0 Replies Latest reply on Nov 14, 2011 7:43 PM by Aoxomoxoa

    DH67CL + 2500k i5

    Aoxomoxoa

      This horrow show began two months ago.  I was asked by my boss to build a new PC for the office to, among other things, serve as our file server.  I settled on the DH67CL, a 2500k i5 CPU, and three WD 1TB drives to make up a Raid 5 array.  Everything seemed to go well with the build with one exception.  The day after building it, I came into work to find the system at the login screen instead of at the desktopo as I left it.  Upon login I was told the system had recovered from an error and got information about a bluescreen.  As a result of crash, the RAID array had to reinitialize.  Initialization is a very very slow process which takes up to 3 days to complete, and the system repeatedly crashed every 1-2 days.  I've debugged the dump files and found no useful information.  The crashes seemed to be caused by various files such as cdd.dll and ntoskrnl.  I tested every component as best I could.  I ran memtest for about 20 hours, ran intels cpu test as well as Prime95.  The only hitch was a crash during one of the Prime 95 tests, but it was not unlike the crashes that had been happening while the system was idle.  The PSU reports voltages all within acceptable range of +/- 5% of target voltage.  I reinstalled the OS several times, updated drivers, BIOS, etc.  Nothing I could do would prevent the system from crashing while idle on almost a daily basis.

       

      At this point I decided to surrender to Intel's Live chat tech support.  I told the agent everything I had done and their first response was that I should RMA the board.  No troubleshotting, no suggestions other than an exchange.  I even tried to coax more information or more suggestions from the agent, but got little help.  Since it was within 30 days of purchase I decided to ask the vendor I bought it from to handle the RMA process, which they did grudgingly.

       

      Replecement #1

      When I finally got the first replacement board 2 weeks later, DIMM socket 0 was dead.  I don't know how a board makes it past intels testing with a dead DIMM slot, expecially when its the one that should always be populated first, but it did.  So talked to live chat again and after having to argue with the agent briefly, I got to talk to a supervisor who agreed to RMA the board again.

       

      Replacement #2

      Another two weeks, and I got the third board (second replacement).  I installed the board and the DIMM slots worked fine.  The only problem was when I booted up, intel desktop utilities claimed that ALL voltages were out of range.  After looking into it I found that IDU had somehow gotten all the voltage threshold values wrong, and I could not reset to defaults or change them manually to fix it.  I removed and installed IDU several times, and even updated to the latest version that had been released since I had the original board, but nothing fixed it.  Aside from the IDU issue, I also notcied that at least twice, the system lost video signal and had to be rebooted to restore it.  At this point I decided I hated myself enough to try to talk to another live chat agent.  This one suggested I update the BIOS from 0125 to 0132, and update that had been released in the weeks that I was waiting for a functioning replacement.  I began the BIOS recovery process using the latest BIOS, but when the system booted...nothing was happening.  I hit reset and quickly found out that I just interrupted the BIOS update that I wasn't aware was running due to the video not working.  Naturally the BIOS was corrupted and the board I had just unpacked not 20 minutes ago was dead as a doornail.

       

      Replacement #3

      I talked to a supervisor again, this time I was offered Advanced (2 day) Replacement, which meant I didn't have to wait another 2 weeks.  I asked everyone I talked to along the way if I could have a new retail box instead of someone else's bad board.  Each time the reply was that they couldn't guarantee it but would request it.  Naturally I didn't get a new board, I again got someone else's returned board.  I installed this (now the fourth board) and the DIMM sockets all worked, and IDU magically had the correct threshold values again and didn't claim my voltages were off.  The RAID array had been through alot previously, so I wasn't surprised to find it was "verifying and fixing" data.  I left the system idle over the weekend while the data was scanned and repaired.  I came in on monday (tady) to find the system running, but with a blank screen.  I rebooted and got that familiar notice that the system had crashed.  To top it off the RAID array was again verifying data and starting off at 1% again.

       

      So after who knows how many hours spent troubleshooting, testing, reinstalling motherboards, boxing and unboxing, and dealing with four bad boards in a row..I'm now back to where I started.  A somewhat functional, but completely unreliable and unstable system.  For all I know, this board I have now is the very same board I sent back through the vendor to begin with.  I feel like intel's RMA method is to assume that user error has occured, yet suggest RMA regardless.  Then they seem to take your board and exchange it with another board that someone else sent in.  I guess the hope is that many of the boards returned will be sent back out to end users and found to be working.  I can't think of any other explanation as to why I would receive a replacement board with a dead primary dimm slot, except that intel doesn't test returned boards thoroughly.

       

      So if you're still reading this, and I haven't lost your attention, I'm back to where I started two months ago.  I have no idea if this board I have now is the very same board I started with or if it just happens to have the same flaws.  At this point talking to live chat or trying to RMA the board again seems like a ludicrous idea, so I'm posting here to see if anyone else has had similar issues.  My next move is probably going to be buy an Asus board and see if all my problems magically disappear.  If they do, Intel will definately get an earful, I will be seeking a refund from them, and it will be the last Intel board I buy.  That last part makes me a bit sad, because for 15 years I've been building systems using only Intel boards.  Some of those 10-15 year old systems are still running today.

       

      In the meantime if anyone has any suggestions or similar experiences with this motherboard, please let me know.

       

      UPDATE:

      After posting this I noticed white noise (static) from the speakers for several seconds when rebooting (after POST and before windows GUI begins to load).  I haven't noticed this with any of the previous 3 boards.  I loaded the latest driver provided by Intel, and haven't noticed anymore of the static during boot, but I also don't hear any sound at all in windows.  I tried all the obvious stuff like testing the speakers on another system and testing other speakers on the system with no sound.  I reinstalled the driver with no success, and tried just uninstalling the driver and letting the windows driver install.

       

       

      Conclusion:

      I feel thoroughly defeated, and Intel does too.  The RMA department suggested a refund before I could even ask for one.  The postage on all this has already exceeded the cost of the product by far.  No ones making any money at this point.  I've been using Intel boards and loving it so long now that the thought of building a system with another brand kind of makes me sick.  I've used intel for many years because despite being limited on features, they were always easy and stable, with good support and drivers.  I've never had to RMA any intel product before this one and the experience is so aweful I feel that buying any other product from Intel would just be poor judgement on my part.  Every replacement sent to me was simply somone else's returned product, not tested and not refurbished.  Three replacements, and each of them was in a much worse state then the original board I sent in.  In one case the board couldn't have passed any sort of simple testing because it could not boot with ram in dimm 0 channel A.  To make it worse, throughout this entire proccess I have to talk to to people that want to question my logic at every step of the way.  Intel need to clean house in the RMA department, start actually testing and throwing away bad boards instead of trying to recirculate everything.  It's not the consumer's fault that production has an inherint failure rate and that some boards slip past quality control.  I feel like I'm being punished by Intel for getting one of those bad eggs.