X370 Gaming X Bios update. Kills Systems.

Final - Final update.

The 5.20 Downgrade bios, is slightly different in a good way.

The Disk Access LED never worked right for me with Raid, is now working as expected. Before was on all the time, now it does the blinky-Blinky like I expect it should.

Also Boinc now flags the CPU at 100% steady, prior it was hitting 100% and then dropping to 98% and then bouncing back up again.

So there is something different in the Downgrade bios, and it seems by different I mean better.

Final Update:

So I wound up Spamming pretty much every option in every support region, no responses after a few days.

Then I spammed every mail address i could find, webmaster, sales etc.. for every region and re-spammed support .

Next day, I was contacted by John from ASRock Support. He gave me a BIOS that allowed me to revert to 5.20

The mail on how to do, seemed that it would be complicated and difficult and dangerous, it was not.

The reversion ran a bit from within windows, then the system rebooted and UFEI Bios did its thing and self flashed to 5.20

All good.

However since the raid was broken by 5.40 I had some issues there. The Bios said there is no raid array but the disks still had a signature on them that they were part of an array.

No big worry, I've been down this rabbit hole before. You just need to low-level the disk and all is good. Except the Hitachi tool though able to wipe the disk doesn't apparently touch the area of the disk where the bits are set that indicate the disk is part of an array.

The fix. I fired up Dban, and Wiped the disks, I did not let it run fully just long enough that I felt confident the areas in question got zero filled. I then set the disks back to AHCI and initialized them under windows as basic disks.

Reboot and reconfigure as RAID and bam I could now initalize the disks and make them part of a new array.

A quick boot then off my cloned disk (before all this brouhaha started.) and reclone back to to my Raid1 Array. 24 hours later... everything is back and working as normal.

Raid... Working.
USB attacked disk Dock (attached via ASRock usb 3.1 Gen2 card). Working

Made 2 more new clones of the boot drive (Just in case.) They'll eventually be rotating boot backups, but for now...

And Stability testing is underway.

So color this solved. 5.40 Bios is problematic, so I'll wait, until the Ryzen3's come out and then flash at that time with whatever bios is out then, and if it doesn't work I have a backout for the thing you apparently cannot backout.

John, Kudos to you.

Harleyevo wrote: Os shouldn't matter since it is happening at the bios where the array first shows broken.

It would have taken less time to say.

Nevermind.

What's everyone running for a processor and OS?

I will pass this info along to my contacts at ASRock.

The new version probably contains support for Ryzen 3k and as usual, something probably got messed up. AMD still has a ways to go with their base code (AGESA) programming and troubleshooting. A quick look around shows issues like this popping up every time AM4 has gotten a generation jump/major update, here and on other manufacturer forums.

I am not defending anyone here or placing blame, just recommending that it is typically best to avoid the first and probably second BIOS release after a major update unless you absolutely need it (ie. you need the CPU support). Generally by the 3rd round of BIOS releases after a generation jump the issues and kinks have mostly been ironed out.

If you take a look around the forums here you will see quite a few threads pressuring ASRock to release updates for the "latest AGESA" and if there are 2 threads on the forums about it you can be sure there have been hundreds of support tickets sent. It's a catch 22, hold off on release till you are sure all is stable and functional, possibly missing a CPU launch and lose a lot of users who feel cheated because X or Y manufacturer has already released updates or rush them out the door and annoy other users with bugs, broken features etc. Personally I prefer the slower release path. This is typically not the case with intel boards so I suspect AMD is still ironing out the kinks in their development cycle and how they coordinate with board partners.