Originally Posted by RyzenChrist
Anyone else getting the 0d on startup? Happened twice already
F9 then 0d? That'd be DRAM issues.
Originally Posted by majestynl
Tried many things but cant get this thing above 2400mhz. Maybe with reference clock, but before i give that an try, i will wait for new bios from @elmor
Im meanwhile i thought i will start to OC the CPU. Before i begin i set everything to stock to start from there. This is my first AMD after few years so need to get used to OC and AMD again.
First of all, when i look to HWMonitor i do see some strange vcore behaviors on stock settings? I was wondering who else has these high voltages on stock ?
Originally Posted by majestynl
So if XFR can do the 1.59 voltage peaks, these are safe limits for AMD. Interesting...
Originally Posted by superchad
I have had two Issues I have come across while trying to overclock using Offset voltages, i don't know what the exact cause is
I did an offset voltage for a 3.9GHz Overclock
1. CPU-Z during a Prime95 run would at time show a voltage value of over 2.0v
2. despite being selected to enables, SMT would be turned off, i have had to reset the BIOS a few times because of this.
also i tried some lower timings for 2666 with a no boot scenario, even though my RAM is rated at 3000
Originally Posted by Snowfox00x
My second Crosshair VI board arrived today.
Everything is on auto and the board is running the vcore voltage at 1.55ish idle and up to 1.65 on load according to CPU-Z.
Idle temps with the corsair h100i are 64 degrees C with the fans at 2200RPM.
Originally Posted by gupsterg
Here is a link
from AIDA64 forum, post by author.
Besides The Stilt explaining how SW doesn't read VID/VCORE for a state properly as there are now "Shadow states" which are "fired" at very very fast speed, it is explained some what there as well.
As CH6 owners we have access to the voltage read points by the 24pin power plug, so a DMM is the way to go IMO.
XFR/CPB should only boost up to about 1.45v by default. I do strongly suspect we have a voltage reading problem on the C6H (large fluctuations), there's no way the system would still function with CPU Core Voltage of 2.0V at ambient temperatures. Can someone verify that's the case with DMM?
Originally Posted by mistax
A bit disappointing. I was messing around with more setting and i can now get a stable 3.95 with just 39.5x100, 1.35vcore, LLC @ Level 3. But Ram speed for my 2x16gb is stuck at 2666. I've only been able to get the speeds up to 3000/3200 with refblck overclocking @ the 2133 base speed
Originally Posted by egandt
Well I have an 1800X and the ROG Crosshair VI, with 64GB of Memory (so 4x16GB, as I do a lot of VM work so memory is king). So far I'm less than impressed, with the 5803 BIOS at least 2666 CAS16,15,15,15,36 is working with 5703 I could only get 2133 to even Boot and then at CAS 18,16,16,16,38.
As for the CPU:
3.925 at 1.325 is as good as it gets anything more and windows does not crash, but the system simply reboot and hangs (not very friendly, as there is no warning simply reboot and hang on reboot), I've gone as far as 1.4065, which reads sometimes as high as 1.5 and that still only gets me 3.925 stable although I can run Cinebench15 at 4.0, anything much else causes the reboot.
I always seem to have bad luck on CPUs (my 7700K only does 4.85 stable and I'm running 4.7 as 4.85 required too much voltage to run 24K). However The 1800X at 3.925 is less than the Boost frequency which is simply depressing, since if each core can handle 4.0 Stable then there is no excuse for the Chip maxes out at 3.925, I'd not even consider that an OC all things considered.
Reading around OC failures on the RyZEN are suppose to be graceful, but even inching up at all over 3.925 (even by few MHZ causes the system to black screen and hang, which is anything but graceful).
I'm on a custom Water loop, so the CPU stays under 72c max I've ever seen is 79c at 1.4v so I do not think the temperature is an issue, and since I get no warnings over 3.925 I'm assume it is just a sad chip.
Until we get an update from AMD that DRAM configuration is limited to 2666 ratio as highest. Refclock is needed to increase frequency after that. Coming update should increase the max frequency for 2x16GB.
Originally Posted by RS87
*My issue resolved!*
Just signed up[ t say a big thanks to Elmor for the bios and the advice from the thread.
I was having issues with the 55 code from out of the box set up. I then realised after a few hours of swapping every RAM module around every slot in every configuration that it wasn't that. I then noticed that the red POST led light (below the q codes) was on (not the green 'CPU is seated correctly' led).
I decided to bite the bullet and removed the seated chip, as this is supposed to be a probable cause and all the pins were flawless, not a single one bent. The H110i cooler was clipped in perfect both times. Nothing. 55 code and red cpu quick led.
i have ram that is not on the QVL list (Corsair Vengenance LED 4x8GB @3466MHz C16) but it went past the yellow quick POST led and stopped on red. So although they are not verified RAM, the mobo had essentially verified it.
pulling hair out reading so many threads about bricked boards and i thought i was one of these unlucky souls.
Alas, page 1, post 1! Elmor's 5804 bios. I extracted it to a FAT32 formatted usb stick, renamed it C6H.CAP (all capitals), stuck it in the correct USB slot (its labelled on the back plate), held the BIOS button for 3 seconds (whilst shut down but PSU on with mono lights on) and then the BIOS button began to flash. it stays flashing whilst its extracting and updating the BIOS. When the light when out, i pressed the reset CMOS button next to it and thought i'd give it one last go.
I seated one stick of 8GB RAM into the second RAM slot (A2), as stipulated in the user guide, and plucked up the courage to give the power button one more press....
the Q codes went mad like normal but then never stopped on 55 this time!!
The yellow LED below it lit, then the red (where it stopped before) long pause on the red..... but then WHITE!!! then a bit more of a flutter with codes and lights and then BOOM! the almighty single beep from the speaker! we are in action, a few seconds later and i was into the bios settings for the first time!
its been largely plain sailing since. i cant get past 2666Mhz with the memory but i now have all 4 sticks in and the system is up and running!
Sooo, my over dramatic story aside THANK YOU Elmor.
(oh and the screen wouldn't ever load all throughout the hair pulling process at until i done the bios updated and the machine POSTed for the first time.)
If this helps just one person then i'll not fret about typing this big ass first post of mine here at overclock.
Great to hear some success stories in between this mess
Originally Posted by LuckyImperial
It sounds like you're a victim of the brick bug. You can try flashing the 5803 BIOS on the first page first post, but don't get your hopes up.
However, Asus has said (if you read Elmor's posts) that they might be able to fix the bricking issue with a future BIOS update.NEW PEOPLE TO THIS TRHEAD - ITS WORTH READING ALL OF IT
Originally Posted by Enzarch
Any chance this might recover some of the "bricked" boards? I booted mine to the dreaded "BIOS updating" screen this evening, eventually resulting in not being able to power up at all.
A chance to not have to dismantle my cooling loop and especially not to have to deal with an RMA would be splendid.
Maybe I was not clear enough. A bricked board cannot be fixed by a BIOS update, we'll only be able to prevent the bricking in the first place.
Originally Posted by Cyberhelix
I have read all and I have tried that,
I even retried the BIOS that came on the DVD
Didn't work for me...
Sounds like RMA time.
Originally Posted by malitze
Maybe the information on how you were able to "jumpstart" an otherwise bricked board can give any clues to elmor. Very interesting indeed.
Originally Posted by Enzarch
A bit of info that maybe will help the Asus engineers.
While a bricked C6H wont respond to the power button, the PSU can be "jumped " and the mobo will attempt to POST. the behaviour thus forth seems to be dependent on the BIOS loaded.
v. 0702 POST codes cycle and quickly halt on 61
v. 5704 POST codes cycle indefinitely
v. 5803 POST cycles a few times, display initializes, and displays the ROG splash, halts there along with POST code 99
Will continue to test and report if there is anything else
This because the BIOS update bug kills the embedded controller (EC) and one of its functions is to manage the power on sequence. You can still make the board POST by forcing a power on as you say, but there will be other issues since the microcontroller is not working properly in the first place.
Originally Posted by newguyagain
Ok I can confirm that setting the DRAM Boot Voltage manually to the setting of the final DRAM voltage does improve detection at boot.
My 4x16GB stickts just wouldn't work and always ended with the 0d error. After setting DRAM boot voltage it runs just fine....still at only 2133Mhz bur 4x16GB with dual rank isn't bad at the moment
That's good advice. This option is available because AMD is running their PSP and applying DRAM Ratio/timings before our BIOS code is able to execute. This means we can only apply higher voltage at that stage. If you boot from a fully powered off stage you'll get the default 1.2V until then. VBoot will tell our EC to apply higher voltage immediately at power on.
Originally Posted by FLCLimax
..IS ASUS working on a real bios update?
I hope we can have a BIOS which prevents this from happening this week, but no promises. We have at least managed to reproduce the issue which should speed things along.
Originally Posted by TyCanadian
Hi guys - question on this board and Ryzen overclocking and temps in general:
Is there a way to overclock ALL Ryzen cores, but have them overclocked such that they only draw power and clock up to that speed when they are busy (not idle)? Sort of like overclocking an Intel CPU's turbo clock while leaving the base clock lower.
I'm currently running this ASUS ROG Crosshair VI board and an 1800X at 4.1 GHz (41x multiplier on 100 base clock for now) at a Vcore offset of +0.05V (another online resource suggested using offset mode should allow idle temps to run lower than if I manually specified my Vcore) and a Vsoc of +0.25V. I'm using a Noctua NH-D15 and my temps are in the mid-50s while the CPU is running the odd little background process in Windows (near-idle), and hits the low-70s when fully Cinebench-stressed. The idle temps in the mid-50s seem high to me and my CPU fan is not as silent as I'd like it to be.
Any thoughts or things that would help me understand the best way to overclock Ryzen with idle CPU temps in mind?
Yes, P-states and C-states are still active in OC Mode. If you use offset mode you'll also get the lower voltages associated with the lower power states. You're not getting frequency drops with your current settings? Is your OS power profile set to Balanced? Additionally Windows has a bug where it doesn't start to drop to lower power states until after 1-2 minutes after reaching desktop.
Originally Posted by agentk7
Elmor's absence/quietness is striking but will hopefully be fruitful in the next few days. Hopefully, they can reproduce this problem in the lab and it's a software/firmware problem thats easily fixed and not some kind of hardware issue.
I am very disheartened by all the drama around the CH6. I had initially ordered the X370 Prime Pro but decided I'd be more future proofed going with the CH6.
I'm still running on the 0702 bios after noticing high temps with the 5704 BIOS. I never upgraded to 5803. So far my system has been working ok, I set my Corsair CMK16GX4M2B3200C16 v5.39 2x8GB kit to 2666 in the BIOS but I haven't really made any other changes.
One thing I never did was install any of the Asus software in Windows. Did any of the people that were bricked do that? I believe that earlier in the thread some people mentioned they tried updating the BIOS in Windows and that may have contributed to the problem.
Pretty busy working on these issues.
Originally Posted by nycgtr
Quick question guys. I got a guy having some c6h problems. The board posted fine for me a couple time using some crucial 2400 and a trident 3200 stick. Now that he has it, using either the beta bios on the website or 5803 hes getting a 03 boot errior with the dram light on. He's tried the vengence lpx, the new flare x, and trident z 3200. It posted one time and never again after that after he flashed the bios within the bios. Tried flashing different bios with the usb and it always goes to 03 dram. Any ideas?
Did he ever get the BIOS updating bricking bug? If not he can try to reseat the CPU.
Originally Posted by newguyagain
Good to hear, please fix the temp issue - I am sure there is one
Can you elaborate?
Originally Posted by SpaceGorilla47
Some OC i could manage to achive on BIOS 5704 with no BCLK change 4GHz @ 3200 CL14:
Doesn't boot all the time, but when it does, it's rock stable.
This settings i'm using now without any problems 4GHz @ 2933 CL14:
Would love to touch the BCLK-settings and upgrade to 5803, but the bricking issue scares me to do any changes what so ever.
Perhaps the DRAM Vboot Voltage can help you get 3200 through POST consistently? See what I wrote above.