• ×
    Information
    Need Windows 11 help?
    Check documents on compatibility, FAQs, upgrade information and available fixes.
    Windows 11 Support Center.
  • post a message
  • ×
    Information
    Need Windows 11 help?
    Check documents on compatibility, FAQs, upgrade information and available fixes.
    Windows 11 Support Center.
  • post a message
Guidelines
Are you having HotKey issues? Click here for tips and tricks.
HP Recommended

Thanks for all your help, SDH. I see something about BitLocker and passwords when I go into BIOS and start that process. I do know that this box came from a company that was running several of them. I don't know if I have anything to worry about.

 

So...I'll put that spare mobo to good use and swap that back in, then try the BIOS internet method of installing 2.54. That one was from an SFF and I see a few differences; the rear fan header has 3 pins instead of 4, there's one less USB port, and there's no heatsink on the big chip by the white PCIe slot; prolly chipset chip. CPU-Z identifies my tower mobo as 8053, and as I recall the SFF it calls 8054. Spitfire rev A. They both use the same BIOS of course, and same mATX form factor.

 

If this BIOS re-install works then I'll see if the systems sees the Hynix 4GB stick in DIMM1; this was the mobo that only liked DIMM3. That should yield some good info about what's going on. Stay tuned; I'm jumping on that in a few minutes,

 

MT

HP Recommended

Alright - I think major progress. In BIOS, for BIOS update, I only see the option to roll back to 2.53, which is fine with me. I do that, and it boots into windows with a Hynix 4GB stick in DIMM3 only. Shut it down, put the other Hynix 4GB in DIMM1, reboot...and YES! It sees both sticks! The 'memory' tab in CPU-Z does show it's operating in dual channel mode. CPU-Z shows that my current BIOS rev is indeed 2.53.

 

This will be a good day...now...do I try the Crucial dual rank sticks, or wait for the specified single rank to arrive? What do you guys think?

 

Thank you so much for this help!

 

Edit; no point in being conservative with this mobo; it's a spare. Just tried the dual rank Crucials in there, and YES...it sees all 16GB. So Crucial isn't exactly telling the whole truth when they say dual rank is incompatible. Those new sticks will go back. Yes, it certainly appears that it was the method of installing BIOS updates that was the issue. Thanks!

 

MT

 

HP Recommended

I guess the HP .bin servers let you roll back 1 level. It must see you already had 2.54 (?) and did not offer to have you reinstall that again.  You might now be able to use the net to upgrade again to 2.54 from within BIOS or do that with my thumb drive technique. Sounds like something got corrupted and now is uncorrupted?

 

Let's see if DGroves has any added questions or tips.  I still don't fully understand what you have installed in memory slots 1-4. Are they all single rank now, and all same type?  Sizes? For my simple mind I try to put all same size, rank and OEM in all my memory sockets and have enough eBay skills to pull that off at a reasonable cost. For example sometimes I'll choose to load 8 x 4GB to fill all sockets rather than 4 x 8GB and leave 4 empty. On the project below, however, the 4GB sticks were actually a bit more expensive than the 8GB sticks (lots more supply of the 8GB sticks at that time due to server retirements).

 

We upgraded about 25 Z420 v2 boxes all with recycled 8GB 1866 MHz server memory sticks at about 11.00 each plus very fast v2 processors, but the memory issues you've had have been odd. We had none of that.

 

I've attached my PDF on upgrading BIOS from within BIOS having the harvested .bin file onto a thumb drive, see below.  I personally would upgrade using that method to 2.54. You may be gun-shy right now, however.

 

Congrats... and lots of credit to DGroves for his advanced knowledge and insights.

HP Recommended

if you have ANY incompatible ram installed,  and it seems to boot normally just wait..... at some point you will power on and get a ram error this may take several months to happen but when it does it simply will no longer boot until you remove the incompatible ram , discharge the cmos and waiting several  hr before powering on with correct ram

 

we wasted many many hours before learning that these flex based systems are just as picky with ram as some of the servers we run (if not more so!)

 

install the correct ram, and the latest bios and be done with it..........................................

HP Recommended

Thanks for the thumb drive instructions; looks like I'll need it. With the original mobo back in, and using the same technique as the other one to update BIOS, this one isn't fixed. I tried both 2.53 and 2.54. I get the same thing as before, with both two Crucial sticks, and two Hynix sticks. I will do the USB method and advise. 

 

On the SFF board, I had one 8GB Crucial stick in DIMM1, and one 8GB Crucial stick in DIMM3. Those modules are dual rank. 

 

MT

HP Recommended

I'd put in exactly the same sticks in the same slots in your currently bad motherboard to test with what worked perfectly together in the other "good" motherboard... only the two Crucial sticks.  It sounds like you're mixing Hynix and Crucial in the motherboard that won't work properly and using just the two Crucial ones in the "good" motherboard that will work.

 

If so, it may be the mixing of Hynix with Crucial that is your real issue. Maybe if you had 4 identical Crucial sticks you'd be fine. What are the ranks for the Crucial vs the Hynix? Read on...

 

Please again read carefully DGroves latest post. He specifically said they had to do a low level clear of the CMOS settings when a memory issue developed and only after that could they get the affected motherboards to work properly again. I think you might be able to do that with the motherboard out of the box by removing the MB battery and press/hold the CMOS clear button multiple times, but not sure. It is also possible one of the two motherboards is partly bad.

 

I respectfully think it is you and your choices in memory that is causing this, plus the very picky memory control of that system, plus corruption of your CMOS by putting in the wrong combo of memory sticks, plus not knowing that you need to clear that corruption before you try again. I've been doing high level HP work for quite a few years now and have never seen or heard of what you're going through. But, DGroves has, and has been through it himself. He has been very generous with his expertise on this forum for years now. He gets paid a fair amount for what he gives here for free...

HP Recommended

what the "OP" and "SDH" may be missing is that all parts have tolerances, and as such having one motherboard boot with unsupported ram is no guarantee that all boards will, same applies to the dimms just because one may work is no guarantee that all unsupported ram will work

 

"SDH" is aware of this issue with the z800/z820 systems with their chipset revisions and the cpu upgrade issues,....... the exact same thing applies here and while the "OP" may have time to spare dinking around with this issue i can guarantee you we don't in a business setting, we're paid to determine the problem, and come up with a fix in the shortest amount of time

 

as such my mindset is geared to the quickest solution and in this case it's installing the correct ram

HP Recommended

Well I never mixed Crucial and Hynix. I tried both Crucials, and both Hynix. I think that's all I better say until the single rank Crucials get here. $75 for a letter and I haven't worked in 3 years...

 

MT

HP Recommended

Ok, well got my sticks from Crucial today, confirmed the 'DFS' P/N on both, installed both with BIOS rev 2.54 still on there, booted up, got the same old thing with only DIMM1 stick recognized. Rolled back to rev 2.53 from within BiOS, got same thing. Turned off PC, removed stick from DIMM3, updated BIOS back to 2.54, booted into Windows, turned back off, re-installed stick in DIMM3, re-booted into Windows, same old thing.

 

???

 

MT

HP Recommended

Don't forget DGroves experience that things would not work normally for them until they did a low level clearance of the CMOS when things were like this for them. To me that points to somehow the CMOS settings getting corrupted and not getting uncorrupted until they did the full clearance method.

 

Very strange situation but they saw it too, and figured out how to clear that... Good luck to you!

† The opinions expressed above are the personal opinions of the authors, not of HP. By using this site, you accept the <a href="https://www8.hp.com/us/en/terms-of-use.html" class="udrlinesmall">Terms of Use</a> and <a href="/t5/custom/page/page-id/hp.rulespage" class="udrlinesmall"> Rules of Participation</a>.