Maybe, still under investigation. The CPU won't be able to boot even, but blank in BMC if R12+.
ES Xeon Discussion
- Thread starter britinpdx
- Start date
Notice: Page may contain affiliate links for which we may earn a small commission through services like Amazon Affiliates or Skimlinks.
Hello, guys.
I bought an MS73‑HB1 and CPUs with stepping A0. In single‑processor mode it works with BIOS R17. For dual‑processor operation, as I understand, is needed to disable ACM or use the old BIOS like R02. Could you send me a link where I can download the old BIOSes? Looking at the Internet Archive I saw that early versions of the R series aren’t there. If you can provide documentation on how to edit the binary to disable ACM, I’d be glad. Yea, I’d like to learn how to do it. I’d also appreciate an already made modification.
I should also say that English is not my native language, so I may make some mistakes. Anyway, I’m open to discussion.
I bought an MS73‑HB1 and CPUs with stepping A0. In single‑processor mode it works with BIOS R17. For dual‑processor operation, as I understand, is needed to disable ACM or use the old BIOS like R02. Could you send me a link where I can download the old BIOSes? Looking at the Internet Archive I saw that early versions of the R series aren’t there. If you can provide documentation on how to edit the binary to disable ACM, I’d be glad. Yea, I’d like to learn how to do it. I’d also appreciate an already made modification.
I should also say that English is not my native language, so I may make some mistakes. Anyway, I’m open to discussion.
R02 will not work, that was a mystery of a user updated from T01(work) to R03 with AFU tools.
PM you now with BIOS and more info.
From what I've heard long time ago - Intel initially thought that E3 would be the production stepping, made a QS, shipped them to customers and then there was some undisclosed hardware bug there that they decided to fix in hardware and that is how we got E5, which is an actual retail stepping for SPR.
As nobody can say (likely don't know) what that bug was - better to get E5 if in doubt.
Thanks, wrote to US seller (who has 8481C) instead of Chinese one. Let's see if sale gets through.
some 'C' OEM non ES have final stepping E3.
all official intel SKU E3 (step 6) are EV-QS( early validation )
while E5 ES (step 8) are official QS.
Interesting, those are without accelerators I assume? So that might narrow down where the change was.
'C' can have more or less, custom models. the OEMs want them early, so only stepping E3.
if you see 8481C , regardless of ES or not, the features are the same.
Hello
Gigabyte MU72 SU0 with Xeon 4189 ES CPU.
Thank you so much for your help and guidance about this motherboard.
I obtained an Intel Xeon Platinum 8360Y ES QWAU CPU - 36Cores 2.2GHz 250W LGA4189 Stepping 6.
It was marked
The motherboard booted without problems using the F23 Bios. I installed Win 10 to evaluate.
HWinfo64 reported this CPU as a Xeon 8362B SRKYM with a D2 Stepping.
I have not investigated further, but this F23 Bios looks like it does not support Intel PPT and a Gigabyte 14 Pin TPM is not recognised with this Bios/CPU.
Has anyone used this motherboad with with a more up-to-date Bios? Current available Bios is F40.
Just a data point, I was surprised to see that with 32GB Ram and 1 small M.2 NVME drive – this motherboard idles at just 51 Watts.
Thanks again for all you help and guidance.
RedX1
can you provide a CPU-Z screenshot ?HWinfo64 reported this CPU as a Xeon 8362B SRKYM with a D2 Stepping.
OEM Intel Xeon 8362B is TDP=225W
it is a mismatch match of HWinfo because they have same specs. in detection table.
Last edited:
Hello
Thank you for the instant attention.
Here is the CPU-Z output.
I do not have the official CTM010 SPI TPM but I used a 14-1 pin SPI TPM module bought from Amazon that works perfectly in a Gigabyte MD71 HB0
Maybe a more recent Bios would unlock these features?
Thanks for your interest.
RedX1
b
simplifies the choice of a core activation mask for dual processor 64x core on socket systems, developed for simplify debugging processors on gigabyte ms73-hb1
git.skylar.tech
BitEditor : #russianantispyware:xmr.se : Free Download, Borrow, and Streaming : Internet Archive
simplifies the choice of a core activation mask for dual processor 64x core on socket systems, developed for simplify debugging processors on gigabyte...
archive.org
simplifies the choice of a core activation mask for dual processor 64x core on socket systems, developed for simplify debugging processors on gigabyte ms73-hb1
Attachments
Hello, new user here. This topic has been incredibly useful.
I'm currently building a 3rd Gen Ice Lake workstation using an ASRock C621A WS (BIOS 1.20) and Optane 200 PMem (4x32GB DDR4 + 4x128GB Optane).
I initially tried an 8358 QVM8, but encountered a missing memory slot (B1). Despite multiple swaps, CPU reseats, and testing with a full DDR4-only setup, I couldn't get slot B1 to be detected.
Based on feedback in this thread regarding missing memory channels on early steppings (pre-stepping 6 like the QVM8), I returned the CPU and ordered the recommended QWAT (8368) instead. However, I experienced the exact same symptom: slot B1 was still not detected ("The following channel did not pass CPU memory test (DDR4_B1)").
I also noticed several DRAM errors in the IPMI logs, and it was impossible to upgrade the BIOS from 1.20 to the 1.27 beta requested by ASRock (the system would freeze when loading the .bin file).
For the community: Both QVM8 and QWAT are correctly identified by the C621A WS on BIOS 1.20. PMem was also detected with both CPUs, though I couldn't test optane "Memory Mode" properly due to the missing channel and IPMI errors.
Given that two different CPUs showed the same B1 failure and BIOS update issues, I suspected a motherboard defect. ASRock support agreed, so I've shipped the board back and am currently awaiting a replacement.
A few questions for the experts:
I'm currently building a 3rd Gen Ice Lake workstation using an ASRock C621A WS (BIOS 1.20) and Optane 200 PMem (4x32GB DDR4 + 4x128GB Optane).
I initially tried an 8358 QVM8, but encountered a missing memory slot (B1). Despite multiple swaps, CPU reseats, and testing with a full DDR4-only setup, I couldn't get slot B1 to be detected.
Based on feedback in this thread regarding missing memory channels on early steppings (pre-stepping 6 like the QVM8), I returned the CPU and ordered the recommended QWAT (8368) instead. However, I experienced the exact same symptom: slot B1 was still not detected ("The following channel did not pass CPU memory test (DDR4_B1)").
I also noticed several DRAM errors in the IPMI logs, and it was impossible to upgrade the BIOS from 1.20 to the 1.27 beta requested by ASRock (the system would freeze when loading the .bin file).
For the community: Both QVM8 and QWAT are correctly identified by the C621A WS on BIOS 1.20. PMem was also detected with both CPUs, though I couldn't test optane "Memory Mode" properly due to the missing channel and IPMI errors.
Given that two different CPUs showed the same B1 failure and BIOS update issues, I suspected a motherboard defect. ASRock support agreed, so I've shipped the board back and am currently awaiting a replacement.
A few questions for the experts:
- Can anyone confirm their experience with the QWAT 8368 using all 8 DDR4 slots? I've since acquired 8x RDIMMs and plan to go full DDR4 for better bandwidth/latency instead of using Optane in Memory Mode.
- Is the QWAT 8368 known to be free of the "missing channel" bug? It seems highly unlikely to get two different CPUs with the exact same dead channel.
- Are there any other known drawbacks to these ES chips, aside from the 200MHz clock speed reduction?
i had almost all QWxx models at home and never had a missing channel. even not with QVxx step 5 or step 4.
yes, the channel is left after some month of work, not missing from factory.
QVM8 is C0 step 5 (should not have mem.channel bug)
QWAT is D0 step 6 (good)
Hello! I bought a QYFS and a Gigabyte MS03-CE0 motherboard. I am stuck with "EE" code during the chipset initialization phase for more than 20 minutes. I have 1 stick of known good 4800 MT DDR5 ECC RDIMM and no peripherals. CPU is torqued to 8 in-lb (0.9 NM). Could someone share F13 BIOS with me?