Mouse
G700 
  hide details  
Reply
post #8508 of 16839
Quote:
Originally Posted by SpecChum View Post

Is this just speculation or do you know this?

Anything official from AMD?

Seems a bit odd.
I speculate this from my own thorough analysis. tongue.gif Since I can reproduce this and spent quite some time on all of this I am pretty sure about the dynamic offsets (+0/+10/+20) and I am even more sure about the "Sense Skew" offset in Asus BIOS and Elmor acknowledged that it sk(r)ews Tctl to the point where it can even keep the CPU from thermal shutdown.
post #8509 of 16839
Quote:
Originally Posted by y0bailey View Post

Yep, all of the new BIOS setups have killed my 3200 R1700 except 083, which takes it from 1T to 2T. So I'm likely going back to 1002 tonight.

I also lost a 100% POST with 3200 - 14-14-14-14-32-1T going from 1002 to 0082
post #8510 of 16839
Quote:
Originally Posted by pantsoftime View Post

Wow... do you have any thoughts you could share on how you got that dialed in? What BIOS are you on? Did you notice the 'memory hole' as you were trying different frequencies? How far do you have to back off to be stable for regular use?

Bios 0082. I wouldn't test stability at the voltages I'm running.
I've only tested 2x8 and 4x8 B-die at 3200 14-14-14-34 1T 1.35V for stability. Basically mimic the G.Skill screen and the best AMD stock bin at 3200C14.
There are some screens I've posted earlier with 100%+ memtest and short blend test.

No problem at 3200 here. In fact 0082 is the best bios for me until now.
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: AMD Motherboards
Overclock.net › Forums › AMD › AMD Motherboards › ROG Crosshair VI overclocking thread