Any ASUS Prime X370 Pro owners out there ... besides me??

[gupsterg]

As Johan45 states.

Pearl of wisdom from The Stilt for Ryzen owners .

The newer FWs used in these bioses are still betas (as are the bioses themselves), so few shenanigans are no surprise really. The newer FWs are also completely different animals to the ones used in earlier AGESA versions, so don't expect them to behave even remotely the same.

Also, there are no "straps" in these CPUs. There are several timings which AGESA tries to keep constant, independent of the MEMCLK. These timings only change to reflect the change required in the resulting cycle time. For example if a certain timing defaults to 8 CLKs at 2133MHz MEMCLK (7.5ns cycle time), at 3200MHz MEMCLK the same timing must be configured to 12 CLKs for the cycle time to remain the same (7.5ns).

I concur Woomack that the X CPU is more likely to hit ~4.0GHz and +. As it is higher leakage CPU, so tends to scale better. So for a bencher or someone who has it set in their mind they what ~4.0GHz then X CPU is more likely better purchase at a premium of cost. When taking the cost into account for everyday use I reckon it's not as worthwhile going for a X CPU. Especially when if you don't need the Wraith Spire you can claw some $ back on purchase. Then the gap between OC/price performance grows again between the non X vs X CPU.

I have seen also several posts on various forums where someone bought an X CPU and when it didn't OC as they expected the devastation is greater.

 

[Keith Myers]

Learned something

Yes the new CHVI BIOS based on the new AGESA have added 2800,3066,3333,3466,3600,38xx and 4000 dividers. They're not straps Keith. A strap is used to set the BCLK and ram at a different speed than the other PCI,DMI busses. X99 platform has straps so you can run CPU/mem at bclk 125/166 without affecting the rest of the system

Well I learned something new today. Never knew the proper definition of what a "strap" was I see. I would assume then the proper term would be a memory divider.

 

[Johan45]

You would be assuming correctly. Look at the read out in CPUz it's expressed as a ratio

 

[Bluefalcon13]

Took a look at downloads page for our board. Was looking for a new bios. No new bios there, but new audio drivers and new version of AI suite. Just a heads up.

 

[Keith Myers]

Since I don't even bother loading audio drivers ..... ho hum. And I still have horrible memories of my one and only install of AI Suite. NEVER again. Still awaiting the "official" AGESA 1.0.0.6 F/W BIOS update for us Prime users.

 

[gupsterg]

Just as a "heads up".

Several of the the "Beta" UEFIs that Elmor posted on OCN for C6H became "Official". When compared, the "Official" were byte for byte / checksum, etc identical as the "Beta" .

 

[Keith Myers]

Well, lets hope that history repeats itself and we get an official AGESA 1.0.0.6 BIOS next week then, since we seem to be about a week behind always the Crosshair VI Hero camp.

 

[Woomack]

Whatever it's called, users are in most cases beta testers. In last years it's pretty common. If official releases were so great and fully tested then there wouldn't be that CHVI BIOS brick issue. Most mobo manufacturers make the same. At least ASUS is trying to fix it. Gigabyte would tell you that all is fine and release new PCB revision 3 weeks later.
Simply if you have any beta BIOS version which is stable for you and you see no issues then just use it.

 

[Keith Myers]

You take a chance of not being able to go back to earlier BIOS

Whatever it's called, users are in most cases beta testers. In last years it's pretty common. If official releases were so great and fully tested then there wouldn't be that CHVI BIOS brick issue. Most mobo manufacturers make the same. At least ASUS is trying to fix it. Gigabyte would tell you that all is fine and release new PCB revision 3 weeks later.
Simply if you have any beta BIOS version which is stable for you and you see no issues then just use it.

I am not willing to risk getting stuck on a beta BIOS that doesn't work as well as the one I am currently on. Reports that I have read is that there were no substantial improvements in the 0612 BIOS from the 0604 BIOS that I am on. I have also read that it is very difficult to go back to a prior BIOS if you try one of the betas. And we are in a very precarious situation where we do not have any BIOS recovery feature from a dedicated USB port like the Hero has.

I am encouraged that there seem to be some of the promised AGESA 1.0.0.6 features in the beta 0801 BIOS. I am a little leery of downloading it from some German user forum and not from an official ASUS site or even a dedicated ASUS Prime forum.

 

[Johan45]

Usually there are certain releases that for what ever reason ASUS/AMD doesn't want you to go back likely because of certain changes they feel are necessary. That doesn't always make them right but most of the time.....
There are ways to revert your BIOS if you really want to but for the masses, at least from my experience/observations the progression has been positive toward better compatibility.

 

[LRG5]

Asus Bios 604 and 612 not only beta's are locked or coded and you can not down grade. 5 or more boards bricked trying . Asus gave no warning at all that they had done
this, on the site or text file with the bios. And that for me was the last straw and I will not buy any more Asus MB. Asus did post all kind warnings about the right way to update bios not to brick. bottom line $$$$$ not service.

Oh , no I did not brick my Prime.

 

[Keith Myers]

Asus Bios 604 and 612 not only beta's are locked or coded and you can not down grade. 5 or more boards bricked trying . Asus gave no warning at all that they had done
this, on the site or text file with the bios. And that for me was the last straw and I will not buy any more Asus MB. Asus did post all kind warnings about the right way to update bios not to brick. bottom line $$$$$ not service.

Oh , no I did not brick my Prime.

Yes, I read about the bricked boards trying to go back to stable, working BIOS revisions like 0515. That is why I am hesitant about trying 0801. I have enough patience to just wait it out. I am stable at 3.85 Ghz and 3200 MT/s memory speed. Of course I would like to investigate whether my 3600 MT/s memory can clock higher for the improvement in the Infinity Fabric. I am hoping that I can get to the 3466 memory divider with minimal issues with the official AGESA 1.0.0.6 BIOS release expected this month.

 

[Keith Myers]

FYI, there is another beta BIOS showing up on the ASUS download site. You have to hunt for it again as usual under all the various Windows platforms. Don't know why ASUS has such a hard time putting the BIOS under every platform it is used on. BIOS is 0803 and still marked as beta.

 

[LRG5]

With that new Beta Bios 803 is there info to what bug fixes, option added, option deleted and is it locking.
To me that how a first class company should be, inform it's customers. Let the customer decide. Not just post the
update because you sign something with AMD saying you would.

 

[Keith Myers]

With that new Beta Bios 803 is there info to what bug fixes, option added, option deleted and is it locking.
To me that how a first class company should be, inform it's customers. Let the customer decide. Not just post the
update because you sign something with AMD saying you would.

All it says is BIOS with AGESA 1.0.0.6. And the preferred method of updating. However, I am still amazed they offer to update the BIOS from within Windows. A recipe for disaster as so many have discovered.

 

[LRG5]

You are right about that, flash from Bios. The AGESE 1.0.0.6 update would have the memory tables you want, but I would check around and see if it locks your Bios. The final 803 should be out soon, just watch the Vcore levels at stock.

 

[Keith Myers]

You are right about that, flash from Bios. The AGESE 1.0.0.6 update would have the memory tables you want, but I would check around and see if it locks your Bios. The final 803 should be out soon, just watch the Vcore levels at stock.

Don't know if the latest bricked board was from trying to go back or go forward to 0801 or 0803. This was from a post about 2 hours old on the OCN "my-experience-with-the-asus-prime-x370-pro" official Prime thread. Don't know if the beta BIOS is locked or not. I suspect it is since it adds so many new AGESA 1.0.0.6 memory timing parameters that the older AGESA 1.0.0.4a BIOS didn't have. And I keep reading that parts of a BIOS are considered core code and is inviolate, so you can't just flash back because the newer BIOS adds memory address blocks to the core code that wasn't there in an older BIOS. I don't know the technical terms for what I am trying to describe.

 

[Woomack]

Big problem on Prime is that BIOS flashback is not working. Clearly ASUS has saved on some things while building X370-Pro. If it was CHVI then force flashing would work in any way.
With Biostar I have no issues at all ... there was no BIOS for 2 months and they don't know when will be available

 

[Johan45]

Don't know if the latest bricked board was from trying to go back or go forward to 0801 or 0803. This was from a post about 2 hours old on the OCN "my-experience-with-the-asus-prime-x370-pro" official Prime thread. Don't know if the beta BIOS is locked or not. I suspect it is since it adds so many new AGESA 1.0.0.6 memory timing parameters that the older AGESA 1.0.0.4a BIOS didn't have. And I keep reading that parts of a BIOS are considered core code and is inviolate, so you can't just flash back because the newer BIOS adds memory address blocks to the core code that wasn't there in an older BIOS. I don't know the technical terms for what I am trying to describe.

This "BIOS locking" as you call it is not something new. I have had many boards that wouldn't let you flask backward. This is IMO mostly for usability reasons. The newer BIOS as I said before introduce things like new AGESA, or memory compatibility , fix USB issues etc... typically the company doesn't want you reverting since all those fixes are gone. The issue is when you get a "bad" release that wasn't tested enough or in some odd combinations will cause severe side effects. It's impossible for a company to test the thousands of HW combinations that are possible.
Re-writing the BIOS back to an older version removes those tables. I have had my CHVI on the latest 9943, 1201, and neither one of those wants you to go backward. But I have reverted to 902, 1101 and the BIOS is just like it was when using that version. All the new settings and mem dividers are gone.

As for the bricking many of those things happen from windows or even from a BIOS flash when settings(memory) aren't really stable and you get a "glitch" during the writing process. That's why it's always recommended that you hit F5 defaults or eve clear the CMOS prior to flashing. This makes the environment as stable as possible. I think the amount of "bricks" is high this time around simply because of the amount of flashing that's being done ( blame AMD/ASUS or whoever for the immaturity of the platform) and many times by inexperienced users.

 

[mackerel]

I just put 0803 beta bios on mine, and I can sum it up in one word: wow!

I still have Corsair LPX3000 2x4GB in mine, and across even Intel systems, I've not found it to be such a compatible set of ram. It runs at rated 3000 on my MSI Z170A board, but it will only do 2933 on an Asus Z170I. Over to Ryzen, with early bios and DOCP, 2400 was the limit which was obviously disappointing. 1.0.0.4 bought it up to 2666, but 2933 remained out of reach. With 0803 bios and 1.0.0.6, I started ramping up the settings once more. All I'm doing is turning on DOCP (XMP) and setting the running speed. 2666, no problem. 2800 was ok too. So already an improvement. How about 2933... yes, it booted! Already I'm happy as this is the highest speed selectable without going past ram rating... I can't stop here, can I? 3066, booted I'm leaving P95 blend on it for now while I do some other things.

Take this from the perspective of a general enthusiast, who wants to get the speed and performance of their ram without detailed fiddling of multiple settings. If this is representative of other fussy ram sticks, we could consider Ryzen platform as reaching mainstream maturity.



Edit: 3066 was unstable. 2933 looks good so far. While the more determined might fiddle with settings to get 3066 stable, I'm still looking at it from the perspective of getting it going with relative ease.