MPG X870E CARBON WIFI Beta BIOS

Sort by date Sort by votes
And I was JUUUST about finished tuning my CO on A27....
I was definitely seeing much worse curve optimizer performance; before I was fairly stable between -25 and -30 in all tests, on A27 I was seeing -15 to -20 crashes on many of my cores. Hoping that A28 will help out with that, but I've been able to tune my memory fairly tightly at not-insane voltage in A27 with DDR5 Nitro enabled and set to "auto".
I'll start out at those same memory timings without a curve optimizer, let AIDA64 with CPU, Memory, FPU and Cache all checked. FPU seemed to be the trigger for failure in A27 for sure. So if I can push some of my worst cores above 20 and it runs stable, I'll call that a win.
I suspect A27 had a pretty minimal voltage setting by default and Curve Optimizer would just dip it too low. What are folks setting their core base voltage to?
 
Last edited:
Upvote 0 Downvote
Updated.


Bedouille69 said:
Hello @Svet ,
4 boots ok and the fifth boom stuck on max 3.0 and 1.1 or 2.0.
I don't understand the issue but I enabled Spread spectrum again.
Click to expand...

Don't know, you can try the BIOS new version .1A29

Aside from possible BIOS issue,
you may have some HW issue, like bent pins or too much pressure over socket from cooler or some issue with the CPU itself.
 
Upvote 0 Downvote
Svet said:
Updated.




Don't know, you can try the BIOS new version .1A29

Aside from possible BIOS issue,
you may have some HW issue, like bent pins or too much pressure over socket from cooler or some issue with the CPU itself.
Click to expand...
If never stuttering with 1A1A, it's not a hardware issue.
I received my computer with 1A17 and all fine during 1 month, then I updated to 1A21 and the issue began. 1A17 was no longer available so I tried 1A1A and never wrong boot for 2 weeks.
If I reboot when it stutters, my computer works fine.

I don't understand how the pci express speed can be wrong on random boots not especially cold ones.

I installed 1A29 and on first boot I see Bluetooth and Wifi but no Realtek audio.
It is not normal to have to do a Clear CMOS to see it again even if I disable it every time.

3rd boot with 5 minutes break between each :
1A29stuck.png

The difference is that now the max speed is correct and the current speed 2.0 is the maximum even with graphic test --> Spead Spectrum enabled to avoid this.
 
Last edited:
Upvote 0 Downvote
Indeed, thanks @Svet. Appreciate you posting the Betas.
Initial testing on A29 (will edit it as it goes on):
M2_1 speed slightly worse than A26: Still short on random writes by about 400,000 on my 990, but now sequential writes are about 100mb/s lower than before across 3 tests.
M2_3 speed back to worse than A27: Lost both Seequential Write and Random Read from A27, about 400MB/s Squential Write Speed, missing 100,000 IOPS on Sequential Read. This is back to what A26 was, losing the gains made in A27. Both are lacking from where A21 was.
Took my stable memory settings from A27 on my Hynix-M die Non-Binary Corsair Dominator Titanium 8000 48GB Kit. Running a full FPU/Memory/Cache/CPU test with no Curve Optimizer or PBO uplifts. So far it seems stable, but was only for about 30 minutes before coming back to my non-test PC to do some after-work gaming. The night is young and I need to see about 6-8 hours
 
Upvote 0 Downvote
I see the exact, and I mean exact same thing as Bedouille69. Identical symptoms with recent BIOS versions. No issues at all with 1A1A.

I understand and believe that not everyone sees this issue, that it's possible it's something specific to our PC configurations. Looking at Bedouille69's signature, the only hardware we have in common are the motherboard, CPU, and graphics card. I have an MSI 4090 Suprim, though mine is a Liquid model.

Updating to a newer BIOS will important in the long run, but for now I am comfortable sticking with 1A1A and hoping the root cause is eventually found and fixed.
 
Upvote 1 Downvote
joshuacant said:
I see the exact, and I mean exact same thing as Bedouille69. Identical symptoms with recent BIOS versions. No issues at all with 1A1A.

I understand and believe that not everyone sees this issue, that it's possible it's something specific to our PC configurations. Looking at Bedouille69's signature, the only hardware we have in common are the motherboard, CPU, and graphics card. I have an MSI 4090 Suprim, though mine is a Liquid model.

Updating to a newer BIOS will important in the long run, but for now I am comfortable sticking with 1A1A and hoping the root cause is eventually found and fixed.
Click to expand...
Did you try to enable Spread Spectrum to see if it's a "fix" also for you ? I never tried 7E49v1A1G so I can't tell if 1A1A is the last BIOS without the issue (A21 was the first bad try).
 
Last edited:
Upvote 0 Downvote
For my cold boot issue, it feel like it is memory related issue. As long as, If I don't do any intense overclocking from DDR5-6000 to 6400 on the memory( enabling EXPO is fine so far). I won't have any cold boot issue.
 
Upvote 0 Downvote
Back
Top