Motherboard for Intel Xeon E5 2687W Engineering Sample

Toggle sidebar Toggle sidebar
J

james-bond

Honorable
Mar 23, 2012
3
0
10,510
Hello,
Does anyone have a suggestion for a X79 motherboard that will work with this ES2 CPU? Doesn't have to be an Asus.
Thanks for the help.
 
Sort by date Sort by votes
Welcome to Tom's Forum! :)

LGA 2011 MOBOs are:
ASUS P9X79 WS (*) - http://usa.asus.com/Motherboards/Intel_Socket_2011/P9X79_WS/#CPUS
Supermicro - http://www.supermicro.com/products/motherboard/Xeon3000/#2011
ASRock X79 Extreme9 - http://www.asrock.com/mb/overview.asp?Model=X79%20Extreme9&cat=Specifications

(*) ASUS P9X79 WS supports a very cool feature 'USB BIOS Flashback' - http://event.asus.com/2012/mb/USB_BIOS_Flashback_GUIDE/ that only requires a USB Flash drive and a PSU (i.e. no CPU, RAM, GPU, etc) is required to update the BIOS

Dual Xeon, one of many:
ASUS Z9PE-D8 WS - http://www.asus.com/Motherboards/Intel_Socket_2011/Z9PED8_WS/#CPUS
 
Upvote 0 Downvote
Sadly the Asus p9X79 WS throws a "5A" internal CPU error. I would like to know if anyone has had success with one of these engineering sample Xeon CPUs on a X79 board. Thanks for the help.
 
Upvote 0 Downvote
I'd recommend that you read the review first and if needed PM Chris - http://www.tomshardware.com/reviews/xeon-e5-2687w-benchmark-review,3149.html

This is the Engineering Sample (ES) used:
cpu-z%20processor.jpg
 
Upvote 0 Downvote
Upvote 0 Downvote
Without proper CPUID the E5 may function but it sure isn't working efficiently or properly on your Sabertooth. It's fine to 'play' but it's nothing that I would use and feel to be reliable.

I have the same MOBO and I wouldn't do it -- play sure why not...
 
Upvote 0 Downvote
Well Asus must have added the CPUID microcode or it wouldn't be working at all. I haven't had one problem with the CPU not functioning correctly. The bios even shows the CPU an when I installed the chipset it was C600/X79, I guess you could say I'm a bit of a risk taker so I'll keep you updated after I run some benchmarks! Thanks for the info tho Jaquith !
 
Upvote 0 Downvote
I'd look at the power consumption and other features i.e. VT-d, etc to see if they still work. The benchmarks should give you a good indication, and I'd use AIDA64 it's more through. ASUS might have added one of the Xeon CPUID so it would post and boot, but I haven't a clue otherwise; they might be sharing BIOS coding with all of their X79's...I don't know.

Yeah, let me know how the testing goes.
 
Upvote 0 Downvote
Hey well when I was home for lunch I checked the power consumption an it was very normal, I went ahead an downloaded AIDA64 ran a full benchmark everything passed flying colors ! It did a comparison to other rigs an it was at the top ! It clocked up to 3.5Ghz during benching! Looks good so far maybe tonite I can post some screen shots!
 
Upvote 0 Downvote
It is >not< the microcode which is causing 5A boot errors with ES2 steppings (B0/B1) on the ASUS Z9PE-D8-WS.

I extracted the microcode data and it contain the proper microcode for the B0/B1 steppings. I also tried replacing the entire microcode block with the data from the P9X79 Pro where ES2 works - and it still won't POST (same error).

I also tried replacing CPU PEI and DXE drivers (because error happens in the CPU DXE initialization phase) - and it does not help either.

So, something else is wrong but I don't have time right now to spend few days digging into this and disassembling the BIOS.
 
Upvote 0 Downvote



I was just looking online about the 5a error look's like it could be a problem with the PCIe lane, look's like some people swapped video cards an it's working for them !
 
Upvote 0 Downvote
^ This is the QS1 sample (C0 stepping) - they should be much less problematic on any decent board as they should be virtually identical to the final silicon except some very small tweaks (ehm... VT-d)

ES2 is one generation earlier, and it has somewhat more differences.

Anyway, it could be PCIe - the EFI boot sequence goes as follows: 67 --> 69 --> 5A

69 is System Agent DXE initialization - system agent in SB-e holds PCIe controller as well... but I can't be really sure at this point what is the problem actually.

In normal circumstances, after 69, the board should progress to 6A -> system agent SMM initialization, and this happens in the, say, ASUS P9X79 Pro board without problems. But with Z9PE-D8-WS, it is just stuck before that.
 
Upvote 0 Downvote


This chip is pretty sweet ! Yah that's what i thought i did a little research on it before an found out that it was a QS (C0) ! Yah the ES2 (B0/B1) are clocked @ 3.0Ghz i did see a few people running these ! If you have another video card laying around give it a try just to see ! I wish i knew more about the bios but im a newb to bios code & how it all works lol ! I know just enought to get me in trouble haha, yah alot of people have been running these chips on the Asus P9X79 from what i've seen !
 
Upvote 0 Downvote


The problem is that QS1 (C0) and QS2 (C1) samples of 2687W are very hard to find on the "open market" (ebay and such) - I suppose people having them are just keeping them as they are almost as good as final (VT-d problem notwithstanding). I guess more of these will start appearing in the next couple of months. Presumably, there are also lots of QS1/QS2 samples that are actually used in HPC production environment as indicated in the press some time ago (http://www.theregister.co.uk/2011/09/19/intel_xeon_e5_follow_up/) - I doubt those will be surfacing on ebay anytime soon...

Btw, with ES2 (B1) - 5A error appears even without any GPU present, so I doubt it has something to do with the graphics card in this case.

Also, another interesting curiosity is that it appears that VT-d is actually working on B1 samples and that the problem was introduced in C0, lived in C1 and fixed in C2 steppings... so VT-d problem is probably a result of fixing something else in between B1 and C0 :)
 
Upvote 0 Downvote
Cool, good to know. As I mentioned Chris had a tough time and his other ASUS X79 failed with his (ES) E5.

This was part of a 'fix' from the latest BIOS. Intel knowingly shipped i7's (3930K & 3960K) with this issue before fixing it with the later versions of the chips, and shipping versions of the LGA 2011 Xeon are all supposed to include this fix on the chip.
 
Upvote 0 Downvote
Yep - the problem is, it appears that the VT-d issue probably appeared in C0 stepping, and by the time it was discovered it was already late for the consumer SB-E launch (C1 stepping).

For Xeons, Intel had more time so they fixed the VT-d in C2... they would probably delay the entire Xeon launch if VT-d was not fixed anyway, as this feature is most needed in the server market.

Coming back to Z9PE-D8 WS and ES2 silicon - I obtained i7 3820, and the boards boots with it. Then, I disabled literally every single CPU and PCH feature, swapped the CPUs... and, no luck - 5A error all over again with the ES2 🙁

So, I think there are only two ways this board can be made to work with ES2 steppings:

a) Someone leaks beta BIOS from ASUS which used to work with ES2

or

b) ASUS actually fixes the issue, which I think is extremely unlikely
 
Upvote 0 Downvote
Trust me I know, I have the older i7-3930K.

I'd say you have a bad MOBO. 🙁

Lessons learned about ASUS's RMA, only Cross-Ship and if they say we don't have one in stock then wait until they do and/or place you on a waiting list. Otherwise, my last X79 RMA took a wapping 30 days total turnaround time - not a happy camper so don't get sucked-in the Que.
 
Upvote 0 Downvote
I wish I had a bad mobo, but this is the second Z9PE-D8 WS I tried.

Furthermore, it has been confirmed by several people that Z9PE-D8 WS does not work with the ES2 steppings of the Xeon E5.

Which is a pity and quite strange, since ASUS X79 boards do work (at least P9X79 WS)
 
Upvote 0 Downvote
You must log in or register to reply here.

TRENDING THREADS

Latest posts

Moderators online

Share this page

COMPANY
RESOURCES
FOLLOW
Top Bottom