X58A UD3R DMI Data pool

[Dark Mantis]

I don't know about your location but if you purchased it from a Dutch eshop then contact them about the RMA procedure. They will then tell you whether to send it to them or Gigabyte Netherlands.

[Trinitrotoluene]

Thank you, I have done so yesterday so I expect a reply from them today.

Memtest still shows coldboot errors, again at 5080,0MB test #6. Just like when I first plugged them in. So this is not a reseating issue, for sure. Even tried 1066Mhz, QPI/VTT voltage on auto, RAM voltage on 1.64, 7-7-7-20 timings as rated for 1333Mhz. Got a nice BSOD on windows login system_service_exception. Upon autorestart, the system would hang at "AHCI bios installed". So I reset it, at that point the screen would stay black and I could hear it hang in an endless reboot. Turned off the power, restarted, some message about overclocking settings etc.

I don't really know where and when this will end. The PSU is supposed to be OK, I've tried two PSU's already. The current PSU is a quality one. Reading about this motherboard from people that have cold boot issues with OCZ memory and everything, I'm not sure what to do. This might go on and on and on up until I just buy memory from lets say Kingston. I can't get a refund anymore I suppose. So I'm pretty much stuck with OCZ and Gigabyte combination. It is on the QVL memory list, but still. Plus I lost my faith in RMA departments and perhaps some brands as a whole. Perhaps when I RMA either of the two to the webshop and ask for an alternative product, things will get sorted quicker than just sticking with OCZ and Gigabyte.

 

[Dark Mantis]

I know some people really like OCZ memeory but we do find a lot of problems on the forum with members who are using it. I think from what I have heard it has gone downhill in quality over the past months.

[Trinitrotoluene]

It might be too early to call it, but if I must follow my gut feeling I'd say I RMA the board to get the coil whine fixed. RMA the memory after having a good chat with my webshop, and have them to swap it with some kingston or corsair ram.

Just think about it, it's rediculous to receive 2 faulty OCZ kits and the 3rd also doesn't work with a QVL motherboard (or so it seems, so far). Anyway lets see what my webshop says on both products. And thanks again for the backup here.

[Dark Mantis]

Yes I would agree with your call and why take any chances if it could be motherboard related as well. As you say you have the coil whine also so maybe you can kill two birds with one stone so to speak. Anyway good luck and let us know the outcome.

[Trinitrotoluene]

Thanks, I sure will! Have been sending a couple of emails back and forth with my webshop yesterday. Haven't come to a final agreement yet on the RMA. But they already offered to take in the motherboard and ram. So we're on the right way. Just want to make some clear agreements in advance.


I would like to add some more info to this thread. Though not to find out whether RMA should still be done, that is obviously a YES.
Yesterday evening I decided to run some more memtest at 1066Mhz 7-7-7-20 1.64v on the ram and all other timings on auto (tRFC is 60 this way with 1T command rate, qpi/vtt voltage on auto also). This was recommended by my webshop so I tried it (whether this OCZ kit likes tRFC of 60 is debatable, but it is at 1066Mhz though). Also, I've put the CPU vCore back to auto as well, to exclude that from causing any problems. Now, memtest would freeze in a seriously messed up way:

http://img8.imageshack.us/img8/2663/20101124190714.jpg (non-coldboot)

Also, I had two freezes yesterday during gaming. One freeze ended in a BSOD with driver irql not less or equal. The same type of freeze I've been having since this new kit (though the first crash with a BSOD). I've seen various people on the net getting this (the type of bsod) also with ram issues. So I'm also leaning towards that still, as my drivers appear to be sorted.
This morning I did another memtest coldboot on the same settings as mentioned above. And guess what, errors at 1080.0MB range again (which also happened at rated specs), this time test #4. Again faulty ram seems plausible as the errors have been reproduced in the same range and solely this range, so far. Unfortunately I will be away for a few days so I won't be able to test these modules individually. But one would expect to see just module 3 from slot 5 (tested in any slot of course) to throw errors when tested alone and the other two to be clean. Here's the screenshot of this morning, after which memtest also froze. I guess if I would let it run for a longer period of time, it might end up as in the screenshot above.

http://img26.imageshack.us/img26/8550/20101125075801.jpg (coldboot)

[Trinitrotoluene]

Just a quick heads up: shop agreed to take in the motherboard (coilwhine), ram (no way to get it stable?!) and cpu (exclude IMC trouble)! So that's good news. Will be able to ship out the hardware tomorrow, I'll keep you posted.

If the OCZ kit is broken or not compatible even though it is on QVL (both options are rediculous), I'm thinking about going with Kingston. Any recommendations are still welcome though, as I haven't made a final decision yet!
Could you tell me if I should worry about bent CPU pins in my case? Just to ease my mind a bit, even though it might just be a guess... I checked Gigabyte RMA and if I read it correctly, bent pins is no longer an RMA thing?
Or might this have been a problem caused by my @bios flash to FB from the start?!

[Dark Mantis]

Well good news that it has all been accepted by the retailer for RMA so hopefully that will cover all bases.

I reallly wouldn't worry about what it could be that caused it as we tried hard to pin it down but without success so more guessing now isn't going to get you anywhere.

Personally I would go for Corsair or Crucial memory as they both have good RAM and they both have websites with a memory finder on them and will guarantee that whatever they recommend will work.

[Trinitrotoluene]

Hmm, you're right! The Corsair and Crucial lists are also very welcome.

One last question, when I get my new motherboard, will my RAID 0 array still work or do I have to totally destroy it before creating it again? Even though it's obviously the same board, I'm just not sure if this will work. I've temporarely copied all my music to that array, because the original drive it was on started to show errors so I took it out. If I have to recreate the array, it's not the end of the world. The error'd drive still works, but stil...


Speaking of which, can you recommend any 1.5TB (or bigger) drives for RAID 1 on this board? I plan on running two of them in the future.

[Dark Mantis]

The RAID should be ok as it will be using the same chip again. That is the important bit. Just make sure that you set all your RAID details again before you go into the OS. You should then be able to do any repair.

As far as HDD go I would probably go for the Samsung F3. I don't have any personal experience of them but have heard some good reports on other forums. They do one of 2GB for around £60 I think.

[Trinitrotoluene]

What kind of repair are you talking about exactly? And all the details I should set are just in the normal bios, selecting raid for harddrive contro etc.l? Or also in the special raid bios ctrl+l? Sorry for the constant stream of questions, but I'd like to be prepared you know.

I've read about raid arrays going bad because people booted into the OS before setting raid details again. I sure as hell won't forget about that hehe!

[Dark Mantis]

The error'd drive still works, but stil...

I thought that one of your drives had errors. That was the repair I was talking about.
As far as I am aware you only have to set up the RAID in the  BIOS again as the rest of the settings should still be there.

[Trinitrotoluene]

Oh right! No, that broken drive is out of the equation here. Sorry if I didn't clear that out!

And by "the rest of the settings should still be there" you mean the files that were written to the two raid 0 drives when I created the array? Keep in mind I might not get my own board back. That's the webshops decision, they might just grab a new board from stock and send that right back to me. Instead of dealing with it via Gigabyte RMA.

Does this mean I'm in trouble?

[Dark Mantis]

I don't think that should be a problem as long as it is using the same chip. It's only when you try to change chips that you have troubles.

[Trinitrotoluene]

Thanks again, appreciate it!

Next couple of days will be interesting. I'll definately keep you up to date on the RMA process!