Discussion in 'Processors and motherboards AMD' started by chispy, Feb 22, 2017.
Do someone have a stable 4200 oc and at what vcore voltage? 1.55? :wanker:
Just a quick update on my build..
Everything is up and running, the C6H is a beast! I love my new H440 case (will post pics soonTM)
Think i got lucky with my chip,
1700 @ 4GHz 1.35v, G.skill Ripjaw V RAM at 3200MHz C14 1.35v!
I have run the above through Cinebench (score of 1778) and wprime over multiple runs, will do more stability testing tomorrow.
CPU temp in HWiNFO64 is maxing at 59c when benching
I did try for 4.1 but it wasn't stable in Cinebench even at 1.4v so I decided to stick with 4GHz.
Really happy with my build so far! just need my EVGA to send me my new GTX1070 after i RMA'd my 980Ti
Your 1700 @ 4.0GHz peaks at 1.35v max while on 100% load or is it the voltage you've set on bios along with some LLC? Either way is good already but if it's the max peak during full load it's even better and you definitely got lucky
Ryzen chips hit a wall around the 4.0GHz area, requiring much more voltage for even a "tiny" 25MHz bump.
I set 1.35 in the bios, LLC set to auto. Under load it drops to 1.309 i think it was showing in cpuz
Took a while to dial in 3.95GHz. Finally hit the four hour mark in Prime95 with no errors. Was weird getting to this point. Would walk away and come back to a black screen and no USB or reset button activity. Watched one time where it would not error out but just lose video and USB functionality.
CPU voltage 1.39375
VDDRCR soc (NB) @ 1.00625
LLC CPU @ lvl 3 and 130%
LLC soc @ same as above
Ram read where it would take 1.400 v and VTTDR @ .91 to get 3200 completely stable on this board.
Once Prime95 finishes, think am going to stay at this for a bit. Peak temp was 65 on air with a NH15 from Noctua.
EDIT: Spoke too soon. Starting over @ 1.4000 on the core. Lasted 4h 38m
BIOS Update Drastically Increases Ryzen’s Game Performance | Benchmarking
The whole construction of nV drivers for Windows Platform is to Maximize 1 Thread !
So ZEN will be slower than High Clocked 7700k for sure, but when we compare ATI Radeon and DX12/DX11.1/Vulcan then situation is oposite becouse ZEN is 8/16 CPU so More 4GHz Threads = Better
Radeons Have H/W Sheduler and nV has Software base Sheduler, the DX11.0 Game which is nV Title will always behave ~10-20% Better :book:
-> Thats why Polaris Have Primitive Discard Accelerator which is supported by VEGA also, besides Vega have HBM Cache to avoid nV Code "bug" for 6GB and more GPU's (Its not Bug just to fill up some 4GB until it stutter even when it's not needed lol, but Radeon drivers since 17.x.x has this patched well)
Primitive Discard Accelerator (becouse DX11.0, to avoid :bang
Improved Shader Efficiency
Even with the threading option enabled? It's on Auto by default.
And another one
Ah i see... i have no CPU overclock at the moment. Its all stock, if you've set your multiplier that i guess it will always be showing 4.0Ghz regardless.
Drop your cpu setting back to stock and see how the 'Balanced' profile behaves. Maybe its just my system. No complaints as such, i just thought that it was strange that the balanced profile is not using the full range of available speeds (2Ghz-4.1Ghz)
Can anyone tell where i'm meant to plug in the corsair link usb header into using the C6H? I have plugged it into the usb2 connector (no. 16 in the manual, next the usb3 one) but its not being picked up in win 10 and I'm getting an unknown usb device error in my device manager..
As far as i can tell that is the only usb2 connector on the motherboard , you might want to contact corsair or asus support as there might be a conflict of some source /incompatibilty or something.
Did you install the AMD USB Driver?
I'll raise a support ticket with both today, might also ask in the hardware section of guru to see if anyone else has had this issue.
Yes, i installed all of the drivers for my board.
Well, after five hours something did not look right with the spread of the tests across all the cores; drift was too great. Stopped and upped the VDDCR SOC one click. The drift was better constrained. Made it to 17 hours. Starting over again. This time raising the core to 1.40625 and VDDR SOC to 1.01875. This should get me to the 24hr mark.
Can you show me the screenshot of the Device Manager at the USB section there?
I can do but it will be later this evening after i finish work, it looks teh same as yours except for the one unknown device error coming from the C-link
oic.. It's because I had to manually install the USB driver.
The AMD driver installer didn't install the USB driver for me.
Since you already installed AMD USB Driver, maybe you can try
to use Microsoft Standard USB driver.
It's because I had the problem like this with my old PC and my phone.
My old PC couldn't detect my phone properly if I use Etron USB driver from
Microsoft but it work fine if I use the USB driver from Etron themself.
who know... maybe you will get the other way around.
You are doing fine and doing it correctly little by little changing voltages upping those when needed will get you a very stable system at lower voltages. Great to see you are taking your time and baby steps for full stability :thumbup: , keep at it.
Curious, does anyone now what the upper limit for VDDCR Soc is for 24/7 operation?
Only six hours into current testing. Hopefully will come back from work to it still running. Have mixed feelings about how it errors out now. Before the one thread would just state an error was encountered leaving the other threads still active. Now the whole system enters an unresponsive state with the fans being the only thing running. Is this because version 29.1 may not be considered a stable build?