ASUS ROG Crosshair VI Hero Ryzen Motherboard Review @ [H]

Daniel Dobrowolski - maybe you can really help here if you still have the CH6. I am not the only one convinced that this board is not correctly enumerating the chipset-driven PCIe slots. In my case it seems that if there are two similar cards (I have two USB3 cards both with the same NEC controller chip) in these slots then the BIOS will only enable one of them. BIOS v1201, Win10 x64, Ryzen 1800X, no OC.

--> Can you run a test to verify or disprove that specifically? Or just run some tests with combinations of cards?

Thanks in advance. I have contacted Asus support with no result, but if you can verify a simple BIOS problem they may pay attention and fix it.

(My 2008 Asus Rampage Formula mobo identifies these cards with no problems so the fact that there is an issue and that Asus can solve it if they want to seems obvious.)

The Crosshair VI Hero died on me. I've got it, but it's dead Jim.
 
The Crosshair VI Hero died on me. I've got it, but it's dead Jim.

Arghh - yes, I figured that out as you were replying. Well, maybe someone else here can experiment a little with the minor PCIe slots. I think if the Asus BIOS team was aware of this, they'd fix it, 'cause they certainly know how.
 
It would power on, lights came up but it wouldn't POST. I don't remember the LED code.
I suppose you tried all this, but just in case anyway:
- Power off, leave PSU unplugged overnight.
- Remove CMOS battery for 1 hour.
- Replace battery, plug PSU back, clear CMOS, power on.

Before all that just for curiosity, your back I/O panel Flashback button+LED is on/flashing/off?

Edit: clear CMOS after replacing battery and plugging PSU back.
 
My one problem with this board has been not being able to run the RAM at rated speeds, I haven't even got to touch overclocking yet. It's extremely finicky on running RAMs speeds. When I updated to BIOS 1002 I was able to run at 2666 speeds. However, when I updated to 1201 it went back down to 2133. And it did not seem to matter what setting or timings I tried it would not clock higher with any stability. I also would get random reboots every once in awhile even at 2133 speeds.

I'm running a kit of Corsair Vengeance LED 32GB (2x16GB) DDR4 3000 memory. Last night I came across this article talking about the new AMD AGESA 1.0.0.6 update: https://www.overclock3d.net/reviews...esa_update_1_0_0_6_-_do_bios_updates_matter/1

This lead me to the beta BIOS for it 9945. You can get it here: https://rog.asus.com/forum/showthread.php?93420-Hurray-for-9945-beta-bios-I-am-at-3600-MHz-at-last!

HUGE improvement so far. I set my RAM speed for 2933 and everything else was auto - it picked up timings to 16,16,16,39,69. I also picked 1T which you could not even change before. So far it's stayed stable running Folding @ Home all night @ 2933 speeds on the RAM and those timings. So they do seem to be on the right track finally, they've also opened alot more memory settings for adjustment then they had.

I do wish they would add RAID 5 back into the RAID portion that they used to have in the older previous gen boards. Curious to see if this BIOS helps anyone else.
 
Last edited:
My one problem with this board has been not being able to run the RAM at rated speeds, I haven't even got to touch overclocking yet. It's extremely finicky on running RAMs speeds. When I updated to BIOS 1002 I was able to run at 2666 speeds. However, when I updated to 1201 it went back down to 2133. And it did not seem to matter what setting or timings I tried it would not clock higher with any stability. I also would get random reboots every once in awhile even at 2133 speeds.

I'm running a kit of Corsair Vengeance LED 32GB (2x16GB) DDR4 3000 memory. Last night I came across this article talking about the new AMD AGESA 1.0.0.6 update: https://www.overclock3d.net/reviews...esa_update_1_0_0_6_-_do_bios_updates_matter/1

This lead me to the beta BIOS for it 9945. You can get it here: https://rog.asus.com/forum/showthread.php?93420-Hurray-for-9945-beta-bios-I-am-at-3600-MHz-at-last!

HUGE improvement so far. I set my RAM speed for 2933 and everything else was auto - it picked up timings to 16,16,16,39,69. I also picked 1T which you could not even change before. So far it's stayed stable running Folding @ Home all night @ 2933 speeds on the RAM and those timings. So they do seem to be on the right track finally, they've also opened alot more memory settings for adjustment then they had.

I do wish they would add RAID 5 back into the RAID portion that they used to have in the older previous gen boards. Curious to see if this BIOS helps anyone else.

Johan45 linked both in here https://hardforum.com/threads/asus-...board-review-h.1936124/page-2#post-1043041285

9943 2 dimm and 9945 4 dimm
 
Whoa. Is this the downfall of Asus?

Has Asus's hardware division finally started to resemble its customer service sector???

I can tell you I have been far less than thrilled with my ASUS Deluxe 170 for my Skylake as well, especially since I shelled out $319 for it - Asus always supposedly was the top enthusiast board, but at this point, I wish I would have stuck with MSI. Asus is nothing but one headache after another.
 
My one problem with this board has been not being able to run the RAM at rated speeds, I haven't even got to touch overclocking yet. It's extremely finicky on running RAMs speeds.

That's an issue with every X370 board.
 
It would power on, lights came up but it wouldn't POST. I don't remember the LED code.

this or something similar seems to have happened already to multiple users in this thread, even with recent bios versions ... so there is a good chance that your board is not dead-dead, more like having a problem with the bios ?
like it has been stated by yourself and others, problems started when overclocking ... it seems the bios has trouble recovering, and needs a lot of time or even a secret handshake (different per bios version) to succeed

some of us have been able to make their board recover, using different methods: clear CMOS procedure, turning computer off and waiting, USB bios flashback procedure ...
this is probably way outside of the scope of a review, but it would be useful in the forum to know if you can get it back to work and how (for those of us that don't mind tinkering with their toys)

this board also has of lot of safety and recovery features, but they don't do much, or nothing at all:
the "safe boot" button, the "retry" button, the "clear CMOS" button ...
it feels like the bios corrupts itself or can't pass it's own internal checks, and fails POST ... something is wrong in that regard and I hope they can do something to solve it
 
this or something similar seems to have happened already to multiple users in this thread, even with recent bios versions ... so there is a good chance that your board is not dead-dead, more like having a problem with the bios ?
like it has been stated by yourself and others, problems started when overclocking ... it seems the bios has trouble recovering, and needs a lot of time or even a secret handshake (different per bios version) to succeed

some of us have been able to make their board recover, using different methods: clear CMOS procedure, turning computer off and waiting, USB bios flashback procedure ...
this is probably way outside of the scope of a review, but it would be useful in the forum to know if you can get it back to work and how (for those of us that don't mind tinkering with their toys)

this board also has of lot of safety and recovery features, but they don't do much, or nothing at all:
the "safe boot" button, the "retry" button, the "clear CMOS" button ...
it feels like the bios corrupts itself or can't pass it's own internal checks, and fails POST ... something is wrong in that regard and I hope they can do something to solve it

Rumour has it the X399 Asus board has a "really really safe boot" button :)

I'm sure Dan reads the forums as well , some people went as far as removing the power and battery from the motherboard for 2 days ...
 
  • Like
Reactions: noko
like this
Rumour has it the X399 Asus board has a "really really safe boot" button :)

I'm sure Dan reads the forums as well , some people went as far as removing the power and battery from the motherboard for 2 days ...
Also a few rain dances thrown in as well. I look at it as like a perfect storm, new platform, ASUS tries to outdo everyone with a killer board then you add the extra features on top of AMD AGESA, then the ASUS software layers and you end up with the CH6.

Love the extra USB ports, the first four are direct from the CPU (lowest latency), the bios flashback feature, BCLK feature, the endless Bios options, great VRMs. It is just getting everything working together that is not working well with the CH6. Dump the ASUS software (all of it) and it becomes much better, albeit we paid for those extras - they are just too bloated and don't work well yet on this platform (my experience). We just have to see if ASUS continues to resolve the issues or do they just keep moving on to other stuff which I hope not. I figure there will be a new revisions of this board, maybe with a new name like CrossHair VII.
 
Rumour has it the X399 Asus board has a "really really safe boot" button :)

I'm sure Dan reads the forums as well , some people went as far as removing the power and battery from the motherboard for 2 days ...

I do and Kyle had already told me about having to do that previously, so I was aware of that "fix" which is bullshit. Even when that works, I have to conclude that something is seriously wrong for that "fix" to be necessary on any motherboard.
 
The BIOS (now v1201) still does not properly identify / enumerate the chipset-driven PCIe slots. This was reported in several forums by Timur Born and I now see it is the cause of all my problems too (described earlier). Slot PCIEX1_3 seems to be the worst in this regard. It is OBVIOUSLY A CH6 BIOS ERROR because two identical x1 cards that STILL WORK TOGETHER IN A 2008 RAMPAGE FORMULA will NOT WORK TOGETHER IN THE CH6. I understand PCIe lane allocation and bandwidth settings just fine, and have tried all combinations. The CH6 BIOS HAS A SIMPLE PROBLEM THEY NEED TO FIX.

Sorry, I used caps only in case Asus is reading - maybe they will actually address this. A ticket I raised with them got the usual "update all your drivers" response.
 
Saw this posted by elmor in overclock.net, not yet in Asus web:
http://www.mediafire.com/file/w8vh4h2v6v5iaae/CROSSHAIR-VI-HERO-ASUS-1401.zip

Beta UEFI thats why it's not on the official site yet, i got it from the ROG forums, i like it from my Corsair 3200CL16 memory i got it running at 3333MHz. As for the people struggling with their memory i would suggest changing the memory settings cause although my 3200CL16/18/18/36 the BIOS would select 16/15/15/36 so cold boots was problematic even tho it ran flawlessly at them latency settings once you got the system booted. I am still playing round with it but running at 3333MHz has laxed the settings a fair bit of 18/21/21/49. Not really had an issues with the board since release day my board has always been able to do 3200MHz except one official UEFI where for some whacky reason they set the CMD to T2 and could only hit 2933MHz but got the modified UEFI from the ROG forums that set it back to T1 and could do 3200MHz again. Oh and on release day i almost bricked my board cause i tried to upgrade to the new UEFI that the reviewers where sent and the instructions for crashfree was horrible but envertally recovered, so much so it put me off in using the windows update utility so now i only upgrade through the UEFI.

Overall tho the 1401 is a pretty good BIOS, but coming from someone that never really had issues with the board to begin with, experiences may vary.
 
Last edited:
this or something similar seems to have happened already to multiple users in this thread, even with recent bios versions ... so there is a good chance that your board is not dead-dead, more like having a problem with the bios ?

Possible.

like it has been stated by yourself and others, problems started when overclocking ... it seems the bios has trouble recovering, and needs a lot of time or even a secret handshake (different per bios version) to succeed

Yes, I'd agree and I'll go on to state that the need to jump through hoops beyond clearing the CMOS is total bullshit.

some of us have been able to make their board recover, using different methods: clear CMOS procedure, turning computer off and waiting, USB bios flashback procedure ...
this is probably way outside of the scope of a review, but it would be useful in the forum to know if you can get it back to work and how (for those of us that don't mind tinkering with their toys)

this board also has of lot of safety and recovery features, but they don't do much, or nothing at all:
the "safe boot" button, the "retry" button, the "clear CMOS" button ...
it feels like the bios corrupts itself or can't pass it's own internal checks, and fails POST ... something is wrong in that regard and I hope they can do something to solve it

I may well be able to get that board to recover, but I've got other fish to fry and other motherboards to grill. We can only allocate so much time to a given motherboard before it's time to move on.
 
  • Like
Reactions: noko
like this
Last edited:
1401 BIOS is working great so far and without the 9945 quirks. System is now perfectly stable with CPU at 4GHz and RAM at 3200. Not cold booting issues or anything. Looks like they are finally getting there!

Update: It's stable when running the AIDA64 stress test, but RealBench kills it. So not really that stable!

Did you try running things at stock with memory at 3200 ?
 
I have to say I was more than concerned after reading this review. I was sitting on some parts for a build including the ch6 for almost a month. I built everything on a test bench last week to check it out.
My build included an r7 1700, 2 x8gb Corsair lpx 3200, and a 960 evo m. 2. At first it would power up but not post. The led code was not in the manual. A quick search led me to bios 1201 which I used the "flashback method" and a thumbdrive to install. That went without a hitch and I was able to boot and install windows 10. So far I have been able to run the RAM at 3200 speeds and even have a mild overclock of 3.8 without any problems. I'll try the 1401 bios later in the week. I have had no problems with cold boot or resuming from s3.
 
Did you try running things at stock with memory at 3200 ?

Before trying the CPU OC, I was running everything at stock/default, except memory. That was using beta BIOS 9945. Now with version 1401, stock cpu and ram, I cannot get RealBench to complete the 15 minutes stress test... AIDA64 stress test can last a few hours without any issues...

I'm seriously considering going back to my Intel setup until a few more BIOS revisions on the CH6 and keep the AMD build as a secondary just for tinkering... At least things are stable under Prime95 and AIDA64. Since it's the Luxmark that is crashing it might be related to my SLI setup (or so I've read on a few threads).
 
Last edited:
Before trying the CPU OC, I was running everything at stock/default, except memory. That was using beta BIOS 9945. Now with version 1401, stock cpu and ram, I cannot get RealBench to complete the 15 minutes stress test... AIDA64 stress test can last a few hours without any issues...

I'm seriously considering going back to my Intel setup until a few more BIOS revisions on the CH6 and keep the AMD build as a secondary just for tinkering... At least things are stable under Prime95 and AIDA64. Since it's the Luxmark that is crashing it might be related to my SLI setup (or so I've read on a few threads).

That sucks :( .

I been semi stress testing my setup running WoW and recording at 50000 bitrate CBR at the same time. That was not a problem , not that this is a real stress test it does require some of the components to "play nice" .
.
 
The unfortunate reality of AM4 today is that if your having issues, it's most likely RAM related. The problem is, few people have six or seven RAM kits to test with. I do and its still a pain in the ass to find RAM that behaves. The Crosshair VI Hero is one board I happened to luck out on because Kyle found out what worked ahead of time. I also happened to have identical modules on hand to work with.

For people with limited budgets or available hardware, it's important to try and confirm modules that will work with a given UEFI and AGESA code version. I didn't change mine from what Kyle had found to work.
 
Last edited:
For people with limited budgets or available hardware, it's important to try and confirm modules that will work with a given UEFI and AGESA code version. I didn't change mine from what Kyle had found to work.

Sadly I don't have a Kyle to test my RAM beforehand! ;)

The memory I'm using is on the motherboard QVL but I guess this does not guarantee it will work with a specific UEFI/AGESA.

I'll make some more tests with only 2 sticks later this evening (with everything at stock/default). If this still fails I have another set also on the QVL list that I can try.

After that, it's back to my Intel setup since I'm not in the mood to throw more money at RAM just to see what sticks...
 
Last edited:
Looks like it. The memory I'm using is on the motherboard QVL but I guess this does not guarantee it will work with a specific UEFI/AGESA.

I'll make some more tests with only 2 sticks later this evening (with everything at stock/default). If this still fails I have another set also on the QVL list that I can try.

After that, it's back to my Intel setup since I'm not in the mood to throw more money at RAM just to see what sticks...
The unfortunate reality of AM4 today is that if your having issues, it's most likely RAM related. The problem is, few people have six or seven RAM kits to test with. I do and its still a pain in the ass to find RAM that behaves. The Crosshair VI Hero is one board I happened to luck out on because Kyle found out what worked ahead of time. I also happened to have identical modules on hand to work with.

For people with limited budgets or available hardware, it's important to try and confirm modules that will work with a given UEFI and AGESA code version. I didn't change mine from what Kyle had found to work.

This is so true and not many have the patience or know how to work with a finicky kit of ram or should I say a kit the Ryzen IMC just doesn't like. I do a lot of benching so playing in BIOS is part of my hobby and I have good Samsung ram. I purposely bought the cheapest set of DDR4 3000 ( Team Delta CL-16-18 Hynix MFR) I could find at my local CanadaComputers. I can see why people get frustrated, it still won't run if I use anything higher than the 2933 divider. There are some settings/timings that are made in the background and the CPU just refuses to co-operate. But I am also stubborn and I know that MFR WILL run faster it just won't on this board. Finally after much testing and using BCLK to go up from 2933 I have it running at 3200, system tested with IBT AVX version set to 12000MB(75%) ram usage. Your average user likely isn't going to go through the countless crash/reboots that I did to get that far but that's just what I do

1600x ibt .JPG
1600x hyn3200.jpeg
 
That sucks :( .
I been semi stress testing my setup running WoW and recording at 50000 bitrate CBR at the same time. That was not a problem , not that this is a real stress test it does require some of the components to "play nice" .
.

The strange thing is that it's also working fine when gaming, and other benchmark application are also completing fine. Oh well...
 
This is so true and not many have the patience or know how to work with a finicky kit of ram or should I say a kit the Ryzen IMC just doesn't like. I do a lot of benching so playing in BIOS is part of my hobby and I have good Samsung ram. I purposely bought the cheapest set of DDR4 3000 ( Team Delta CL-16-18 Hynix MFR) I could find at my local CanadaComputers. I can see why people get frustrated, it still won't run if I use anything higher than the 2933 divider. There are some settings/timings that are made in the background and the CPU just refuses to co-operate. But I am also stubborn and I know that MFR WILL run faster it just won't on this board. Finally after much testing and using BCLK to go up from 2933 I have it running at 3200, system tested with IBT AVX version set to 12000MB(75%) ram usage. Your average user likely isn't going to go through the countless crash/reboots that I did to get that far but that's just what I do

View attachment 27895 View attachment 27896

AM4 needs a few more months it seems for it to become less clunky. The C6H seems as if it is the worst of the bunch (to sensitive for certain settings) right now. But once you get it working it is fine (if you stop updating bios because your OC might not work on a newer version).
The strange thing is that it's also working fine when gaming, and other benchmark application are also completing fine. Oh well...

Have you seen video by Chew how he does the stress testing while running the Heaven benchmark in a loop ...
 
Let me put it this way, I've had really good experiences with most of ASUS' motherboards over the last ten years or more. I've encountered a couple of models I may not have liked for whatever reason, but they worked. The worst motherboard I've encountered up until the Crosshair VI Hero from ASUS was the Striker Extreme. That board was better than a lot of the fucked up 680i SLI reference boards, but it wasn't good either. The Crosshair VI Hero reminds me of that experience, but worse. It's easily the worst motherboard ASUS ever put in my hands. That's not saying its the worst motherboard ever, but it's definitely subpar for ASUS as a company.

They had just as long to get it right as GIGABYTE, but their effort wasn't as strong. That's puzzling to me but there it is.
 
Let me put it this way, I've had really good experiences with most of ASUS' motherboards over the last ten years or more. I've encountered a couple of models I may not have liked for whatever reason, but they worked. The worst motherboard I've encountered up until the Crosshair VI Hero from ASUS was the Striker Extreme. That board was better than a lot of the fucked up 680i SLI reference boards, but it wasn't good either. The Crosshair VI Hero reminds me of that experience, but worse. It's easily the worst motherboard ASUS ever put in my hands. That's not saying its the worst motherboard ever, but it's definitely subpar for ASUS as a company.

They had just as long to get it right as GIGABYTE, but their effort wasn't as strong. That's puzzling to me but there it is.
Hello

Any idea how far I could overclock if I had one of the motherboards you consider good?



3600_1.jpeg
 
Have you seen video by Chew how he does the stress testing while running the Heaven benchmark in a loop ...

I haven't seen the video. Do you have a link?

I'm down to 16GB of RAM and the results are the same (I've shuffled the sticks to make sure it was not cause by a bad one). The TridentZ is next so we'll see how it goes.

It's strange because RealBench lets me test all 32GB of RAM but when I only have 16GB installed, it tells me that there is not enough available memory to test and forces me to go with a smaller RAM preset.

Update: same result with the TridentZ (I'm getting the errors even when running the RAM at 2133). With my luck I'll get the same results with the MSI Z270 Xpower and the 7700K. :cool:
 
Last edited:
I haven't seen the video. Do you have a link?

I'm down to 16GB of RAM and the results are the same (I've shuffled the sticks to make sure it was not cause by a bad one). The TridentZ is next so we'll see how it goes.

It's strange because RealBench lets me test all 32GB of RAM but when I only have 16GB installed, it tells me that there is not enough available memory to test and forces me to go with a smaller RAM preset.

Update: same result with the TridentZ (I'm getting the errors even when running the RAM at 2133). With my luck I'll get the same results with the MSI Z270 Xpower and the 7700K. :cool:

If you're still at 4.0 I would suspect the CPU/cache as the culprit. Try using Prime 95 29.1 set to custom with 13000MB memory allocation. If you drop threads it's the memory. If BS/crash then cpu need more volts or less speed depending on thermals. For my 24/7 I'm pretty thorough for stability. This is the same CPU I posted before but running P95 custom for 11 hours

p95 1600x 11 hrs.JPG
 
I haven't seen the video. Do you have a link?

I'm down to 16GB of RAM and the results are the same (I've shuffled the sticks to make sure it was not cause by a bad one). The TridentZ is next so we'll see how it goes.

It's strange because RealBench lets me test all 32GB of RAM but when I only have 16GB installed, it tells me that there is not enough available memory to test and forces me to go with a smaller RAM preset.

Update: same result with the TridentZ (I'm getting the errors even when running the RAM at 2133). With my luck I'll get the same results with the MSI Z270 Xpower and the 7700K. :cool:



he shows a configuration with mixed ram kits ( both G.skill) on older bios 1001(or 1002).
 
  • Like
Reactions: noko
like this


he shows a configuration with mixed ram kits ( both G.skill) on older bios 1001(or 1002).


Thanks for the link!

On my end, upping the CPU voltage while still at default clock (offset +0.02500) and SOC voltage (offset +0.02500), and the RAM (3200) voltage (1.36v) Prime95 (blender) can run for hours without issues (so far!) with BIOS 1401. It still fail at completing the 15 minutes RealBench stress test... This thing is a mess!

Update : AMD published an official list of memory modules that are tested to run reliably at rated speed and F4-3200C14D-16GFX is on the list... How nice... (https://www.amd.com/system/files/2017-06/am4-motherboard-memory-support-list-en.pdf).
 
Last edited:
Back
Top