Question Has anyone updated their AMD X570 board to the latest AMD ComboPI1.0.0.3abba BIOS?

jon96789

Great
Aug 17, 2019
145
17
95
1
Watching this video, JayZTwoCents updated his MSi X570 MEG GodLike motherboard and ran into some major issues (the board keeps getting CPU errors, runs at a persistent 1.5 volts along with system crashes). Even after reverting to the previous BIOS, the problems persisted on his system.

I have the MSi X570 MPG Gaming Carbon Pro Wi-Fi board and the latest AMD PI1.0.0.3abba BIOS (7B93V13) update was released on 09/18/19.

The video is making think twice about updating the BIOS. I had already updated my BIOS to the previous version (7B93V12) and ran into a lot of issues with CPU thermals as well as extremely high voltages, it also did not accept any changes in Power Options in the Control Panel... I was able to revert to 7B93V11) which restored my system to a normal running condition. If anyone else has updated their BIOS on their X570 boards, please post your experiences... Thanks.
 
Why would X570 run different from B450 with a 3700X?

I updated to ABBA....I don't get any crashes or bad behaviour. In light "bursty" loads (browsing around the desktop, opening closing apps, reading email and browsing web) core voltage (SVI2) does peak around 1.48 or 1.49 V as the processor is boosting to 3.75-4.4 Gig repeatedly. AMD alerted us to this and stress that it is NORMAL for the processor to ask for up to 1.5 V when boosting.

But in heavy workloads the voltage drops...and drops. In Prime95, small FFT, AVX on, it drops to 1.287-1.295 V. But core frequency is down around 4.1G too; still perfectly stable. But hot... P95, small FFT with AVX on is just a power virus.

Also, when i just sit idle, no mousing at all, volts drops to right about 1.0 V. After a no-stress, light, "bursty", session average Vcore in HWInfo is only 1.3 V or so.

I should prolly watch the vid, but he's prolly just being Jay2Cents. setting you up for another vid where explains all he did wrong. Actually, not bad strategy to both keep viewers coming back and getting a point across about how to do it right.
 
Jul 25, 2019
61
1
45
1
Watching this video, JayZTwoCents updated his MSi X570 MEG GodLike motherboard and ran into some major issues (the board keeps getting CPU errors, runs at a persistent 1.5 volts along with system crashes). Even after reverting to the previous BIOS, the problems persisted on his system.

I have the MSi X570 MPG Gaming Carbon Pro Wi-Fi board and the latest AMD PI1.0.0.3abba BIOS (7B93V13) update was released on 09/18/19.

The video is making think twice about updating the BIOS. I had already updated my BIOS to the previous version (7B93V12) and ran into a lot of issues with CPU thermals as well as extremely high voltages, it also did not accept any changes in Power Options in the Control Panel... I was able to revert to 7B93V11) which restored my system to a normal running condition. If anyone else has updated their BIOS on their X570 boards, please post your experiences... Thanks.
Similar board that I had just updated the BIOS and can say that i'm not experiencing too much of a difference in crashes or major thermals per say but the power options don't really seem to affect my CPU clocks but that could just be cause I have XMP on.
 
Watching this video, JayZTwoCents updated his MSi X570 MEG ....
Finally went to watch the video... from comments (some from HWUnboxed, another youtube review channel) suggest he's using his Reviewer's board which is a pre-production model. Not at all representative of what's out there now.

Also MSI's had problems getting any kind of BIOS updates out for their AM4 boards, much less stable ones. Justifiably so for a pre-pro board but even so it's an MSI problem and hardly an AMD problem...whether with CPU, chipset or AGESA/microcode.

It's gotten so bad with MSI that The Stilt has started making modded BIOS's available for many of their boards that get people in the ABBA game. That's really, really sad.
 

jon96789

Great
Aug 17, 2019
145
17
95
1
I just updated my MSi MPG X570 Gaming Carbon Pro Wi-Fi motherboard BIOS to the latest 7B93V13 posted by MSi on September 18 with the latest AMD PI1.0.0.3abba update. The update still ignores any settings in the Control Panel's Power Options, with the CPU clock speed running at 100% even though I set it to a maximum of 99%.

I am currently running Prime95. The CPU voltage averages ~1.3 volts and peaks at 1.494 volts. The CPU speed averages at 3.725 Ghz with one core hitting a max of 4.55 GHz (0.05 GHz faster than before), and another hitting 4.45 GHz. The CPU temp averages 70 degrees C and peaked at 95 degrees. The VRMs hit 90 degrees C and stays there. The X570 chipset is running a constant 60 degrees C.

At idle with only FireFox running with HWiNFO64, the CPU voltage still runs about 1.488 volts. The CPU speeds idles at 3.6 GHz. The CPU temp fluctuates all over the place, from 41-60 degrees, the VRMs settled down to 55 degrees and the X570 chipset still runs at 60 degrees.

Closing down HWiNFO64 and using AMD's Ryzen Master shows way different stats. The CPU temp idles at 40 degrees, CPU speed shows 0.5 GHz. i cannot find any other info other than CPU PPT at 35 watts, CPU TDC is 9A and CPU EDC is 10A (whatever those mean)

Overall, I find this BIOS update a lot more stable than 7B93V12 but 7B93V11 was even better. That version let me cap the CPU max speed at 99% (or 3.72 GHz) which also dropped the CPU core voltage to 0.9 volts. The CPU temp would be stable at 35-40 degrees at idle.

When stressing the CPU with Prime95, the CPU would run mostly at 3.725 GHz, literally no different than running the old BIOS at 99% CPU power. The CPU runs a lot cooler and peaks at 75 degrees. To me, this shows MSi seriously needs to test the BIOS updates better as it shows bugs which should not be there. I am seriously considering reverting to BIOS 7B93V11. The CPU ran more consistently and cooler which I think will prolong the CPU's life...
 
...
I am currently running Prime95. The CPU voltage averages ~1.3 volts and peaks at 1.494 volts. The CPU speed averages at 3.725 Ghz with one core hitting a max of 4.55 GHz (0.05 GHz faster than before), and another hitting 4.45 GHz. The CPU temp averages 70 degrees C and peaked at 95 degrees. The VRMs hit 90 degrees C and stays there. The X570 chipset is running a constant 60 degrees C.
...
Wow. 90C on the VRM's... what CPU are you running? I'd heard many of MSI's X570 VRM's were lame, now it shows.

If it was still in a return windows I'd sure take it back. You're just as well off with a B450, better with some X470's, and for a lot less cost.

I'm not really sure what you're trying to achieve by 'capping the CPU at 3.72Ghz' or what point your trying to make about it being buggy or unstable. HWInfo, nor any utility other than Ryzenmaster, is not currently capable of reporting the processor sleep states correctly. AMD has told us that, so there will be reporting disparities at idle conditions, and accurate VCore reporting is always a challenge with any Ryzen. Also, we're to expect the processor to ask for voltage up to 1.5 V during light bursty work loads as it is boosting. AMD says that's perfectly normal.
 
Last edited:

jon96789

Great
Aug 17, 2019
145
17
95
1
The problem is that the 3900X CPU is almost always running neat 1.5 v. Also the latest BIOS ignores any of the Control Panel Power Options. I like to use the Power Saver mode in order to use the CPU in a more conservative state and use a different power plan when needed. I don't need to have it running 100% all the time.

I use the PC primarily for video encoding which is not that often. Even with the CPU throttled at 99%, the encoding time difference is minimal, so I am fine with that. Like i mentioned, even with the CPU running at 100%, it runs about at an average of 3.725 GHz with occasional burst to about 4.2 GHz. With the CPU restricted to 99%, it runs 3.72 GHz, not much of a difference.
 
Wow. 90C on the VRM's... what CPU are you running? I'd heard MSI's X570 VRM's were lame, so now it shows.

If you can take it back I'd do it. You're just as well off with a B450, better with some X470's, and for a lot less cost.
The problem is that the 3900X CPU is almost always running neat 1.5 v. Also the latest BIOS ignores any of the Control Panel Power Options. I like to use the Power Saver mode in order to use the CPU in a more conservative state and use a different power plan when needed. I don't need to have it running 100% all the time.

I use the PC primarily for video encoding which is not that often. Even with the CPU throttled at 99%, the encoding time difference is minimal, so I am fine with that. Like i mentioned, even with the CPU running at 100%, it runs about at an average of 3.725 GHz with occasional burst to about 4.2 GHz. With the CPU restricted to 99%, it runs 3.72 GHz, not much of a difference.
Then it doesn't sound so much buggy or unstable as just not configurable in the manner you wish, and apparently the earlier one did.

There are other ways to gimp performance for lowering power consumption you might want to consider. Probably the easiest is to disable cores and SMT for instance. Either or both drops CPU power consumption massively. You can have a 4 core/4 thread processor for the knock-around stuff and then turn back on all 12 cores/24 threads to get the encoding done fast. I was surprised how responsive my 1700 remained for web browsing and desktop apps when configured that way (4/4). But it idled with such low power consumption anyway it was not really worth it.
 
Last edited:

jon96789

Great
Aug 17, 2019
145
17
95
1
If the BIOS does not respond to any changes in the Control Panel Power Options, then i would consider it to be buggy. The BIOS should not override any OS controls...
 
Jul 15, 2019
76
19
35
0
I'm in the same quandary with my MSI X570 Gaming Plus. It shipped with the A2 BIOS, based on AGESA 1.0.0.3a. It ran my Ryzen 3600 right out of the box, and has worked well for the month I've had it. If HW Monitor is to be believed, the core boost speed is very close to the advertised 4.2GHz. I skipped the A3 BIOS (AGESA 1.0.0.3abb), but MSI just released A4 with 1.0.0.3abba.

I know the usual rule is not to update the BIOS on a system that is running well. But 1.0.0.3a was a work in progress, and there are now two updates to it. There is a (probably small) risk associated with BIOS updates, but I don't understand the benefit of abba enough to decide whether the update is worth the risk.
 

jon96789

Great
Aug 17, 2019
145
17
95
1
TBH, the latest 7B93V13 BIOS is "supposed" to fix two things...
  1. Fix t he Turbo problem where the CPU does not reach it's claimed turbo speed
  2. Fix the problem where diagnostic programs do not reflect the actual CPU and voltage data from the motherboard.
In reality, the Turbo boost increases the CPU speed by 0.025-0.050 GHz, so my CPU went from 4.5 GHz to 4.55 GHz. But the turbo boost speed only occurred for a fraction of a second and never reached that speed again. The average CPU speed remained the same running Prime95, approx 3.72 GHz (I could get a higher CPU average with Cinebench 20, about 4.2 GHz).

The second issue was not really resolved. Core Temp and HWiNFO64 still has data jumping all over the place. What bothers me is that the core voltage ALWAYS registers ~1.48-1.5 volts which is too high and the CPU temps usually is about 50 degrees...

With the earlier BIOS 7B93V11, the voltage would drop to 0.95 volts when the CPU is not under load. The CPU temp sits at ~35-39 degrees (when Power Options was set to Power Saver mode).

I decided to rollback the BIOS to 7B93V11 until MSi can release a BIOS that is less buggy and allows me to control the CPU speed which the latest one ignores.
 
If the BIOS does not respond to any changes in the Control Panel Power Options, then i would consider it to be buggy. The BIOS should not override any OS controls...
I've never known any BIOS ever (OK, ever since the original IBM PC/AT) that responds to OS controls (or utilities run from within the OS) and I would consider it a buggy BIOS if it did. By that I mean: once something is set in BIOS it should boot the computer that way everytime until it's over-ridden from within the OS. That seems to me a very highly desireable behaviour in order to prevent an errant OS app from rendering your computer unuseable.
 

jon96789

Great
Aug 17, 2019
145
17
95
1
What I mean is that the Power Options in the Control Panel no longer works after BIOS 7B93V11... No matter what setting you choose, the system ignores it...
 

jon96789

Great
Aug 17, 2019
145
17
95
1
Yes... I do have the latest AMD Chipset drivers. I installed them when I got my PC a couple of weeks ago. I also have the latest Windows 1903 build which has optimized sections for the AMD 3000 CPUs.
 
Yes... I do have the latest AMD Chipset drivers. I installed them when I got my PC a couple of weeks ago. I also have the latest Windows 1903 build which has optimized sections for the AMD 3000 CPUs.
But did you install the chipset drivers before or after the BIOS update... i assume it's an ABBA BIOS? I've read that it's important to re-install after updating to ABBA. It sounded to me like it's a conditional install based on the SMU revision level in BIOS.
 
Jan 12, 2019
13
0
10
0
I never had that high temp problem with Ryzen 3700x. Max 70 degree under heavy load, even with the stock cooler. It runs at 1.45 V and 1.35 V under full load. By the way, I have updated to AMD's latest chipset driver. This fixed lots of not recognized, not working properly SSD, NVMe SSD and NVMe SATA SSD problems for me.

Should I update the bios for Gaming Edge X570?
 
Jul 15, 2019
76
19
35
0
I just updated my MSI X570 Gaming Plus to BIOS version A4 (AGESA 1.0.0.3abba). I used M-Flash with no problems. So far, the only thing I've noticed is that it POSTs and starts Windows faster. I already had the latest AMD chipset drivers.

I went into the BIOS after the update reboot to load optimized defaults, turn on XMP to run the 3600MHz RAM at its rated speed, and disable the built-in LAN and serial port, which I'm not using. When Windows started up it had to re-associate several USB devices with their drivers, as if they were installed for the first time. Windows also started up with only one of my two displays, although the second display turned on after Windows reinitialized the USB devices. I also had to manually set the DisplayCAL profile loader to the existing display profile files, which Windows no longer recognized as system defaults for color management. But on subsequent reboots Windows used all those settings correctly. I'm not sure why the BIOS update affected Windows that way, but don't be surprised when it happens.
 
Aug 22, 2019
3
2
15
0
I have the msi x570 pro carbon wifi and the 3700x. At the moment, my temps are great, but my cpu only reaches 4.25Ghz max and not the advertised 4.4ghz. Should I update the bios?
 
Sep 27, 2019
1
0
10
0
What I mean is that the Power Options in the Control Panel no longer works after BIOS 7B93V11... No matter what setting you choose, the system ignores it...
hi i had a similar issue , fixed it by removing that msi "dragon center" software. Might be worth a try if you have it installed. i have the stock bios and can adjust frequcies in the power settings with powersaving mode, balanced and high performance settings. if you install amd chipset drivers "amd_chipset_drivers_v1.8.19.0915 " it also shows options for amd balanced and amd high performance settings.

hope this might help
 
I just updated my MSI X570 Gaming Plus to BIOS version A4 (AGESA 1.0.0.3abba). I used M-Flash with no problems. So far, the only thing I've noticed is that it POSTs and starts Windows faster. I already had the latest AMD chipset drivers.
....
One thing I've picked up from AMD community guidance. They recommend that even if you had previously installed the updated chipset drivers you un-install them and then re-install them after updating BIOS with a later AGESA. I suspect the installation is conditional and adds different components depending on which AGESA version it sees, or something similar. It's only an issue if you get the BIOS update after the chipset drivers... but since MSI was so slow getting updated BIOS/AGESA out, we all installed chipset drivers out of sequence.
 
Oct 4, 2019
6
0
10
0
I've rolled back to Bios V11 and it's still happening. CSGO will blackscreen and PC will restart. Wolfenstein Youngblood carked it at 3am... I just gave up then and went to bed. Hardware log files showed nothing out the ordinary. 3600 was moving around the 70-76 Degree mark, RTX2060 super was a cool as a cucumber at 60 degrees and boosting constantly around 1950Mhz. Lost now.
 

ASK THE COMMUNITY