ASUS Z10PA-D8 - Dual Socket 2011-3

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

gb00s

Well-Known Member
Jul 25, 2018
1,177
587
113
Poland
Did you guys notice, running the board outside a case, that CPU0 is always 3-5C warmer then CPU1? Is this due to PCI lanes are all tied to CPU0? I tested in several cooler configurations and setups Supermicro 1U, Intel & Supermicro 2U cooler + Noctua P9 or P12 and switched CPU's from one socket to another etc.. I was testing with CPU's >> 2630 v3's and 2673 v3's .

Tested with 4 boards. All the same. CPU0 always 3-5C warmer than CPU1.

Regards

Mike
 

Guillaume14

New Member
Oct 27, 2018
5
1
3
Hi

I have ordered a board from that vendor and i have a problem with memory detection (at start some pins were bent on CPU2 socket but i managed to fix them and both CPU's are now recognized )

I tried with:

4 sticks of Kingston 16GB DDR4 ECC Reg (KVR21R15D4/16): B7 error. Same thing with only 2 sticks in A1/E1 slots
Those sticks are listed in the z10pa-d8 memory list and they are working great with another SuperMicro board

4 sticks of Kingston 32 GB DDR4 ECC Reg (KSM26RD4/32HAI): same error (sometimes its working with 2 modules but not every time)

2 sticks of Crucial 32 GB DDR4 ECC Reg (CT32G4RFD424A): same problem

I have inspected the pins in the memory slots and i cant see any damaged pins.

Do you think that the board may have a problem ? The vendor is gonna sent me a new one (very good seller by the way !) but i want to now if someone got the same problem (i am using two Xeon 2680 V3 CPU's)

Regards
Guillaume
 

gb00s

Well-Known Member
Jul 25, 2018
1,177
587
113
Poland
B7 error is an initialization error. I had the same problems as I was very unfamiliar with it at the beginning.

If you change/install or uninstall peripheries you can experience B7 initialization errors and the board hangs at start-up. I always was able to fix the issue if the board is upgraded to the newest Bios version. In order to initialize everything the correct way I would start with:

1. Get a Live-CD/USB (e.g. PMagic)
2. Add one CPU only with one memory stick in A1 and start the board ... if initialization was successful (00 or AA code) and LiveUSB loaded fully, shutdown correctly. Do not just pull the plug.
3. Add the second CPU with a memory stick in E1. Same as above ...
4. If step 3 was successful then add another RAM stick to each socket ... and so on.
5. Only then start to add PCIe cards

I hope this helps. Once the board has initialized everything it should be fine.

Regards

Mike
 

Guillaume14

New Member
Oct 27, 2018
5
1
3
Hi Mike

Thanks a lot for your time: i am gonna try that as soon as i get the new board

Best regards
Guillaume
 

EagerToLearn

Active Member
Feb 25, 2020
115
25
28
Hello,

I got two E5-2690 v3 running on the board. With Prime 95 CPU+Memory test (both @ 100%) the temperatures were stable at 67°C (both CPUs).

Now I was thinking about doing the all core boost. Can anyone reccomend a good tutorial for this mod?
 

malloot

Member
Oct 24, 2013
36
14
8
We bought two of these boards a month ago and tested them, they worked great and accepted dual Xeon V4 even before bios updates.

We decided to move ahead and buy nine more boards, the shipment of course was drama but the boards survived.
Now to the real issue

None of these accepted a single Xeon V4 to begin with, only after putting a V3 in them and updating the bios to the latest version did they accept single a V4. But in no way are we able to get them to run dual Xeon V4, we tried every bios version, ram combo, psu, jumper setting and other tricks to get them to POST with a dual Xeon V4 config. Sadly nothing works, the only hint i found to the issue online was a post regarding the WS version of this board with a quote from Puget Systems

"The Z10PE-D8 WS works fine with the v4 processors... as long as it was manufactured after October of 2015. Regardless of the BIOS version installed, if the board is older than that it will not work with the new processors. We've run into this once, but Asus had warned us about it. I have been told that the only solution is to send the board back to them (Asus) to be updated."


These boards we got in the 2nd shipment of 9 boards are clearly of a different batch, the revision of both boards in batch 1 and 2 is 1.02 sadly i cannot find the date of production as described in that post.

So buyers beware: If you want to run dual Xeon V4 on these, make sure you get the newer version from the seller.


Update: Got an RMA from ASUS for this known issue, costs me 9 euro's a board to fix so i send them to them. Really hope they will return working.
 
Last edited:
  • Like
Reactions: bob_dvb

gb00s

Well-Known Member
Jul 25, 2018
1,177
587
113
Poland
On eBay US it's still available. The main seller in the EU (Netherlands) got sold out quickly after this thread opened.
 

EagerToLearn

Active Member
Feb 25, 2020
115
25
28
Basically I wouldn't care if there were extra holes, but I can't find any, that cover the ones I need...
 

EagerToLearn

Active Member
Feb 25, 2020
115
25
28
Thanks for the tipp, but since I am located in Europe, the total cost would be ~30€ which I am not willing to pay for an I/O shield :)

Maybe I leave it out all together...but your idea is good too
 

gb00s

Well-Known Member
Jul 25, 2018
1,177
587
113
Poland
The original shields sold by the guy in the Netherlands cost 15EUR as well excl. shipping. What's rare in the market isn't cheap.
 

gb00s

Well-Known Member
Jul 25, 2018
1,177
587
113
Poland
So this is mainboard number 4 failing. I received another board from the eBay seller in exchange to a 3rd broken board. The 3rd broken board just stopped working from 1sec to another. Started again, no screen, no postcode. Nothing. Now this board arrived on Friday and I build it yesterday. Worked, installed Proxmox, nice.

Today again, stopped working from one sec to the other second. Black screen. No post-code. Nothing. Dead. I post a short vid about what happens when I start it again. I checked the PSU, a Seasonic SS-1050-XM. Works. I checked it with other PSU's. The mainboard does the same. It comes up, PSU clicks audible loud, no post-code. Fans stop spinning. CPU's work on another board. So cleared that out.

Vid starting the mainboard >> VID_20200620_161843.mp4

The same parts installed on both boards where the CPU's (2x 2630v3 from same seller), the PSU, the HBA (Dell Perc), and the RAM. Tested all. Working
 
Last edited:

gb00s

Well-Known Member
Jul 25, 2018
1,177
587
113
Poland
My problem is I have a project and can't get it running due to all the issues. Yes, it's used hardware. But where ever I buy used hardware, eBay, or here, I kind of 'always' have/had trouble except for my 1st purchase, 4x 32GB DDR4 Samsung ram from SF in US. Otherwise, always issues with something. Either the wrong part, or not working at all, or doesn't come at all or very late or part is not as described and ....

I suspect something else here ... All boards worked for 4 years in a 'Rendering Factory'. Seems like without issues. They all had the original BIOS from 2016 on it. I updated all boards with the newest BIOS and firmware when they came in. After some days of testing, they stopped working. Not stress testing, but installing OS, setting up work environment with scripts to run tests for some time before I wanted it to get into production. One failed after 25days, this one after 1day. Is it the BIOS? Is it the new firmware that breaks it? I don't know. But I'm not the only one here with problems with this board.

I tested all components on other boards, tested specifically the PSU. It's super healthy and even under stress without any flaws. The CPU's coming from the same seller. They just work. The HBA (Dell Perc) came from a seller here. But I don't suspect it to break a motherboard. Same for the RAM. 4x 32GB Samsung (on pair Cisco labeled).The only thing I noticed with the RAM it doesn't like to register all of the 4 sticks on one socket if just 1 CPU is installed. Just 2 of the sticks.

I tested the power cables. I tested and exchanged the power terminals. Nothing.

I'm mad. I wanted to work on the weekend. Again, ..... And there will be a point the seller will ask me 'Are you f.... kidding me ....?'. Cheap becomes very expansive right now ...
 
Last edited:

gb00s

Well-Known Member
Jul 25, 2018
1,177
587
113
Poland
I have a small update. After @Rand__ literally pushed me to make some more testing, I started some investigation. I checked the iKVM and there is no serious information. All logs are clean like after a reset.

Screenshot_2020-06-20_20-58-37.png Screenshot_2020-06-20_20-57-51.png Screenshot_2020-06-20_20-57-28.png Screenshot_2020-06-20_20-57-16.png Screenshot_2020-06-20_20-57-04.png Screenshot_2020-06-20_20-56-48.png

I made another iKVM reset and BMC reset, took out 1x cpu + the ram installed in the slot. Additionally, I pulled one more memory stick from the first socket. The motherboard booted and showed a POST CODE 04 for a very short time. 04 is an indication for the initialization of the Southbridge for PEI preparation. One more thing I noticed is the motherboard now stays on, the fans keep running, but still without additional boot codes. The strange thing is, the iKVM shows Post Code 00, which is an indication for an already successful boot. I pulled the other ram stick from the board and noticed the board doesn't complain about the missing ram. If the board is healthy, it does beep and complains about it.
 

Stephan

Well-Known Member
Apr 21, 2017
920
698
93
Germany
Is it the BIOS? Is it the new firmware that breaks it?
I can feel the pain. Is the Aspeed out of bound chip still running and accessible? Can you flash back to the old BIOS? The boards really shouldn't die in series and that fast. Maybe there is an undiscovered short between board and case somewhere? Different PSU same issue?
 

gb00s

Well-Known Member
Jul 25, 2018
1,177
587
113
Poland
iKVM is fully accessable. I was able to go back to BIOS 3107 from 2016. After that, I get additional postcodes. That's progress. After a second boot I get 19 before the start is interrupted and the board powers down. Postcode 19 means 'Southbridge initializes before installed memory'. On the third power up the boot process goes a little bit further up to postcode 60, which means 'DXE Core Started'. This brings me to RAM :rolleyes: ?

Boot & Post >> VID_20200620_214108.mp4

Edit: Changed RAM and I get to postcode b7 which is 'Memory initialization'.
 
Last edited:

Rand__

Well-Known Member
Mar 6, 2014
6,626
1,767
113
Sure thats 60 and not b0? that also would be a memory code... I had that a lot with a broken slot on my Auss board - maybe try another one?