ASUS releases bios updates for x570 boards

erek

[H]F Junkie
Joined
Dec 19, 2005
Messages
10,785
(no way i'm trying this until i hear about a truly stable release... been a nightmare on these bioses... see: https://hardforum.com/threads/asus-...v2-pi-1-1-0-0-patch-b.2002763/post-1044788792 )

BIOS
Version 2402 Beta Version
2020/11/04 20.13 MBytes
ROG CROSSHAIR VIII HERO (WI-FI) BIOS 2402
"1. Update AMD AM4 AGESA V2 PI 1.1.0.0 Patch C
2. Improve system stability
3. Improve DRAM compatibility

Example: https://dlcdnets.asus.com/pub/ASUS/...FI/ROG-CROSSHAIR-VIII-HERO-WIFI-ASUS-2402.ZIP

d3athf1sh
 
I also did some looking at Reddit and the 1.1.0.0 AGESA also seems unstable on some peoples X570 Asrock boards. I'm sticking with my older one for now.
 
  • Like
Reactions: erek
like this
I also did some looking at Reddit and the 1.1.0.0 AGESA also seems unstable on some peoples X570 Asrock boards. I'm sticking with my older one for now.
I'm running the latest AGESA (patch C) on my Asrock x570 Taichi with no problems. Stable and smooth with my 5800X.
 
"After installing 2311, coming from 1302, I had nothing but BSODs on startup, finally followed by Windows telling me my boot drive was corrupt. Had to wipe it clean and reinstall all my software which took almost a week to get everything dialed in again.

I'm afraid to touch a beta BIOS from the same source, thinking my motherboard could literally burst into flames."

https://rog.asus.com/forum/showthre...osshair-VIII-Hero-(WI-FI)&p=818833#post818833
 
I'm running the latest AGESA (patch C) on my Asrock x570 Taichi with no problems. Stable and smooth with my 5800X.
The accounts I heard were from 3000 series users, I'll likely update myself at some point in the future. Glad to hear it's stable with a 5000 series, I'll likely get one... eventually....
 
The accounts I heard were from 3000 series users, I'll likely update myself at some point in the future. Glad to hear it's stable with a 5000 series, I'll likely get one... eventually....
any links?

i'm tempted to try it but i'm on a 3950X .. :(
 
For what its worth, i'm on 3.40 and my system is absolutely stable.
https://www.reddit.com/r/ASRock/comments/jj1b6q/x570_taichi_bios_v360_released_today/

It's hard to find too many things dedicated to the X570 Taichi, and, of course, there's confirmation bias here. People with perfectly stable systems don't have any reason to make a comment about it.

Just tried the Patch C Beta BIOS and it’s unstable for me. Apparently have to manually set the DRAM voltage to 1.36 to get it stable which is unacceptable to me. Also didn’t solve some other problems i was hoping for. Reverting back to 2206
 
The accounts I heard were from 3000 series users, I'll likely update myself at some point in the future. Glad to hear it's stable with a 5000 series, I'll likely get one... eventually....
Until I get a 5000 series CPU mine is staying at 2.80, the asrock redditt threads have scared me off flashing for now.
 
  • Like
Reactions: erek
like this
Just tried the Patch C Beta BIOS and it’s unstable for me. Apparently have to manually set the DRAM voltage to 1.36 to get it stable which is unacceptable to me. Also didn’t solve some other problems i was hoping for. Reverting back to 2206
Funny you should say that. I just ran memtest86 and I lost stability. I did this after noticing my Aida copy score was way less than on the old BIOS 1201. Everything else was pretty close, but copy on BIOS was like approx. 52000 on 2311 where it was approx. 60000 on 1201. I've tested various power settings, upped SOC, etc. The higher I make SOC, the faster the errors show up. I put SOC back on auto, which runs it around 1.02 I think and had the fewest errors so far. So now, I'm testing with auto SOC and DRAM voltage at 1.36 and all power setting at default. We'll see if it finishes the test in 10 hours.

I'm curious, why would you refuse to go 1.36v on the RAM? Seems pretty mild to me.
 
Funny you should say that. I just ran memtest86 and I lost stability. I did this after noticing my Aida copy score was way less than on the old BIOS 1201. Everything else was pretty close, but copy on BIOS was like approx. 52000 on 2311 where it was approx. 60000 on 1201. I've tested various power settings, upped SOC, etc. The higher I make SOC, the faster the errors show up. I put SOC back on auto, which runs it around 1.02 I think and had the fewest errors so far. So now, I'm testing with auto SOC and DRAM voltage at 1.36 and all power setting at default. We'll see if it finishes the test in 10 hours.

I'm curious, why would you refuse to go 1.36v on the RAM? Seems pretty mild to me.
why should i have to manually intervene at totally stock XMP settings for stability?
seems unacceptable
 
why should i have to manually intervene at totally stock XMP settings for stability?
seems unacceptable
While agreeing with your point, I gotta say--just tongue-in-cheek--that that question isn't very [H].
 
Be glad that you didn't have an X370 board, I spent 3 years manually intervening to keep things stable. Every single BIOS update I had to adjust something.

It never *once* was stable at XMP default.

Heard Ryzen 1 was rough
 
why should i have to manually intervene at totally stock XMP settings for stability?
seems unacceptable
Well, I guess that's a good point. My B-Die RAM wouldn't even boot at XMP settings, but then again, it's not on the QVL. It doesn't matter anyway, I got errors when I tested, just not as many and they took longer to show up. They actually put out a newer Patch C BIOS today. However, I just went back to 2206 which I had never actually used. I got the best STOCK CB20 score I've ever had after applying my settings from BIOS 1201 to BIOS 2206. 9225 multi and 530 single core. I'll run memtest overnight and see what happens.
 

Attachments

  • 3950x_CB20_BIOS_2206_64GB_11_09_20.jpg
    3950x_CB20_BIOS_2206_64GB_11_09_20.jpg
    299.7 KB · Views: 0
Heard Ryzen 1 was rough
I used to manually overclock and I enjoy tinkering around in the BIOS, so while it was annoying at times, getting that stable setup felt great. I still managed to keep up time to "only reboot on windows updates."

I could have left things on an earlier BIOS and just let it be, but that's not as fun for me. 😅 Always wanted to see if a newer AGESA would finally get me better RAM speed.
Turned out the problem was the motherboard afterall, once I switched to X570 I had the same processor and the same RAM and I just hit that XMP button once and it worked great and I did a few days of Prime95 memory torture without an error.

...so of course I had to fiddle with it after that. 😉
 
Well, I guess that's a good point. My B-Die RAM wouldn't even boot at XMP settings, but then again, it's not on the QVL. It doesn't matter anyway, I got errors when I tested, just not as many and they took longer to show up. They actually put out a newer Patch C BIOS today. However, I just went back to 2206 which I had never actually used. I got the best STOCK CB20 score I've ever had after applying my settings from BIOS 1201 to BIOS 2206. 9225 multi and 530 single core. I'll run memtest overnight and see what happens.
Well, Prime 95 large FFTs failed after about 2 hours, so back to the drawing board. Adding a little SOC, up to 1.0625 now to see what happens....
 
Swappable BIOS chips would simplify, for the consumer, these issues. It seems like you could even produce a drop-in 32MB BIOS chip. (I'm not sure if the pin-count would allow it?)
 
With the ASRock X570 Steel Legend and so far with BIOS's and my 3700x w/ 2 x 16GB Hyper X Predators 3200@3600

2.60 Great
2.80 Hot Trash
3.00 Great
3.30 Hot Trash
3.31 Hot Trash
 
I updated to the latest beta on my Prime X570-Pro and it works fine with my 5600x.
 
Using Patch B since I got the CPU at launch. Everything is stable, 3200mhz memory at 3600mhz. I am not touching it unless something is a game changer.
 
My b550m/csm also has patch c, but I haven't tested it. It's more or less stable with the original bios, so I'm hesitant to upgrade.
 
(no way i'm trying this until i hear about a truly stable release... been a nightmare on these bioses... see: https://hardforum.com/threads/asus-...v2-pi-1-1-0-0-patch-b.2002763/post-1044788792 )

BIOS
Version 2402 Beta Version
2020/11/04 20.13 MBytes
ROG CROSSHAIR VIII HERO (WI-FI) BIOS 2402
"1. Update AMD AM4 AGESA V2 PI 1.1.0.0 Patch C
2. Improve system stability
3. Improve DRAM compatibility

Example: https://dlcdnets.asus.com/pub/ASUS/...FI/ROG-CROSSHAIR-VIII-HERO-WIFI-ASUS-2402.ZIP

d3athf1sh
BIOS
2. Improve system stability
3. Improve DRAM compatibility

This is pretty much the patch notes for every ASUS motherboard BIOS release in the last ten years. It's stupid.
 
With the ASRock X570 Steel Legend and so far with BIOS's and my 3700x w/ 2 x 16GB Hyper X Predators 3200@3600

2.60 Great
2.80 Hot Trash
3.00 Great
3.30 Hot Trash
3.31 Hot Trash
*sigh*

Still a year off from building a new system, but I really want to be able to go AMD, but it's weak support like this that still gives me pause. Think I will wait for the next socket/chipset, or give Intel time (more? fuck) to get their shit together.
 
  • Like
Reactions: ncjoe
like this
*sigh*

Still a year off from building a new system, but I really want to be able to go AMD, but it's weak support like this that still gives me pause. Think I will wait for the next socket/chipset, or give Intel time (more? fuck) to get their shit together.
Not my experience, FWIW. I'm running a few Gigabyte Aorus somethings...and they've been rock-solid. I've updated BIOSes...because, and still rock-solid.

If you decide to skip this generation of AMD, you're missing out on a pretty great tech. But, your call. (Who knows? Maybe Intel will pull one out with their next generation...)
 
How is ASUS with their AMD boards?

Looking at the TUF Gaming X570-Pro Wifi 6

This is my first AMD CPU since my Athlon 64 3200+ (Venice)

I'd like the VIII Dark Hero cause having a chipset fan feels very 2003-ish, but trying to bring build cost down where I can.
 
Last edited:
*sigh*

Still a year off from building a new system, but I really want to be able to go AMD, but it's weak support like this that still gives me pause. Think I will wait for the next socket/chipset, or give Intel time (more? fuck) to get their shit together.
Just because ASRock had some misses with a few BIOS updates .. that does not equate "weak support" .. The 3.30, and 3.31 are mainly for 5000 series Ryzen's of which I do not run .. so I don't know if I could run my memory past XMP with those if I had a 5000 series or not ..but no go with my 3700x.

I HIGHLY recommend AMD ..and even more so with their new CPU's out now .. Best of Both Worlds... Best all around gaming-content creation, PCIe 4.0 for graphics & storage.. Going Intel for a new build just doesn't make sense right now.
 
Ive got an asrock x570 and my 5600x is running like butter on this C patch

But thats on 32GB 3600 c18

I cant seem to get it stable at c16 but again im populating all 4 dimm slots so im probably electrically unstable and not timings.
 
  • Like
Reactions: erek
like this
Just because ASRock had some misses with a few BIOS updates .. that does not equate "weak support" .. The 3.30, and 3.31 are mainly for 5000 series Ryzen's of which I do not run .. so I don't know if I could run my memory past XMP with those if I had a 5000 series or not ..but no go with my 3700x.

I HIGHLY recommend AMD ..and even more so with their new CPU's out now .. Best of Both Worlds... Best all around gaming-content creation, PCIe 4.0 for graphics & storage.. Going Intel for a new build just doesn't make sense right now.
I don't think it is ASRock or any mobo manufacturer, I think it's the microcode from AMD, the Agesa part, and sticking with the same socket for so long. Keeping it all compatible is difficult and has played a part in the unstable releases.

I still think it could be a good system, but I value stability. I don't need to replace the i7-6850k just yet, still does well in games/porn. It's only 4 years old and I got 7 out of the i7-920, so figure another year and a half, then get at least an 8 core cpu. Still want to spend about $550 max on the cpu. The i7-920 was like $285 iirc. Amazing for getting 7 years out of it, and that mobo still works, but is nearly 12 years old now and no spectre fixes came out for it.

But yeah, will be fun to upgrade at some point.
 
I just installed a Strix x570-e last night for my 5600x. Would not post for shit. Used Biosflashback feature to the latest C agesa and system runs great!
 
I don't think it is ASRock or any mobo manufacturer, I think it's the microcode from AMD, the Agesa part, and sticking with the same socket for so long. Keeping it all compatible is difficult and has played a part in the unstable releases.

I still think it could be a good system, but I value stability. I don't need to replace the i7-6850k just yet, still does well in games/porn. It's only 4 years old and I got 7 out of the i7-920, so figure another year and a half, then get at least an 8 core cpu. Still want to spend about $550 max on the cpu. The i7-920 was like $285 iirc. Amazing for getting 7 years out of it, and that mobo still works, but is nearly 12 years old now and no spectre fixes came out for it.

But yeah, will be fun to upgrade at some point.
I think you're right about the agesa code being the root issue and it being due to them using the same socket for multiple generations but it's still up to the board makers to properly integrate the updates and some have done worse than others(especially when they've run out of space). Unfortunately it has been an issue with each of the ryzen launches but to be fair any real problems have been sorted out and often the updates even included some minor performance bumps along the way.

If you don't want to deal with these things it's probably best to wait until there's a good BIOS out for whatever board you're going to use or go a different route. My i7 920 lasted me even longer than yours and that was a great CPU but personally I'm glad we're back to a point where upgrading every few years is at least worth considering.
 
I think you're right about the agesa code being the root issue and it being due to them using the same socket for multiple generations but it's still up to the board makers to properly integrate the updates and some have done worse than others(especially when they've run out of space). Unfortunately it has been an issue with each of the ryzen launches but to be fair any real problems have been sorted out and often the updates even included some minor performance bumps along the way.

If you don't want to deal with these things it's probably best to wait until there's a good BIOS out for whatever board you're going to use or go a different route. My i7 920 lasted me even longer than yours and that was a great CPU but personally I'm glad we're back to a point where upgrading every few years is at least worth considering.

Version 2502
2020/11/13 20.13 MBytes
ROG CROSSHAIR VIII HERO (WI-FI) BIOS 2502
"1. Update AMD AM4 AGESA V2 PI 1.1.0.0 Patch C
2. Improve system performance
 
Version 2502
2020/11/13 20.13 MBytes
ROG CROSSHAIR VIII HERO (WI-FI) BIOS 2502
"1. Update AMD AM4 AGESA V2 PI 1.1.0.0 Patch C
2. Improve system performance
2402 with XMP on tanked my 5800x scores on this board. CB20 was between 4200-5200 MC and 520's SC. Sequential runs seemed to get worse. WHEA errors filled the Application Logs in WIN10.

XMP off (all default Bios) my back to back runs of CB all netted 5992-6060 MC and 623 SC.

Updated to 2502 and Im seeing consistent 5998-6016 MC and 620's SC with XMP enabled. Seems they figured something out.
 
Version 2702
2020/11/24 20.13 MBytes
ROG CROSSHAIR VIII HERO (WI-FI) BIOS 2702
1. Improve system performance
Before running the USB BIOS Flashback tool, please rename the BIOS file (C8HW.CAP) using BIOSRenamer.
 
Back
Top