Crosshair VII Hero Essential Info Thread

[Tunza]

Hi All,

I have finally gotten some sort of consistent stability after the 9234th BIOS flash (now on 0702) and 3rd re-install of windows 10. This time around I have not been using any monitoring software (which is really frustrating as I want to know my temps and voltage for overclocking), or corsair link/iCUE (also frustrating as I have no custom control over my expensive peripherals) after reading these might be causing issues. I am just wondering if anyone knows if there is a 'safe' way I can run this software without my board chucking a tantrum. I expected to be able to use this software on a board that cost me $450ish AUD, especially considering I have never had issues like these on boards that have cost less than half! This is my first ASUS/ROG board and I am very dissapointed to say the least. I decided on this board as I wanted the very best on AM4.

My other specs include R7 2700x cooled by corsair H100i V2(stock settings, due to lack of monitoring software), 32GB Gskill TridentZ-RGB 3200mhzC14 clocked at 3200mhz (only god knows how I managed to finally get it stable), ROG Strix Vega64 8GB (stock), 1TB samsung evo960, 1tb samsung exo850 SSD and 1000W corsair RMi PSU.

Any help is greatly appreciated.


Regards,

Tunza down unda

[gupsterg]

AFAIK the Corsair iCue does not support SMBus locking, see section named here for some info.

I would not use UEFI 0702, especially if you use a lot monitoring SW/ones that use the SMBus/access Super IO chip. It has a feature called ASUS WMI, but is bugged in that UEFI.

Do you happen to have a C7H WiFi?

[Tunza]

AFAIK the Corsair iCue does not support SMBus locking, see section named here for some info.

I would not use UEFI 0702, especially if you use a lot monitoring SW/ones that use the SMBus/access Super IO chip. It has a feature called ASUS WMI, but is bugged in that UEFI.

Do you happen to have a C7H WiFi?

Hey Gupsterg,

Yes it is the wifi version. I changed back to this UEFI revision as I could not get a succesful OS handover with any other edition. which part of the board causes the major issue? Can the WMI be disabled? I heard that even after a flash the bios can sometimes remain corrupted. maybve that is why 0804, 1101 and 10XX all failed for me.

[gupsterg]

I have never experienced a corrupted BIOS on any of the ASUS boards I own, I still have a Q6600 on an ASUS P5K Premium; purchased at launch and still in use/OC'd . I use Flashback method on the boards which have it, I would advise you to use that.

I had no issues with 0804 as long as I did not go overboard on monitoring SW in use. Avoided AIDA64, etc, only used HWINFO. UEFI 0012 onwards has ASUS WMI fixed (near top of OP is link to folder of UEFIs), this was a beta prior to 1001 release, again held same OCs/profiles for me and worked without a hitch.

I believe the vendor ID for the onboard sound changes with UEFI 1101, which may actually stay in future UEFIs so you would just have deal with it as needed. This also has AGESA 1.0.0.6.

ASUS WMI can't be disabled. If an application does not use it, it will lead to fan PWM issues. If you could disable it you'd still have fan PWM issues. The fix is as stated two fold, UEFI needs ASUS WMI and applications need to use it.

[Tunza]

I have never experienced a corrupted BIOS on any of the ASUS boards I own, I still have a Q6600 on an ASUS P5K Premium; purchased at launch and still in use/OC'd . I use Flashback method on the boards which have it, I would advise you to use that.

I had no issues with 0804 as long as I did not go overboard on monitoring SW in use. Avoided AIDA64, etc, only used HWINFO. UEFI 0012 onwards has ASUS WMI fixed (near top of OP is link to folder of UEFIs), this was a beta prior to 1001 release, again held same OCs/profiles for me and worked without a hitch.

I believe the vendor ID for the onboard sound changes with UEFI 1101, which may actually stay in future UEFIs so you would just have deal with it as needed. This also has AGESA 1.0.0.6.

ASUS WMI can't be disabled. If an application does not use it, it will lead to fan PWM issues. If you could disable it you'd still have fan PWM issues. The fix is as stated two fold, UEFI needs ASUS WMI and applications need to use it.

Thanks heaps for all of your help. I am going to try UEFI 1002 as you suggested in another post. And I will use the flashback method as you have suggested . I do hope that the sound issues do get fixed in version 1101. In its current state, supreme FX does not function properly . I don't think I could live with that, as the sound features are something that I paid for when I got the board. The fancy advertising won me over LOL.

I will let you know how I go with 1002. Thanks again!!

[gupsterg]

No problem .

I doubt sound will be rendered useless on motherboard and just expect this to be blip currently. It maybe as the UEFI has not become "official" yet all the OS software side has not yet been updated for it. But dunno, I am just an avid owner/user and have no connection with ASUS .

What I can tell you as an owner of gen 1 Ryzen/Crosshair VI Hero from launch day in early 2017, is that ASUS have really supported the C6H well. So well that I still can not put the board away or say it is gimped vs the C7H. I expect the C7H to me supported as well as that TBH.

mtrai has good experience with using Ryzen and also dabbles in UEFI modification, he's had a good experience on UEFI 1101, read from here onwards in OCN thread.

Look forward to reading how 1002 is for you .

[wonderiuy]

Latest from Elmor: PinnaclePi-1.0.0.6 BIOS releases are delayed and targeted for mid November 2018
and (sadly) Elmor quitted from Asus:
https://www.overclock.net/forum/27694588-post38965.html

[Tunza]

No problem .

I doubt sound will be rendered useless on motherboard and just expect this to be blip currently. It maybe as the UEFI has not become "official" yet all the OS software side has not yet been updated for it. But dunno, I am just an avid owner/user and have no connection with ASUS .

What I can tell you as an owner of gen 1 Ryzen/Crosshair VI Hero from launch day in early 2017, is that ASUS have really supported the C6H well. So well that I still can not put the board away or say it is gimped vs the C7H. I expect the C7H to me supported as well as that TBH.

mtrai has good experience with using Ryzen and also dabbles in UEFI modification, he's had a good experience on UEFI 1101, read from here onwards in OCN thread.

Look forward to reading how 1002 is for you .

Hey again gupsterg,

Flashed to 1002 Now. Mostly things seem to work. I have had HWinfo open and running, Ryzen master too. Still yet to try iCUE, CPU-Z, HWMonitor. Am I able to run more than one software at a time? I have still been sticking to the one programme for now. Does iCue use WMI? Should I be safe with it now?

A few things I did have happen to me in UEFI 1002:
1. During post/training the board seems to turn completely 'off' after giving me a single 'all clear' beep. After about 5 seconds it starts up again and complets POST. This seems to be a little different to how the cold boot used to happen for me. Could be my imagination though.
2. Occasionally (Far far less than what I used to have happen to me) I still get stuck in-between what I believe to be POST and OS handover. I get the single 'all good' beep, then I get stuck on the splash screen. POST code usually reads 'b4'. I am unable to enter UEFI or OS once this happens, it just hangs. I have to reset the PC to get anyway. As with what I used to experience, it can sometimes get stuck on 'b4' (or A2, A0, d0... it seems to be a random freeze) over and over and over. At this point I have to cut the power to the PSU and do a 'cold boot' to get the board to win10 successfully. This has always confused me as I have only ever gotten the 'all clear' single beep when this happens.
3. My BCLK is reading 99.8 in HWinfo. In the UEFI i have BCLK set to DOCP/ fixed 100mhz (not auto). The clock seems to stay fixed at 99.8. When I change it back to AUTO, I experience the stand slight fluctuations from approx 99.8 to 100.4 when read from HWinfo with periodic polling enabled.
4. If I change my windows power plan to 'ryzen balanced' (manually changing minimum processor state to 5% to match the standard balanced plan) I seem to get uncomfortably high fluctuations in voltage (ryzen master claims slightly above 1.5 vcore). if I change to high performance plan, the fluctuations are far smaller, and the maximum vcore is approx 1.4V. One thing that really stumped me was that I was reaching the EDC limit on this power plan (according to zen master), but in the balanced power plans it stayed at about 5%. I did not touch any UEFI settings, just the power plan.

With this in mind am I better off using the asus performance enhancer function or the PBO function in the AMD CBS menu? Do these settings work together or does one setting simply override the other deeming it null and void (and if so which one overrides?). I have also set a very slight negative vcore offset to see voltages I am more comfortable with. Is this a good idea? Maybe I am being too overcautious haha. This is my first AMD build, and it has been a very very steep learning curve after been on intel for so long! Maybe I am too used to intel voltages and everything is OK? FYI my SOC is set manually to 1.0625 (seems to push itself down to 1.05 according to HWINFO).

Looks like I will stick with 1002 now after my small amount of testing. If any strange things happen, I will probably post here! LOL.

Thank you again for helping me get things going. I was nearly ready to chuck in the towel, but my tinkering energy has been renewed!

[bigtop1967]

Elmor did say that the problem with sound and gfx caused by incorrect device sub-IDs in 1101 is fixed and will be ok in next release, so I'm looking fwd to that..

I'm sticking with 1002 for now, seems fairly stable. Was playing around trying to get 3600MTs again ( could do it on earlier builds) and discovered something that I have encountered before but did not want to mention in case I had imagined it.

Although not a cold boot - pressing reset button while machine already on - it booted REALLY fast.... I mean really fast - <5 sec (fastboot is off). I'm trying to work out which setting it is that is making this behaviour, but not worked it out yet. I think it's to do with the cpu having enough juice at boot time but need to experiment more. At the moment i'm concentrating on getting 3600 100% stable, but if I can find settings that get me a nice fast boot time as well, that would be great.

[gupsterg]

Hey again gupsterg,

Flashed to 1002 Now. Mostly things seem to work. I have had HWinfo open and running, Ryzen master too. Still yet to try iCUE, CPU-Z, HWMonitor. Am I able to run more than one software at a time? I have still been sticking to the one programme for now. Does iCue use WMI? Should I be safe with it now?

A few things I did have happen to me in UEFI 1002:
1. During post/training the board seems to turn completely 'off' after giving me a single 'all clear' beep. After about 5 seconds it starts up again and complets POST. This seems to be a little different to how the cold boot used to happen for me. Could be my imagination though.
2. Occasionally (Far far less than what I used to have happen to me) I still get stuck in-between what I believe to be POST and OS handover. I get the single 'all good' beep, then I get stuck on the splash screen. POST code usually reads 'b4'. I am unable to enter UEFI or OS once this happens, it just hangs. I have to reset the PC to get anyway. As with what I used to experience, it can sometimes get stuck on 'b4' (or A2, A0, d0... it seems to be a random freeze) over and over and over. At this point I have to cut the power to the PSU and do a 'cold boot' to get the board to win10 successfully. This has always confused me as I have only ever gotten the 'all clear' single beep when this happens.
3. My BCLK is reading 99.8 in HWinfo. In the UEFI i have BCLK set to DOCP/ fixed 100mhz (not auto). The clock seems to stay fixed at 99.8. When I change it back to AUTO, I experience the stand slight fluctuations from approx 99.8 to 100.4 when read from HWinfo with periodic polling enabled.
4. If I change my windows power plan to 'ryzen balanced' (manually changing minimum processor state to 5% to match the standard balanced plan) I seem to get uncomfortably high fluctuations in voltage (ryzen master claims slightly above 1.5 vcore). if I change to high performance plan, the fluctuations are far smaller, and the maximum vcore is approx 1.4V. One thing that really stumped me was that I was reaching the EDC limit on this power plan (according to zen master), but in the balanced power plans it stayed at about 5%. I did not touch any UEFI settings, just the power plan.

With this in mind am I better off using the asus performance enhancer function or the PBO function in the AMD CBS menu? Do these settings work together or does one setting simply override the other deeming it null and void (and if so which one overrides?). I have also set a very slight negative vcore offset to see voltages I am more comfortable with. Is this a good idea? Maybe I am being too overcautious haha. This is my first AMD build, and it has been a very very steep learning curve after been on intel for so long! Maybe I am too used to intel voltages and everything is OK? FYI my SOC is set manually to 1.0625 (seems to push itself down to 1.05 according to HWINFO).

Looks like I will stick with 1002 now after my small amount of testing. If any strange things happen, I will probably post here! LOL.

Thank you again for helping me get things going. I was nearly ready to chuck in the towel, but my tinkering energy has been renewed!

1. What can occur at times is an embedded controller (EC) firmware is updated with a UEFI update. Usually this is initiated after POST, usually you will see a message like "Please DO NOT turn off power...". Once these updates are done the board may opt to do repost or power down up from what I have seen. This is only thing that crosses my mind.

2. Perhaps bad profile? If you go to UEFI page Tool > ASUS User Profiles > Load/Save to USB > press CTRL+F2 you can save UEFI settings as txt to USB, attach file to post and will have a view.

3. Ryzen has no HW for accurate BCLK read back. SW monitoring uses other methods. You should use HWINFO with periodic polling disabled. See appropriate section in original post of this thread.

4. I don't use/install the "Ryzen Balanced", read too many issues with it TBH. I use MS power plans, balanced for W10 and high performance in W7 (min.proc. 5%). Both OS I expose Core Parking setting only via a registry edit and set it as 25% IIRC. W7 is less affected by core parking, but W10 for say 3DMark I'd see more consistent benchmark results by use CP 25%. So test and see what is best for you.

I don't use Ryzen Master. I pretty much setup rigs via UEFI 99% of the time, so I really can't comment on that aspect. With the type of OC'ing we'd be doing manually or using Performance Enhancer/Precision Boost Override I can't even envisage the VRM being taxed at all. So I'd just ignore what RM is stating on EDC limit measurement.

Excursions 1.5V when using stock CPU and or PE/PBO setups can be expected IMO. As the SMU/FIT is controlling voltage you should only see this when low loads are on CPU. So low amps/watts through silicon and it will be accounting for temperature to see if it can or can't sustain boost, etc. So I guess you have to decide what you prefer.

ASUS Performance Enhancer is say Precision Boost Override presets section. God knows what [Auto] defaults to . I use Default if keeping CPU stock or doing a straight up PState 0 OC. The other PE presets info is in original post.

I have very rarely dabbled in doing my own PBO settings and when I do I set PE to Default and then change values in PBO menu of AMD CBS to what I'd like. If use a PE preset like 3/4 you may find changes in PBO section do not override them. There was a discussion in the OCN C7H thread and a member created his own separate thread, see end of section ASUS Performance Enhancer of original post for link.

I found when I measured the VCORE, using a digital multimeter on ProbeIt point, when applying a negative offset it did really seem to make much of a difference to it. The test was using PE1/2/3 IIRC, I shall repeat testing soon . I also recall anything higher than ~ -0.081V lead to POST issues, so I sorta give up on using controlling PBO voltage using negative offset.

I too prior to Ryzen was Intel user, so has been a learning curve, we'll get there .

The effect of lowered voltage on sat SOC as you state is VDROOP, nothing wrong with this. Again there is a section in OP in regard to this but context as VCORE, but same principles. Hope that helps .