If it's anything like New Zealand in terms of consumer law, your contract of sale is with the supplier so you should get in touch with the company who sold it to you new. A quick check using Intels website didn't show an Aussie contact for motherboard warranty support, only one in the Philipines (for Asia Pacific customers). The only simple method of contact I found was to use the chat button (if you find it goes green for you there - perhaps at certain times of the business day)
If you wanted help to prove there's definitely a hardware fault (rather than something you may be able to fix yourself), what's gone wrong?
Thanks for the answer. I bought this MB online via Sacheto.com . I thought I can contact Intel directly since this motherboard is a retail product.
The problem is that under heavy load (mainly I/O) computer reboots or freezes. For example, if I build big software project on my machine (8 processes in parallel), then after 2-3 minutes of work it just reboots. No BSOD, no nothing. Nothing in the system event viewer either.
Also, I installed Intel Desktop Utilities and I'm not sure it reports true data. For example, it shows CPU temperature as 60-69C most of the time, which is ridiculous for an idle machine. I opened the case and touched the heatsink's base where it attached to the CPU. It's icecold.
In some forums people say that the north bridge chip may be overheating. I can touch the sink, it's warm, but not too hot, actually.
I don't overclock, and I use default BIOS settings. All I need is stability. Intel motherboards were renown for that once...
What brought this problem on or has it always done this since you recently installed it? What memory (make, part no and capacity), CPU, PSU, Graphics card and storage (including space remaining) does the system have? I believe these Extreme Series boards can be particularly picky about memory modules. What does Intel Desktop Utilities show is happening with all the different rail voltages around the time of the spontaneous reboot - are any fluctuations present? I wouldn't worry about the CPU temp being misreported in this particular case because some boards do have known issues like this. Remember that the CPU temp reported may be the 'thermal margin' rather than the actual temp ie the gap between the actual temp and the maximum the CPU can run at so a high no in this case is good - check what IDU syas about this to be sure with your particular motherboard. My DG45ID misreports one of the minor temps (can't remember which and its not installed at the moment so I can't check), as being at 58 degrees C even when first switched on in a cold room. What does the BIOS show the temps (and voltages) are like if you manually reboot and enter the BIOS after a speel of normal operation? Is it only building this specific software project that results in the problem - what about copying a huge file over from the the LAN or an external HDD - do these things cause rebooting as well (and if so, again, how does Intel Desktop Utilities behave at the time)?
I did find these Aussie Intel contacts (after typing in www.intel.com.au) Intel Offices in Australia so maybe that'll be some help (should you NEED to get the board replaced - be aware of the note at the bottom re they're not tech support however). I think it'd be worth exploring other options as above first though as you'd want to make sure it is the board to blame to save yourself alot of wasted time and money posting working boards away etc.
Thank you for pointing out my mistake regarding the thermal margin. Yes, it says CPU thermal margin ~69-72C after boot. I thought it was Intel jargon for CPU temperature. It makes sense now.
I bought this motherboard a couple of days ago, so I use it less than a week. The reboot problem occurred first time I tried to build Boost library (if it matters). Boost library is a huge C++ library, which takes about 30 minutes to build with 8 parallel processes running simultaneously. I used to build it with the same hardware I have now with my old motherboard, which died because of physical damage.
I tried to reduce build processes to 4, it still reboots. I succeeded to build the library by limiting build to 1 process. Intel Desktop Utilities don't show anything suspicious at all. All temperatures are good and green. Voltages are green. Everything is OK, but after 2-3 minutes of build the machine just freezes. It started to freeze since I connected my HD and CD drive to the blue SATA connectors (Marvel SATA controller). Before that they were connected to the black ones - in that case I experienced reboots rather than freeze.
Copying big files are OK. I copied my backups from laptop and and external drive without any problems. Some files were of size of several GB's. Moreover, video games run OK, too. Though games are not very I/O intensive applications.
Here is the full spec:
Product Name: DX58OG
Serial Number: BTOG104002LE
BIOS Version: SOX5820J.86A.0888.2012.0129.2203
i7 920 @ 2.67GHz (Bloomfield)
3 x 4GB (total 12GB)
Zotac NVIDIA GeForce GTX 295
FSP Aurum CM Gold 750W (AU-750M)
Western Digital, 1TB, SATA 3 Gb/s
Thanks for the further details and I trust that you found my (edited in later) Intel Australia non-tech-support office details.
Attempting to pin this issue down further, is it ONLY when working with these boost librarys that you get this problem? Given you said this is a replacement board for another that was damaged physically, are all the remaining comonents the same ones that worked with that board and was the old one also a DX58OG configured exactly the same way in the BIOS, of the same hardware revision (AA number) and using the same BIOS revision? Are all the latest drivers installed on the board and is the HDD controller in AHCI mode in the BIOS or just IDE? I've had difficulty tracking down the specs for the memory module part number you listed so are you sure it meets the requirements of this motherboard listed here: DX58OG System Memory ?
Was the PC running at the time it was damaged and if so have you tried running any CPU testing apps on it to give the system a really good workout? How does running the Windows Experience Index on this machine go and what does task manager show CPU usage on each core is like when the PC is worked hard? If its a very complex setup you have on your HDD with many programs installed etc, do you have another blank HDD (a different one will be fine as long as it connects to the SATA cable) that you could swap in for testing purposes just to do a Windows/windows updates//driver and Boost library program install and try running just that problem app to see if the problem recurrs on a fresh install? If its a similar board you are replacing this shouldn't make a difference but if the 'old' PC was running at the time (or has unknowingly been run since it was damaged), who knows what damage/corruption may have been done to the contents on that HDD and that may explain the rebooting problem you have now.
Thanks for your help.
The old board was from Gigabyte and I used it quite a long time until I started to have unexplainable problems (random reboots etc). Then I started to investigate the problem and I found that a couple of pins in the CPU socket were bent. They were too tiny to fix them, so I decided to replace the MB. I bought new motherboard and new memory as well.
Now that you asked me about memory I actually went to the manufacturer's website and checked it: G.SKILL-Products . Apparently they don't have X58 chipset in the list! Moreover, on the page you posted (DX58OG - System memory) I noticed now the following requirement:
Unbuffered, single-sided or double-sided DIMMs with the following restriction: Double-sided DIMMs with x16 organization are not supported
I suspect that I've got this dreaded x16 organization, since I have 16 small chips on each memory block (8 on one side and 8 on the other side). I saw this page so many times and it never occurred to me that this may be the problem. I'll go tomorrow to the shop I bought it from and try to replace it with some of memory from the list.
Also, I tried to run Intel Processor Diagnostic Tool. First time I ran it - it was able to trigger the problem with the "CPU Load" test. After reboot I ran it again from command line to save the log and it passed.
Also, in the other thread on this board people suggest to install KB-2664888 hotfix.
I'll try to replace the memory first, then I'll see what's going on.
P.S. Boost library is not a product in the common sense of the term. It's a software library which is distributed as source files. You need to build it (compile it with C++ compiler and then link it linker program) to get executable binaries. The build process is very CPU and memory consuming, and also puts a lot of strain on disk I/O system. So, practically it's a kind of a test.
P.S. My system rating is not available yet. since I just installed Windows and everything a couple of days ago. But with older MB it was around 6.2-6.5 if I remember correctly.
ETA. Yes, I saw the Australian office link you posted. Thanks for this. I'll contact them if I'm not able to fix the problem myself.
My system rating is not available yet. since I just installed Windows and everything a couple of days ago. But with older MB it was around 6.2-6.5 if I remember correctly.
Thanks for the further info. You can easily force Win 7 to run the assesment - there's a link somewhere on the screen and certainly you can re-run it after it's been run the 1st time using a similar link there.
I wonder what caused the old mobo pins to bend and if running it like that has damaged the CPU? Given it's a different brand of board, did you install all the latest Intel provided drivers for it after plugging it into the HDD or did you just let PnP instal Win 7 native drivers? If the latter, you should really install the Intel ones (and in the correct order, chipset ones going first). I don't know if that memory 'organisation' related term equates directly to the number of chips so your modules may well still be perfectly compatible and it's just an omission from the manufacturers website. It'll be interesting to see what the supplier says about it (or even the manufacturer if you email then your query).
Frankly, I don't know how pins got damaged. I think one of them was damaged long time ago (maybe since the first day of CPU installation), and another two were damaged recently when I tried to figure out what's going on and installed/removed CPU several times.
Once I've got new board I reinstalled Win7 from scratch. I never use old installation. With latest Intel drivers, BIOS, etc.
Apparently, when I tried to run Windows rating test, the machine rebooted again. It seems that CPU load triggers this quite consistently. It's not the CPU temperature, since it doesn't grow that much before machine reboots. It also happened with the Intel Processor Diagnostic Tool. During the CPU load test of IPDT the reported temperature is about 50-47 degrees before maximum and then system reboots. I'm wondering if the CPU got damaged after all, because of these bent pins.
I'll replace the memory anyway to something from the compatibility list.
Will keep you posted.
Did the CPU eventually pass the Intel Processor Diagnostic test or does that never complete due to the spontaneous rebooting? As you mentioned the best plan of action is to correct any possible system incompatibilities first and see if the problem still occurs. Its good to know you performed a fresh OS install when the new motherboard was installed (presumably by never using the original installation you mean after formatting the HDD rather than an 'over the top' of the existing installation quickie).
Yes, I never install OS "on top". I always format HD first.
Intel Processor Diagnostic test succeeded only once out of 3 or 4 times I ran it. It always causes reboot when it reaches CPU load test. Even though CPU temperature is still within safe margin (50-45 degrees before maximum). So, I'll replace the memory to see if it happens. Then, if it doesn't help I'll try to find another working CPU and test with it.