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

ROG Crosshair VI overclocking thread - Page 920

post #9191 of 17681
Thread Starter 
Quote:
Originally Posted by gupsterg View Post

@elmor

Well looks like this evenings tinkering highlighted above had some odd effect frown.gif . So as highlighted in quoted post above all good today, after "tinker"/bench runs I shut the rig off with my "known" good 3.8GHz 3200MHz C14 setup in UEFI. Win EVT log show 22:46 last entry, I booted up ~23:33, mobo power up > weird chirp from speaker > power down > 3x mobo run through Q-Codes without powering down > clean post beep > enter UEFI = 1.5V.

AMD CBS reset, manually set entries in Extreme Tweaker still there.


AMD_CBS_0081.zip 464k .zip file


Loaded optimised defaults now, gonna do flashback and setup base profile > base profile + 3200MHz C14 > 3.8GHz+ 3200MHz C14 as done last night.

Your DRAM fails to train from cold boot and causes AMD CBS settings to reset, including P-states. BIOS is not aware of failed DRAM training and has all other settings still intact, meaning you get default voltage + your specified offset. Not sure what we can do about that at this point. Future updates might help with the cold boot issue and maybe a reliable way to detect DRAM training failures. It's also possible to have a fixed P-states. An alternative at this point would be an OS application which applies your P-state settings (FID/VID).
post #9192 of 17681
@gupsterg

Raise your SOC volts. It fixed my cold boot up issues. I never had trouble rebooting, but cold boots were a problem until I raised my SOC.
Ryzen - Elmor
(13 items)
 
  
CPUMotherboardGraphicsRAM
AMD Ryzen 1700 @ 3.8Ghz ASUS ROG CROSSHAIR VI HERO GeForce GTX 1070 32GB DDR4-3200 14-14-14-34 
Hard DriveHard DriveHard DriveHard Drive
Samsung 960 EVO 1TB Samsung 830 SSD 256GB Hitachi 2TB HGST 4TB  
CoolingOSMonitorPower
Noctua NH-D15S Windows 10 Pro Acer Predator XB271HU SeaSonic PRIME Titanium 750W 
Case
Fractal Design Define R5 
  hide details  
Reply
Ryzen - Elmor
(13 items)
 
  
CPUMotherboardGraphicsRAM
AMD Ryzen 1700 @ 3.8Ghz ASUS ROG CROSSHAIR VI HERO GeForce GTX 1070 32GB DDR4-3200 14-14-14-34 
Hard DriveHard DriveHard DriveHard Drive
Samsung 960 EVO 1TB Samsung 830 SSD 256GB Hitachi 2TB HGST 4TB  
CoolingOSMonitorPower
Noctua NH-D15S Windows 10 Pro Acer Predator XB271HU SeaSonic PRIME Titanium 750W 
Case
Fractal Design Define R5 
  hide details  
Reply
post #9193 of 17681
Quote:
Originally Posted by Reikoji View Post

I suppose try flashing 0081 or 0082, and if it persists it may be a board issue. probably not a major board issue if you can ignore it, its just unlikely that the socket is actually hotter than the cpu.

0081:


0082:


Both are the same.
In BIOS itself the Hardware monitor CPU temp (right side) starts at 60°C while the left side (monitor tab) starts at 70°C, then they fluctuate both back & forth. Hardware monitor side is slower but catches up eventually between 52 - 62°C.

So it's still there...
Do you know where those two sensors are?
Edited by Merboe - 4/13/17 at 10:25pm
post #9194 of 17681
Quote:
Originally Posted by elmor View Post

[...]

So the official bios works pretty well, even allows my ****ty 2400Mhz Crucial 16Gb kit to hit 2666Mhz with tighter timings. Out of the new ones though, first I tested 82 since many people said it was the best for them. It was trash for my RAM as was 81. Played with settings and still nothing. Refused to boot at anything even more than a hair over 2400. 79 however is just as good as the official bios for my RAM. I haven't tried pushing it further yet but it's better than any of the others for my parts.

The Aura LEDs still stay on ignoring my settings when it's off though. Other than that it's currently pretty annoying when the OC fails (or the RAM won't boot) and it resets the p-states. They do however save in the OC profile so saving/restoring is pretty easy if annoying.

Quote:
Originally Posted by madweazl View Post

My personal opinion is that none of the X sku's are worth the money if you intend to overclock. In the end, the realistic outcome is about 100mhz. If the chance of squeezing another 100mhz is worth the money, go for it.

I decided to spend the extra $20 (thanks to Micro Center's bundle pricing) to get the 1700x because of the fact that based on the info from Silicon Lottery, it looked like a 1700x had/has a better chance of clocking a bit higher and even if it doesn't it should take less voltage. Well worth it when the difference was that small. If it had been more I would have skipped it and got the 1700 instead.
    
CPUMotherboardGraphicsRAM
Intel i7 2700k Asus Sabertooth P67 Asus Radeon 7970 Ref. Crucial Ballistix 
RAMHard DriveHard DriveOptical Drive
Corsair Vengeance Crucial M4 Hitachi HDS721010CLA332 TSSTcorp CDDVDW SH-S223F 
CoolingCoolingCoolingCooling
EK Supreme HF Copper/Red Acrylic CPU Block EK 7970 Reference Nickle/Acetal Black Ice SR1 360 Rad Phobya G-Changer 280mm Radiator 
CoolingOSMonitorMonitor
XSPC Dual Bayres W/ D5 Vario @ 17V Windows 7 x64 Vizio SV320XVT Samsung SyncMaster S27A550H 
MonitorKeyboardPowerCase
Samsung SyncMaster S27A550H Logitech G510 Corsair TX950W Azza 9000B 
MouseAudio
Logitech Marathon Mouse Asus Xonar DX 
  hide details  
Reply
    
CPUMotherboardGraphicsRAM
Intel i7 2700k Asus Sabertooth P67 Asus Radeon 7970 Ref. Crucial Ballistix 
RAMHard DriveHard DriveOptical Drive
Corsair Vengeance Crucial M4 Hitachi HDS721010CLA332 TSSTcorp CDDVDW SH-S223F 
CoolingCoolingCoolingCooling
EK Supreme HF Copper/Red Acrylic CPU Block EK 7970 Reference Nickle/Acetal Black Ice SR1 360 Rad Phobya G-Changer 280mm Radiator 
CoolingOSMonitorMonitor
XSPC Dual Bayres W/ D5 Vario @ 17V Windows 7 x64 Vizio SV320XVT Samsung SyncMaster S27A550H 
MonitorKeyboardPowerCase
Samsung SyncMaster S27A550H Logitech G510 Corsair TX950W Azza 9000B 
MouseAudio
Logitech Marathon Mouse Asus Xonar DX 
  hide details  
Reply
post #9195 of 17681
Quote:
Originally Posted by Merboe View Post

0081:


0082:


Both are the same.
In BIOS itself the Hardware monitor CPU temp (right side) starts at 60°C while the left side (monitor tab) starts at 70°C, then they fluctuate both back & forth. Hardware monitor side is slower but catches up eventually between 52 - 62°C.

So it's still there...
Do you know where those two sensors are?

I unfortunately do not. My guess is the CPU socket temp sensor is most likely on the back of the board in the center area of the backplate, or part of the socket. Another more technical than I would know for sure.
post #9196 of 17681
Thread Starter 
false
Quote:
Originally Posted by Reikoji View Post

I unfortunately do not. My guess is the CPU socket temp sensor is most likely on the back of the board in the center area of the backplate, or part of the socket. Another more technical than I would know for sure.

CPU Socket Thermistor is located here

post #9197 of 17681
@elmor do you have more bios update for us to play with this weekend?
post #9198 of 17681
Thread Starter 
Quote:
Originally Posted by haryadiaja View Post

@elmor do you have more bios update for us to play with this weekend?

One more day added to the counter smile.gif (For those who keep count it's at +2 right now, you guys are doing better than I thought)
post #9199 of 17681
I am using a 4 GHz P-state overclock, no LLC, The issue I am having is that now although I can see the the various core clocks fluctuating, some at minimum, some at 4 GHz, etc, even with minimal demand going on if only one or two cores are above minimum, I see maximum core voltage being used. Light system demand triggers all eight core frequencies to go to 4GHz.

This basically negates one of the prime rationales for using P-states. How can I get P-states to be less "aggressive" in the threshold to kick into full bore, and how do I get less voltage to be used when the system is basically loafing while using a browser, etc.?

Is this the reason for all the P-states, to be able to make transitions to be more gradual?

Is this impossible to do anyway, sense apparently that mean being able to set up P-state VID for each state, which I guess does not function yet?

Right now I have only the 0 P-state set, all others are left to auto.


Edited by CeltPC - 4/14/17 at 12:42am
post #9200 of 17681
finally my com is stable at 3.9@1.375 with LLC3
should i change to LLC2 with some +v ? or safe with this setting ?
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: AMD Motherboards
Overclock.net › Forums › AMD › AMD Motherboards › ROG Crosshair VI overclocking thread