Just to know - I loaded the 32MB dual-bios R34-M07 to see if it still picked up that ES Milan CPU. It looks like it did, but just curious, is there a way to check in the BMC (or maybe the bios settings), if the 32MB bios was actually loaded, or I'm still using the 16MB M07 only?
MZ32-AR0 REV 1 > 3
- Thread starter deathvenom369
- Start date
Notice: Page may contain affiliate links for which we may earn a small commission through services like Amazon Affiliates or Skimlinks.
Ok, I was wrong. This ES does NOT boot from the dual-BIOS. It will only boot from the Milan.rbu M07. BTW, in looking at this try to boot 15-20 times now, I notice that when there is even a BIOS that is compatible with this CPU, the wattage goes way up immediately. If the cpu is compatible with the installed bios, the wattage went up to 120-130W immediately. If not, it stays at 70-80W. Anyway...
with MIlan.RBU there is only one BIOS to choose. with dual BIOS the first must handover to the next if the wrong one...
supported: base clock, code execution.
not rupported: 700mhz no x86 code executed
My MZ32-AR0 with (4) sticks of 16GB suddenly only sees 3 of them. The same thing happened with my first one (512GB), where it would mysteriously start registering only 14 sticks. It seems like processing speed nosedives when this happens. Does anyone have an idea as to the root cause of this behavior? I know that the system performs better when you have at least 8 sticks populated, but this is just some random drop of memory sticks....
I know what you're talking about...I've seen that on my other system - 1 slot will have a lot of ECC corrections. But this system doesn't have anything in the logs about ECC errors. The mobo simply disabled one of the sticks. It's shown as slot A0, but in fact it is in slot H0, where the manual says it's supposed to be.
Attachments
there is another BIOS only threshold for corrected ecc errors. i guess it comes from there.
you have to clear cmos /NVRAM to get the slot back.
the ecc error must not be the RDIMM intself, there can be a bad connection here: CPU->socketpins->motherboard->DIMMslot->RDIMM
Wrong slot name can happen often, will be solved in future BIOS/BMC Firmware.
check this, same RAM type, same issue here maybe the micron's are not good supported.
Yes, it was H1, not H0.
That one line diagram about a bad connection made me wonder, so I completely reseated the CPU, reset the cooler with new thermal paste, cleared the bios, then restarted it - and it worked. It wasn't the memory at all. Since I did not want an unstable situation, I decided to remove the cooler, reseat the cooler, and it reverted back to recognizing 3 sticks only. So there is a marginal contact issue of the CPU to socket.
The question is: What can be cleaned to remove this issue, and how is that done? All contacts eventually become dirty, but I've never cleaned CPU contacts nor socket contacts.
Thanks Rollo. Yes, everything is installed correctly. Here's a pic of the problematic cooler - standard 4U cooler, really.
I think the CPU and socket contacts are dirty. I ruined a mobo one time cleaning a socket, so seriously adverse to doing it. But if it has to be done....
Is there a recommended process for doing that?
Attachments
how about bent motherboard after cooler installation ? reseat RAM after cooler install.
must not. the error threshold can be reached in first run, and then DIMM slot was disabled at 2nd run.
i have seen multiple user with similar issue with same micron RAM than yours. have you other RAM ?