Drag to reposition cover

ASUS Pro WS W790E-SAGE SE + Intel Xeon Sapphire Rapids SPR-SP

Notice: Page may contain affiliate links for which we may earn a small commission through services like Amazon Affiliates or Skimlinks.

Max1024

Member
Jun 23, 2024
33
3
8
All my BMC settings are - OFF, I do not use this functionality at all. Now I use 10G lan card from Asus on Aquantia AQC107 and all is fine.
I've seen reports that if the board warms up, the problem with Intel LANs disappears, as if it were a problem with the resistors, similar to the one that the Asus Dominus Extreme had. Very sad story.
 

Max1024

Member
Jun 23, 2024
33
3
8
@RolloZ170

Are you planning on creating a new thread for the ASUS Pro WS W890E-SAGE SE with GNR-SP Xeon-6 LGA4710? I saw in the processors thread that you responded to ES samples of the GNR-SP. I think this board will also be very popular, and I'd like to learn more about it ;)
 
  • Like
Reactions: RolloZ170

RolloZ170

Well-Known Member
Apr 24, 2016
10,685
3,418
113
germany
@RolloZ170
Are you planning on creating a new thread for the ASUS Pro WS W890E-SAGE SE with GNR-SP Xeon-6 LGA4710? I saw in the processors thread that you responded to ES samples of the GNR-SP. I think this board will also be very popular, and I'd like to learn more about it ;)
intel confirmed the pinout is same, so we can't brick anything. but it is not easy like as with W790.
GNR-SP are SoC and don't have external chipset.
GNR-WS are started(bringup) by Intel ME in PCH W890.
in the meanwhile we should all pray that the DMI4 lanes are free available at the GNR-SP processors...
 
  • Like
Reactions: Max1024

Not Sure

New Member
Jan 22, 2026
10
3
3
Hi everybody - my first post - my kingdom for a bios screen.

I've got this board, with SPR-SP QYFS ES.

The board doesn't boot, power button does nothing, nor does the reset button.

Board is brand new, PSU is a brand new ASUS Pro Workstation 3000W, board power connections as per manual, case fan is connected to satisfy CPU fan requirements (its on full water), with 2x32GB ECC DIMMS, one each side nearest the CPU (A1 & E1), and a random 1050Ti for video.

Its on the bench for a pre install test, not yet in a case, as I had a feeling this wouldn't be smooth.

BMC appears to work just fine:

1770920360236.png

I try to update the bios via the USB flash port, being sure to run the renaming tool before copying the cap file to the media.

It 'appears' to flash... runs for around 5 mins of blue light flashing, and eventually goes off.

But at no time does the BMC recognize this change (see above).

So only the 'Safe Boot' button does much of anything - spins up the GPU, but Q-code remains at 00 (not used).

All the switches on the board as they were leaving the factory test.

So what am I missing here?

I have a Dell 7960 Tower with 3495X, but its busy, so I'm reluctant to pull that CPU yet - especially given how much 'fun' LGA4677 is to cage and mount.

Any ideas for me guys?

Thanks.
 

Max1024

Member
Jun 23, 2024
33
3
8
So only the 'Safe Boot' button does much of anything - spins up the GPU, but Q-code remains at 00 (not used).
I had the same behavior when a pair of 8-pin connectors (with 1 PSU) were connected not to the top connectors themselves, but to those lower down (i.e. incorrectly)
 

Kizune

Member
Dec 2, 2022
97
69
18
Hi everybody - my first post - my kingdom for a bios screen.
First of all - you will not see new bios version in BMC until system fully gets through the POST process - it’s BIOS who updated BMC with the version string not the other way around.

The 00 means system did not recognize the CPU and refuses to power it up. Either CPU is damaged or you have a bad contact with the sense pins - problem with pressure applied by the cooler or with dirty/oxidized contacts on the CPU. As a troubleshooting step I would get the dumb cheap (maybe even passive) cooler and replace the water block with it - at least to get trough the POST process. And swapping these CPUs is no big deal if you swap them together with the cooler - after they mounted they are kept pretty good on the cooler by both bracket and the thermal paste.
 

Not Sure

New Member
Jan 22, 2026
10
3
3
I had the same behavior when a pair of 8-pin connectors (with 1 PSU) were connected not to the top connectors themselves, but to those lower down (i.e. incorrectly)
Thanks guys for your help.

I had stumbled across my error of my own doing.

I had created my own confusion by intending to run dual-PSUs, and had remembered what I needed to connect based on that. so when I saw that the lower case watercooling would not quite fit dual PSUs, I ordered the giant single one - the WS 3000W Pro. And long-story short, connected power as ATX12 1_1 and ATX12 1_2.

Correcting that to ATX12 1_1 and ATX12 2_1 saw it post first time...

Apologies for wasting your time & appreciate the quick responses.

Erik.
 

Not Sure

New Member
Jan 22, 2026
10
3
3
Well guys,

I'm not out the woods yet. First two dimms reported, bios reports as updated.

So I added the 2nd pair of dimms and the 905p on a quid U.2 carrier.

It looked like it was going to boot, but it didn't.

Shutdown, entered bios, only 3 of 4 dimms recognized.

Q code is AA.

Shutdown, reseated the 2nd pair of dimms, just the same - 3 dimms in bios and AA.

I pulled the 905p and added a Ubuntu 24 usb stick, it gave me the option to Install or Try Ubuntu, but I'm still suck at Q code AA and a black screen.

Maybe reseat the CPU (done already the other day)....?

BMC:

1770927941154.png

Appreciate any thoughts...

Regards,

Erik.
 

RolloZ170

Well-Known Member
Apr 24, 2016
10,685
3,418
113
germany
Shutdown, entered bios, only 3 of 4 dimms recognized.

Q code is AA.

Shutdown, reseated the 2nd pair of dimms, just the same - 3 dimms in bios and AA.
you remove AC power while swapping right ? clear cmos after all changeds before power up.
 

Not Sure

New Member
Jan 22, 2026
10
3
3
you remove AC power while swapping right ? clear cmos after all changeds before power up.
Yes power removed before changing anything.

Well it seems I got a bad dimm... tried suspect dimm in all other available slots, and I don't see it in bios.

Also, I still get the splash screen for Ubuntu, but never make it beyond making my choice, and then Q-code AA again.

Thats it...

Well 4:15am. I'll look at this tomorrow...

Regards.
 

Max1024

Member
Jun 23, 2024
33
3
8
Also check the CPU clamping force; it may be necessary to tighten the clamping force by tightening the bolts more tightly.
 

RolloZ170

Well-Known Member
Apr 24, 2016
10,685
3,418
113
germany
Also, I still get the splash screen for Ubuntu, but never make it beyond making my choice, and then Q-code AA again.
AA is checkpoint, not error.
AA System has transitioned into ACPI mode. Interrup controller is in APIC mode.
normaly seen while OS boot.
 

Not Sure

New Member
Jan 22, 2026
10
3
3
AA is checkpoint, not error.
AA System has transitioned into ACPI mode. Interrup controller is in APIC mode.
normaly seen while OS boot.
Hello and thanks for your comments.

I could boot 'nothing' with the ES chip, and had CATERROR CPU & one dimm not 'seen' by bios.

I tore apart my beloved Dell for the CPU this evening, seated the 3495X, set secure boot to Other OS and it booted the UB24 iso immediately.

So I seated my 4x U.2 carrier with one of the 905p's and it booted to desktop first time (even though that was a Dell install) and here I am chatting on it.

I took pics of the CATPROC ERRs, sent to the seller, and asked if either they had another or a refund.

I really appreciate the time you guys give to those of us in need.

Desktop! .....

1771182600555.png

Oh, and no more CAT PROC ERRs anymore in the BMC.

Some complaints about power supply connections, but I can't see what that could be.

I assume in the previous posts there are some 'safe limits' for initial overclocking?

Regards,

Erik.
 
  • Like
Reactions: RolloZ170

Not Sure

New Member
Jan 22, 2026
10
3
3
Hi and well, I'm back.

I put in 384GB of RAM in the thing (all read first time - Dell would sit on that for multiple boots and maybe 10 mins!), set the bios to Watercooled mode. And got some good numbers on CPU-Z, like 751 ST and 44,000+ on MT with all 8 channels populated (it was around 41,000 / 42,000 with four dimms).

But what I never saw right, was the storage performance. It seems as though there must have been some high latency affecting it. Each of my 905p's would have like 9MB/s in ATTO / 4kB, and then sometimes 180MB/s.

I pulled the 2TB SN8100 from the Dell (again that booted right away), but was also horrible. It could peak sometimes near 11GB/s or even 14GB/s but performance was all over the place. And also roughly the same in the 4kB numbers, maybe a bit better... PCIe issues on this board I was suspecting.

Bored of chasing that, I played with the Intel's ITU, and made just a +200MHz change to each bin, applied it, and locked up. Rebooted it and it never came back.

I had some Qcode message about PCH before microcode [or something similar], then PCIE initialization. Now I get stuck at Qcode 92 [PCI Bus Initialization is started].

With manual in hand, I locate what looks like a bios battery underneath the rear IO board, but there's no way my fingers can reach that. Not sure i see a bios rest button either, just a RETRY, and a SAFE BOOT. But neither has saved the day. So how does one get to the CMOS battery?

BMC reveals a CATPROCERR asserted at my tinkering-time, but also it was de-asserted later. Did I really fry such an expensive CPU?

My last resort is either a bios reset, or restore. Then I'm going to have to try the 3495X back in the Dell.

As an aside, the seller of the failed ES CPU I had, refunded without a single word of communication, nor asked for his bad chip back. I guess at this point it would have been nice of him to send another as asked, but c'est la vie.

Appreciate any / all ideas on how to move forward.

Regards.
 

Not Sure

New Member
Jan 22, 2026
10
3
3
BIOS has OC features, why use iXTU ?
Hi,

Yeah, tried the clear cmos button, but was hoping to be able to pull the battery I cannot reach. I'll try it again though...

Intel's ITU is something I've used for over a decade, so that is why I went there... sorry I did.

Thanks,

Erik.
 

Not Sure

New Member
Jan 22, 2026
10
3
3
With mixed memory SPR make like this.
Dont make 32gb modules with 64 gb modules.
Pcie init error and malfunction.
Hi, all eight modules are 5600MHz 48GB each.

4800MHz 32GB modules are back in the Dell.

WELL, I pulled the 5090, and it posted & booted FIRST TIME using a 3090 in the same slot.

The 5090 was also overclocked, so I guess maybe that died and was holding up the PCIE bus initialization.

I guess I'm happier losing a 5090 GPU than a 3495X CPU if thats the case...

Thanks again guys, lots more experimentation to do...

Erik.
 
  • Like
Reactions: RolloZ170

Not Sure

New Member
Jan 22, 2026
10
3
3
So I finally found some time to play with this again.

Re-flashed the bios, and tried min to many peripherals, to see what would work, and what wouldn't.

So it seems I have PEX errors, and lots of them. The 5090 works again, but I can see PCIe PEX Errors Recovery Counter numbers. Like 46 for the 5090 (even pegged the 5090 to pcie 4.0 - Which improved the numbers, but I have PEX errors no matter what HW is installed it would seem).

5090 46 errors
3090 21 errors
3090Ti 19 errors
A10 (enterprise card reports very little to Windows Workstation) so sat this one out.

All the above from HWMonitor.

HWInFO64 doesn't even initialize past the bus scan:

1773585708409.png
So did I managed to get a bad board?

As an aside, a buddy claimed my 3495X Xeon would game horribly, so I looked up his 7800X3D, and found I placed top 1% globally for 3Dmark's Steel Nomad DX12 and 5090.

Doesn't seem so 'horrible' to me:

1773586125110.png

Just more people who don't know what they're talking about.

Regards.
 

Attachments

  • Like
Reactions: Zakes