cancel
Showing results for 
Search instead for 
Did you mean: 

[FIRMWARE] Intel ME (H810/B860/Z890)

MoKiChU
Level 40

Hi everyone,

- Intel ME Consumer Firmware :

Firmware : 19.0.5.1948 [18/02/2025]
PMC Firmware : 1900.21.0.1059
PCHC Firmware : 1900.21.0.1010
SOCC Firmware : 1900.25.0.1009
ACE Firmware : 20.40.1483.0
PHY S Firmware : 13.0.1.7093
PHY N Firmware : 19.2.0.7014
IOM Firmware : 49.20.0.0
TBT Firmware : 19.0.0.1501

Download : Link

Check/Update Process :

Check current version : Right click on "Check.cmd" > Run as administrator

Install/Update : Right click on "Install.cmd" > Run as administrator > Automatic restart when install is done


OS requirements : Windows 10 64 bit or more recent.
Hardware requirements : Motherboards with Intel H810/B860/Z890 chipset.

10,751 Views
28 REPLIES 28

Dimm fit is a bios configuration not in windows not a app, it force trains the memory for about 5 to 6 hours.

its almost like this me firmware update is preventing the cu-dimm ram kit from using its external clock generator.

 

i am running windows 11 pro 24h2

 

bitlocker is disabled i dont use that feature

Ncola0, 
Okay, but were you able to Safe Mode into Windows and do a System Restore to undo the problem?

Go do crazy some place else, we're all stocked up here!

So windows is fine that I fixed fairly easy using cmd in system repair mode after a few to many failed boot attempts.

 

what seems to have appeared to happen is that this me update changed something with the ram side of things.

 

cu-dimm have a built in clock generator was rock stable at 8800mhz c42 at 1.45v for about 2 weeks prior till after I updated this me firmware.

manually clocking the ram down tonight at 7000 made it error free, is why I’m thinking it has to do with it now using the mem clock generator in the mem controller on the cpu die. Games are no longer crashing and it’s an amazing experience.

when back at 8800mhz ram via xmp  using Occt I saw 21,588 errors when testing the ram.

i ran out of time tonight, but tomorrow I’m going to dig deeper into the cu-dimm bios settings see if something is not on auto like it was prior to updating.

its the one set of settings I have not looked at yet. 

previous to the firmware update everything was perfection.

fun times I had playing gta5 enhanced lol

 

WOW this is Great!  You have come a long way to solving this issue. 
In my own research and experience with BIOS & ME Firmware updates; often an update will reset some back to default states in the BIOS. There are even messages that appear to check it.

Go do crazy some place else, we're all stocked up here!

MoKiChU
Level 40

Hi everyone,

- Intel ME Consumer Firmware :

Firmware : 19.0.5.1948 [18/02/2025]
PMC Firmware : 1900.21.0.1059
PCHC Firmware : 1900.21.0.1010
SOCC Firmware : 1900.25.0.1009
ACE Firmware : 20.40.1483.0
PHY S Firmware : 13.0.1.7093
PHY N Firmware : 19.2.0.7014
IOM Firmware : 49.20.0.0
TBT Firmware : 19.0.0.1501


Thread updated.

NcokeO
Level 9

@MoKiChU thank you for this, put everything to stock, hoping this has brought back stability again.

 

 

NcokeO
Level 9

updated Intel me firmware still receiving thousands of cpu and memory errors, how ever the new one is not causing game crashing as often.. 

took apart the pc and found physical damage to the cpu rear side pads below where memory controller is located.

 

question is it possible during the Intel me firmware update that due to using cudimm ram that it decided to turn on functions of the cpu mem controller that were being controlled my the on memory clock driver?

ended up returning the cpu mem and motherboard today. That was short lived excitement. 

IMG_3899.jpeg

LOL Yes, I am laughing out loud!  How in the world did you ever think to check the physical hardware of you PC? Brillian Discovery!

Go do crazy some place else, we're all stocked up here!

I’ve only ever seen this type of errors with bad ram or other physical issues.

 

wasn’t expecting to see the damage on the bottom pads of the cpu.

they were out of 285k CPUs at my local store so I swapped to a 9950x3d because they had one in… bummed but excited lol, now to just wait for them to get a asus hero motherboard in

 

also in my local store we tested the ram for about with memtest86 on my 285k and the extreme board I was using failed immediately no matter the settings. Put the ram in a different board with a different 285k “the owners” and all was good so at least the cudimm ram was not damaged. Ended up picking a slower kit for ryzen that is expo ready.

 

using my 285k and ram we got into windows and ran the intel cpu testing tool, cpu passed lmfao like the fk