ROG Crosshair VI

[Alex@ro]

@elmor

 

Loading DOCP Standard, then the XMP (3733C17 mems) boots at 3200.

Then, tried raising bclk to see if 3600 will boot (skipping the mem hole) and it didn't boot.

This leads to infinite reboot cycles and the board can't recover. Using clrcmos does not help, turning PSU power off doesn't help. Leaving just one module doesn't help. CPU is on ambient water.

Happens for the second time, will probably need to remove battery again.

 

I found out a quick way to recover will be to use safe mode,then quickly after it ramps up you press safe mode again and then a reset after second start,i did confront with your issues too but if i double safe-mode and a reset it always gets to boot again.

[I.nfraR.ed]

Can't get it boot this way. Leaving it without battery for ~30min worked and all settings are still there in bios.

It only happens with DOCP, manual is able to recover.

[Masterchief79]

If the settings are still there after you pulled the battery out for 30mins I'm sure you didn't unplug the PSU, right?

[I.nfraR.ed]

I have turned it off from the power switch. It booted at default, but docp setting were still there.

[elmor]

I have turned it off from the power switch. It booted at default, but docp setting were still there.

 

When you press the Clear CMOS button, does the board have standby power? If not please try it with standby power on, and hold it at least for 3 seconds.

[I.nfraR.ed]

Yes, of course. It has standby power when I try the clearcmos button.

Anyway, running without a battery now.

 

Btw, things are getting better

 

3733 is the maximum I can get at the moment, regardless of timings.

Mems are capable for sure, maybe it is just my CPU's IMC.

 

 

[Kanuki]

This is my first post here but I just gotta say Elmor, sterling job with BIOS 1001!

 

I'm using the Corsair CMK16GX4M2B3200C16 v5.39 kit and up till now I'd been unable to POST @3200mhz using DOCP Standard, I tried 0038 first and it gave me the same behaviour as 0902 but I decided to take a punt with 1001 with DRAM Startup Voltage @1.4v and it bloody well worked!

 

It's completely opposite to what I was expecting as 2T should be MORE lenient, but you're not going to hear me complain about it. Keep up the good work mate!

 

Is your RAM Single Rank or Dual Rank?

[flanker]

thats very solid man

[xoqolatl]

Bravo, that's a spicy mem speed, seems above average of my 3 chips

[HattoWW]

Hi, this is my first post in this forum. Its nice to be in this community I'm using ROG Crosshair VI Hero board with r7 1800x. Could you pls share your bios settings for the ram settings in your post? I can only overclock my CPU to 4.100 Mhz and my rams to 3200mhz with 100 mhz bus speed. This was stable in 0704 bios but its not stable in the bios vers 1001 even with 1.45v CPU and 1.35V ram. Im open to your advices.

[Gorod]

I.nfraR.ed Beautiful , Very impressive ! How about graphics PCIe lanes , still @ 3.0 with 138 bus clk ?

[Kanuki]

Hi.. All.

 

I'm using R7 1700X with Crosshair VI Hero (BIOS v1001)

and Corsair Vengeance LED 16GB (8GBx2) 3000MHz CL15.

 

Right now my RAM stuck at 2666MHz 14-14-14-14-30 by

only set DRAM voltage to 1.35v.

 

 

Any advice?

[chispy]

Infrared that's some awesome speeds right there :celebration:

 

Lots of improvements with the latest Bios , thank you Jon ( Elmor ) for making this possible , without your support this would not be possible.

 

Kind Regards: Angelo

[Kanuki]

Hi...

 

Does BIOS v1001 already include new AMD AGESA Microcode?

[loccothan]

When we can get:

New AMD AGESA Microcode & New UEFI

[matose]

The 1001 is really great but I have some problems with NVMe SSD: I have crashes with any 4K workload with QD > 8. Secvential and 4K QD < 8 work fine.... is anyone else experiencing this?

 

Thanks,

matose

[xoqolatl]

I have crashes with any 4K workload with QD > 8. Secvential and 4K QD < 8 work fine.... is anyone else experiencing this?

Just tested QD=12 4K random data with IOMeter and QD=32 4K random in CrystalDiskMark and both worked fine. This is on 1001 bios and Samsung 950 Pro in M.2 slot.

[matose]

Could you please run Anvil Pro in loop 3-4 times and see if you have any BSOD / lock-ups? I tried M.2 slot and PCIe slot 2 with adapter, swap SSD's, memory, etc. I have random crashes in 4K high-QD workloads...

 

P.S: SSDs tested are Toshiba OCZ RD400 and Apacer Z280 with Phison controller. Everybody is testing with Samsung NVMe so I don't know if it's a SSD compatibility problem or CPU / MB issue!

Edited March 21, 2017 by matose

[Wizerty]

Thanks Angelo, can you also confirm the voltage using DMM? I've managed to replicate it but instead my VRM would shut down.

I also had this issue weeks ago. It happened when I used turbov (bios 5803).

boot at 1.1 V (or 1.4 v) with LLC auto or LLC1, I don't use LLC4or5 as it's too high.

 

-First time the bug happened I tried to apply 1.2 v and I was looking on T°(not on vcore). I saw T° going from IDLE to +120°C in less than a 1second when I started cinebench R15. I thought I had set the wrong value.

-2nd time same happened, +114°C and shutdown again really quick, no time to check voltage (also tried to set 1.2V on Turbov).

-3rd time it happend when I tried to apply 1.1V. I saw T° going to 98°C but benchmark was still able to run, so I used my DMM and read +1.7V.

 

My guess is if it's +98°C with 1.7V running CBr15, previous time when CPU reached +120°c in 0.1second voltage was probably more than 1.9 V.

 

...

 

I did 1900 MHz on mem also with 5803 (didn't test new bios yet). But it's difficult (at least on my setting). Anything between 3400 and 3650 is no go. anything higher than 3680 is no go.

The "windows" is really small, 1 MHz on ref_clk. this is with 2600 ratio, didn't try on other ratio, maybe it's more easy. I Will try it on new BIOS later

[elmor]

I also had this issue weeks ago. It happened when I used turbov (bios 5803).

boot at 1.1 V (or 1.4 v) with LLC auto or LLC1, I don't use LLC4or5 as it's too high.

 

-First time the bug happened I tried to apply 1.2 v and I was looking on T°(not on vcore). I saw T° going from IDLE to +120°C in less than a 1second when I started cinebench R15. I thought I had set the wrong value.

-2nd time same happened, +114°C and shutdown again really quick, no time to check voltage (also tried to set 1.2V on Turbov).

-3rd time it happend when I tried to apply 1.1V. I saw T° going to 98°C but benchmark was still able to run, so I used my DMM and read +1.7V.

 

My guess is if it's +98°C with 1.7V running CBr15, previous time when CPU reached +120°c in 0.1second voltage was probably more than 1.9 V.

 

...

 

I did 1900 MHz on mem also with 5803 (didn't test new bios yet). But it's difficult (at least on my setting). Anything between 3400 and 3650 is no go. anything higher than 3680 is no go.

The "windows" is really small, 1 MHz on ref_clk. this is with 2600 ratio, didn't try on other ratio, maybe it's more easy. I Will try it on new BIOS later

 

The TurboV Core issue is solved and should be in the next release.

[chispy]

Elmor , i'm having a really tough time getting anything decent in the Ram department with my Hynix Dual Rank ADATA AX4U2800W8G17-DRZ. The highest i can get them is 2666 at 18-18-18-36 2T at 1.35v and VDDSOC Voltage Override at 1.20v

 

I have tried many combinations on my Hynix ram for testing purposes. The best i can do is 2666 at 18-18-18-36 2T. I have left Cas alone at 18 as anything lower does not work at all , so that confirms there is some kind of hole at anything below for that , same findings as infrared and Chew, can you comment on those findings ?. I cannot get any other timing changed as soon as i start to do baby steps starting to tight latencies it's a no go on the ram. Tried 18-17-18-36 did not boot , it hangs. tried 18-18-17-36 same , 18-17-17-36 same and so on ...

 

Tried voltages from 1.20v to 1.50v for v.ram and v.ram boot voltage and did not work , tried VDDSOC Voltage Override from 1.0v all the way up to 1.25v did not make a difference. Any advice to get Hynix memory some decent speeds and timings ?

 

Testing last night for a few hours and i end it up with a completely trashed and corrupted windows 10 due to many blue screens. I need to re install windows today to keep on memory testing on Hynix Dual Rank. Thanks in advanced.

 

Regards: Angelo

 

 

 

[matose]

Tried many DDR4 kits and the conclusion is: anything else than single-sided B-die is a big fail on Ryzen. I guess the IMC is setting things really tight and the B-dies are the only IC that can handle that. Maybe Elmor could confirm?

[flanker]

seems E-die works relative OK also But B-die are the best (eg as new AMD Flares from GSkill). Waiting for mobo now, I have few E-dies, Hynixs, one B-die kit

[elmor]

Elmor , i'm having a really tough time getting anything decent in the Ram department with my Hynix Dual Rank ADATA AX4U2800W8G17-DRZ. The highest i can get them is 2666 at 18-18-18-36 2T at 1.35v and VDDSOC Voltage Override at 1.20v

 

I have tried many combinations on my Hynix ram for testing purposes. The best i can do is 2666 at 18-18-18-36 2T. I have left Cas alone at 18 as anything lower does not work at all , so that confirms there is some kind of hole at anything below for that , same findings as infrared and Chew, can you comment on those findings ?. I cannot get any other timing changed as soon as i start to do baby steps starting to tight latencies it's a no go on the ram. Tried 18-17-18-36 did not boot , it hangs. tried 18-18-17-36 same , 18-17-17-36 same and so on ...

 

Tried voltages from 1.20v to 1.50v for v.ram and v.ram boot voltage and did not work , tried VDDSOC Voltage Override from 1.0v all the way up to 1.25v did not make a difference. Any advice to get Hynix memory some decent speeds and timings ?

 

Testing last night for a few hours and i end it up with a completely trashed and corrupted windows 10 due to many blue screens. I need to re install windows today to keep on memory testing on Hynix Dual Rank. Thanks in advanced.

 

Regards: Angelo

 

Hynix ... Dual-rank ... What do you expect when we don't even have subtiming adjustments?

[Bruno]

Hi, it seems I hit a wall and can't figure out how to solve it:

Tried Ryzen 7 1700 and C6H on LN2 and it goes fine up to 4500MHz. After that the system reboots no matter the BCLK*multiplier combination.

141*31.75 works fine, but 141*32 reboots, no matter the voltage. Same goes for other bus*multiplier combinations.

It just reboots when going above 4500MHz.

BIOS is 1001 and I tried different temps, from -20 to -110. Loaded 141 bclk profile, then tried Extreme profiles (lowering the vcore and the multiplier according to temp). No go for all above 4500MHz.

Tried all voltages from XOC Guide. No go.

 

Any ideas?

 

Thx,

Bruno