Latest BIOS for the ASRock B650E Steel Legend is so broken it turns your Board into a B650, breaks a USB Port and flips your RGB Colors permanently.

[GerKNG]

What ASRock just did is incredible...
The B650E Steel Legend got a new Bios Update (not a Beta or anything... a full release.)

Now the "funny" thing is:
it turns your B650E Board into a B650 Board, turns your third M.2 Slot into a Gen 3 slot, it permanently turned the 10Gb/s USB-C Port into a 5Gb/s Port and it flips Red and Green RGB (permanently as well!)
CMOS Resets do nothing and flashing it back through the bios is impossible since it just says that it did not found a valid file.
i had to use BIOS Flashback and i got it back to 3.08 but the RGB is still flipped and the USB Port is still 5Gb/s (maxing out at ~600MB/s)

HOW did they got this through QC? (if there is any...)
just take a look for yourself...

 

[chrcoluk]

Still remember their first REBAR bios that had no post screen with REBAR enabled (I think had no display at all). What saved it, was a cmos reset made it visible so could recover.

 

[eidairaman1]

What ASRock just did is incredible...
The B650E Steel Legend got a new Bios Update (not a Beta or anything... a full release.)

Now the "funny" thing is:
it turns your B650E Board into a B650 Board, turns your third M.2 Slot into a Gen 3 slot, it permanently turned the 10Gb/s USB-C Port into a 5Gb/s Port and it flips Red and Green RGB (permanently as well!)
CMOS Resets do nothing and flashing it back through the bios is impossible since it just says that it did not found a valid file.
i had to use BIOS Flashback and i got it back to 3.08 but the RGB is still flipped and the USB Port is still 5Gb/s (maxing out at ~600MB/s)

HOW did they got this through QC? (if there is any...)
just take a look for yourself...

View attachment 369687

View attachment 369688

View attachment 369690

View attachment 369692
View attachment 369693

Contact asrock with your findings

 

[Redwoodz]

This may sound like a stupid question to you but can I ask you what made you install a bios made for supporting a 9000 cpu on a 7000 series?

 

[natr0n]

welcome to bios adventures

 

[Vincero]

This may sound like a stupid question to you but can I ask you what made you install a bios made for supporting a 9000 cpu on a 7000 series?

There are other fixes with the updated BIOS and AGESA other than just supporting newer processors.
Things like the Sinkclose bug for example will only arrive with a BIOS update and required AMD AGESA 1.2.0.1 which would only have been in place with the BIOS version 3.08 according to the screenshot in OP's post.
Also, potentially improvements to RAM support, etc., may exist as well as potentially some other fixes they didn't bother to mention. Some vendors list a lot of changes and others list the bare minimum.... sometimes you just need to suck it and see...

 

[Dr. Dro]

I have a suspicion that they switched the BIOS names around when posting to their site. The BIOS they labeled "B650" may very well be the B650E... if this board has BIOS flashback capability as insurance, try flashing the one they posted for the B650 - might be the right BIOS there.

 

[tabascosauz]

What ASRock just did is incredible...
The B650E Steel Legend got a new Bios Update (not a Beta or anything... a full release.)

Now the "funny" thing is:
it turns your B650E Board into a B650 Board, turns your third M.2 Slot into a Gen 3 slot, it permanently turned the 10Gb/s USB-C Port into a 5Gb/s Port and it flips Red and Green RGB (permanently as well!)
CMOS Resets do nothing and flashing it back through the bios is impossible since it just says that it did not found a valid file.
i had to use BIOS Flashback and i got it back to 3.08 but the RGB is still flipped and the USB Port is still 5Gb/s (maxing out at ~600MB/s)

HOW did they got this through QC? (if there is any...)
just take a look for yourself...

I was just thinking about the new updates yesterday lol. Glad I did not entertain that thought further.

tbf, as to QC, I think I've had my fair share of gripes equally about Asus, Gigabyte, ASRock and MSI BIOSes. Lots of copy+paste happens whenever a new release comes up.

This may sound like a stupid question to you but can I ask you what made you install a bios made for supporting a 9000 cpu on a 7000 series?

I love how we can't win.
When we have inexplicable issues or bugs (e.g. with memory training), the internet tells us to update to a newer BIOS cause new AGESA magically just "gets better".
When we have unforgivable bugs on a new BIOS release, the internet ridicules us for not staying put and updating the BIOS.

It's a release BIOS, designed for the board in question, and it should work, period.

 

[A Computer Guy]

What ASRock just did is incredible...
The B650E Steel Legend got a new Bios Update (not a Beta or anything... a full release.)

Now the "funny" thing is:
it turns your B650E Board into a B650 Board, turns your third M.2 Slot into a Gen 3 slot, it permanently turned the 10Gb/s USB-C Port into a 5Gb/s Port and it flips Red and Green RGB (permanently as well!)
CMOS Resets do nothing and flashing it back through the bios is impossible since it just says that it did not found a valid file.

i had to use BIOS Flashback and i got it back to 3.08 but the RGB is still flipped

The RGB is surprising as that's a separate chip and separate firmware. For your motherboard make sure to install the latest RGB software. Open a command window and go to "C:\Program Files (x86)\ASRock Utility\ASRRGBLED\Bin" and run "WriteFW.bat". That should upgrade/reset the firmware for the RGB. This is especially useful in case it get's stuck with a bad setting. I had to do this when I was using RGB splitters and some RGB's got flipped probably from confused addressing.

and the USB Port is still 5Gb/s (maxing out at ~600MB/s)

I believe I saw in ASRock UEFI/BIOS speed controls for the USB ports. Look for them and see if you can flip it back to 10Gb/s.

HOW did they got this through QC? (if there is any...)
just take a look for yourself...

Before flashing did you make sure to turn off TPM?

 

[tabascosauz]

The RGB is surprising as that's a separate chip and separate firmware. For your motherboard make sure to install the latest RGB software. Open a command window and go to "C:\Program Files (x86)\ASRock Utility\ASRRGBLED\Bin" and run "WriteFW.bat". That should upgrade/reset the firmware for the RGB. This is especially useful in case it get's stuck with a bad setting. I had to do this when I was using RGB splitters and some RGB's got flipped probably from confused addressing.

Before flashing did you make sure to turn of TPM?

Is TPM off a new requirement now? Can't remember that being a thing on AM4, or AM5. I shopped around a little bit on bios versions for my PG-ITX, and I am too lazy to turn TPM off since the 5800X3D days.

 

[A Computer Guy]

Is TPM off a new requirement now? Can't remember that being a thing on AM4, or AM5. I shopped around a little bit on bios versions for my PG-ITX, and I am too lazy to turn TPM off since the 5800X3D days.

Not sure why for some reason ASRock has a big ass warning screen to remind the user to disable TPM when upgrading the UEFI/BIOS. It will re-enabled after the upgrade is complete.

 

[eidairaman1]

Not sure why for some reason ASRock has a big ass warning screen to remind the user to disable TPM when upgrading the UEFI/BIOS. It will re-enabled after the upgrade is complete.

To reduce any potential for flash problems

 

[Dr. Dro]

I'm not sure why TPM needs to be disabled to update the RGB controller firmware, BIOS I understand due to fTPM keys being potentially erased during flash process and Bitlocker being picky, but I wonder if there's a problem with AsRock's code signing in general, their timing configurator tool is blacklisted on Windows 11 24H2 as a known unsafe driver, including the latest 4.0.16

If it doesn't work, this might be the reason why and for some reason you're not getting the warning

 

[A Computer Guy]

It's a release BIOS, designed for the board in question, and it should work, period.

The hazards of a long term socket I suppose.

I'm not sure why TPM needs to be disabled to update the RGB controller firmware,

It doesn't for the RGB firmware just main UEFI/BIOS.

BIOS I understand due to fTPM keys being potentially erased during flash process and Bitlocker being picky, but I wonder if there's a problem with AsRock's code signing in general, their timing configurator tool

timing configurator tool? Where do you find that?

is blacklisted on Windows 11 24H2 as a known unsafe driver, including the latest 4.0.16

If it doesn't work, this might be the reason why and for some reason you're not getting the warning

 

[Vincero]

They should be warning about bitlocker also - they should really be advising people to disable it before flashing and enable it afterwards.

 

[Dr. Dro]

timing configurator tool? Where do you find that?

It's an XOC tool, like ASUS ROG MemTweakIt. My Intel KS club thread has links to both - doesn't need to be an AsRock board. Careful that editing memory timings with the OS live can corrupt things pretty badly.

 

[Launcestonian]

3.10 bios for my Asrock Steel Legend X670E board still does not have HVM (high voltage mode) for DRAM, yet 2 B650 boards I have here (Asus & MSI) have it & have for several months already but those boards are not high end either. Another weird surprise is that the 3.10 bios for my board is not up there now, it was for like barely 24hrs & then pulled down. 3.08 is the latest bios at time of writing this.

Running Zen timings v1.32 shows disturbances in how the app access the power table so incorrect data for voltages on 3 rails ( compared to what is set in bios) but no problem with 3.08 bios. Who the heck makes these bios & decides to put them up for public usage is giving Asrock a bad reputation.

I'm also puzzled why this Steel legend board does not support anything over 7600MT/s for 9000 series, yet my Gigabyte B650 Aorus Elite v1.0 has 8000MT/s support with over 10 different 32GB DIMM kits for 9000 series on the support page & this board came out in late 2022 like all the other 1st gen B650 boards.
Something is not right with Asrock when bios gets put up (non beta) & then pulled back down again literally in less than a day.

They should be warning about bitlocker also - they should really be advising people to disable it before flashing and enable it afterwards.

They do on my X670E steel legend board & also on other non Asrock AM5 boards I have here.

 

[A Computer Guy]

It's an XOC tool, like ASUS ROG MemTweakIt. My Intel KS club thread has links to both - doesn't need to be an AsRock board. Careful that editing memory timings with the OS live can corrupt things pretty badly.

Oh ok. Looks like a tool for Intel boards.

Well there's a UEFI/BIOS v3.10 (AGESA 1.2.0.2a) for LiveMixer but with tomorrow being a workday there is no way in hell I'm installing that tonight just to see what happens.

Running Zen timings v1.32 shows disturbances in how the app access the power table so incorrect data for voltages on 3 rails ( compared to what is set in bios) but no problem with 3.08 bios. Who the heck makes these bios & decides to put them up for public usage is giving Asrock a bad reputation.

I don't understand. Can you illustrate this issue for us?

 

[GerKNG]

This may sound like a stupid question to you but can I ask you what made you install a bios made for supporting a 9000 cpu on a 7000 series?

why should it be a problem to install the latest bios for my motherboard provided from the manufacturer.
and i plan to buy the 9800X3D.

I have a suspicion that they switched the BIOS names around when posting to their site. The BIOS they labeled "B650" may very well be the B650E... if this board has BIOS flashback capability as insurance, try flashing the one they posted for the B650 - might be the right BIOS there.

the B650 does not have 3.10.

 

[3DVCash]

This may sound like a stupid question to you but can I ask you what made you install a bios made for supporting a 9000 cpu on a 7000 series?

Not to pile on here, but all BIOS post AGESA 1.1.7.0 have supported 9000 series cpus.

This is 100% on Asrock.

 

[_roman_]

This may sound like a stupid question to you but can I ask you what made you install a bios made for supporting a 9000 cpu on a 7000 series?

There are different changes always in uefi updates. Sometimes it increases or decreases stability, security and supported features.

--

If you do not understand encryption and have ways to decrypt, I recommend not using encryption at all. Regardless if it's bitlocker or something else. Someone should know how his box boots in the last details.

 

[Redwoodz]

why should it be a problem to install the latest bios for my motherboard provided from the manufacturer.
and i plan to buy the 9800X3D.



the B650 does not have 3.10.

I just wanted to know if you were having issues beforehand. Maybe you have a faulty board, idk. I have this exact board and I'm running 3.08. Maybe I'll install it and see what happens, but I have 9000 series cpu.

 

[GerKNG]

I just wanted to know if you were having issues beforehand. Maybe you have a faulty board, idk. I have this exact board and I'm running 3.08. Personally I do not change bios if the system is working normally.

i upgraded to every release bios and over the last couple months i haven't had any issues. all of this happened with 3.10.

 

[Redwoodz]

i upgraded to every release bios and over the last couple months i haven't had any issues. all of this happened with 3.10.

You sure you didn't install B650 bios by mistake? They are right next to each other on the website and you have something blurred out in your screenshot of downloads.

 

[Launcestonian]

You sure you didn't install B650 bios by mistake? They are right next to each other on the website and you have something blurred out in your screenshot of downloa

i upgraded to every release bios and over the last couple months i haven't had any issues. all of this happened with 3.10.

Likewise, your not alone with Asrock AM5 high end boards doing this, I'm going to revert back to 3.08. Another annoying issue is Aida64 Cache & benchmark run now shows an additional 10ns increase in latency with exact same RAM settings as before in 3.08 bios. Someone at Asrock mucked up big time here imo.