P67A-GD65 freezing up

[sccarlso]

I've never seen a "Forum" that wants new topics for the exact same problems (spreading potential fixes for the same issue all over the place instead of in one thread), but apparently if I post to another topic from someone with identical problems (the one right below this one with the same issue) I'm called a thread hijacker.

Seems like a ton of people are having these issues with lockups with the P67A-GD65

My setup - all new components

P67A-GD65
I7 2600k SandyBridge 3.4Ghz
G.SKILL F3-14900CL9D-8GBXL  (2 x 4GB sticks in slot 1 and 3)
Nvidia GeForce GTX 460
two WD Black sata III 1TB drives in a RAID0 in sata ports 1 and 2  (SATA 6Gb)
two WD Green sata II 1TB drives in a RAID0 in sata ports 3 and 4  (SATA 3Gb)
one WD Black sata III 1TB drive in sata port 5 (system resides on this non raid drive)
Bluray drives in sata ports 6 and 7
Corsair 750W 80 PLUS Power supply

Then go to another machine and RDP to it, no problems.  Wait about an hour, RDP and you get "Welcome" and it just hangs.

So if I go to the console, it tells me someone else is trying to login, but no drive activity at all, and won't let me do anything and I have to hard power off.

If I'm at the machine, just doing different "normal" things, browser windows lockup, applications lockup, all requiring a hard shutdown by the button since I can't select the option to reboot normally.  No BSOD's, no OC, the OC Genie is OFF

My memory is even certified for this damn board on bios 1.B which I have updated to.  But in "auto"  doesn't even get the memory timings correct.

http://www.gskill.com/products.php?index=378

I'm Going nutz trying to figure out what the f--k is going on.  I even ran the on board memory test and it passed.

I'm thinking that quality control on this production line is crap.  I've seen a ton of complaints of BSOD's, rebot loops, and random lockups which can be reproduced easily.

MSI?  Are you going to fix this!!!

How about posting the serial numbers of the bad run from manufacturing so users can see if they have a bad board?  I've been building systems for 15 years and I've never had so much trouble with a board.

Has anyone had a success story that wasn't from an RMA?  I haven't found a post where a user had these issues and made a few bios changes and they came back with "great, that fixed it".

Thanks
Scott

 

[Jack the Newbie]

I've never seen a "Forum" that wants new topics for the exact same problems (spreading potential fixes for the same issue all over the place instead of in one thread), but apparently if I post to another topic from someone with identical problems (the one right below this one with the same issue) I'm called a thread hijacker.

Now you have seen it.  Get over it and also please skip this kind of stuff:

this damn board

what the f--k

this production line is crap

... because if you can't, I will close this thread immediately and delete the entire thing a.s.a.p. This latent aggressive attitude is not going to shed one bit of useful light on the problem. >>Please read and comply with the Forum Rules.<<

Also:

I've been building systems for 15 years and I've never had so much trouble with a board.

I've heard that kind of thing a million times.  Be that as it may, it does not really tell us anything about the actual cause of the problem. Furthermore, if you are so sure it is the board, why didn't you simply send it back/RMA it/have it exchanged after you have ruled all other explanations out systematically?

-------------------------

Seems like a ton of people are having these issues with lockups with the P67A-GD65

There are thousands of freeze issues on all different hardware configurations.  However, more important than the symptoms are the precise hardware configuration the freezes occur in and the actual causes for the common symptom.  Once the causes are truly found out it makes sense to compare similar issues.  It should not be done, however, because of a common symptom.  And that is exactly why your problem (and any other problem that is just a symptom at first) deserves its own, seperate thread.

---------------------------

P67A-GD65
I7 2600k SandyBridge 3.4Ghz
G.SKILL F3-14900CL9D-8GBXL  (2 x 4GB sticks in slot 1 and 3)
Nvidia GeForce GTX 460
two WD Black sata III 1TB drives in a RAID0 in sata ports 1 and 2  (SATA 6Gb)
two WD Green sata II 1TB drives in a RAID0 in sata ports 3 and 4  (SATA 3Gb)
one WD Black sata III 1TB drive in sata port 5 (system resides on this non raid drive)
Bluray drives in sata ports 6 and 7
Corsair 750W 80 PLUS Power supply

Please reduce the complexity of your system to the most basic configuration needed to run your operating system.  Use one stick of RAM only, make sure it is set to run at a frequency that does not exceed the specifications of the processor's memory controller (DDR3-1333 max.!).  Unplug all the hard drives, optical drives, SSDs, card readers, add-on cards, USB devices etc. that are not needed to work with the operating system (you need just mouse, keyboard, video card, processor, one stick of RAM).

Once you have reduced the system complexity and please unplug your PSU from A/C power, >>Clear CMOS<< and then load "Optimized Defaults" in UEFI Setup.  Then apply your custom settings (e.g. RAID Mode settings  & Memory Speed = DDR3-1333).

Then retest.

In case the freezes still occur, then please disable all CPU related power saving settings in UEFI Setup (C1E, C-States, EIST) and retest.  Also, try if setting the phase control mode from "APS" to "Intel SVID" mode.

 

[HU16E]

Jack's recommendations & suggestions of course, but I have added some additional thoughts. Seems things got off to a rocky start, but the goal of a new thread was to assist you specifically with your freeze up problem.

My memory is even certified for this damn board on bios 1.B which I have updated to.  But in "auto"  doesn't even get the memory timings correct.

Excerpt from the G.Skill link; "*G.Skill guarantees 2 Dimms dual channel operation would reach announced specification." Spin doctering at it's finest. I'm sure they do in a controlled lab test. :lol_anim:

The timings set correctly? SPD profiles govern that. Auto (default) for the Sandy Bridge platform as currently released is native support for CPU IMC of DDR3 1333 RAM speed. Typical standard timings for DDR3 1333 are 9 9 9 24, tRFC 74-88 or so @ 1.50v's. The CPUID utility CPU-Z should show timing information in the SPD Tab. Please try setting the RAM to 1333 with the timings & voltage I mentioned manually in the BIOS. CR can most likely be set to 1, but only testing will tell.

two WD Black sata III 1TB drives in a RAID0 in sata ports 1 and 2  (SATA 6Gb)
two WD Green sata II 1TB drives in a RAID0 in sata ports 3 and 4  (SATA 3Gb)
one WD Black sata III 1TB drive in sata port 5 (system resides on this non raid drive)
Bluray drives in sata ports 6 and 7

Possibly the SATA devices arrangement may be a freeze factor. Some feel the Intel controlled SATA III ports to be better than the Marvell controlled ports. It may be worth a try of placing the OS HDD into the white Intel controlled SATA III port #1. The SATA III WD Blacks in RAID in the white Marvell controlled SATA III ports 7 & 8. Presume the SATA III WD HDD now in a SATA II black port #5 has a jumper installed to work properly? Unless WD has made a change recently to their HDD's, a jumper was required for use in a SATA II port. Curious as to why hamstring a SATA III HDD that has the OS on it in one of the slowest ports? Again, that SATA III HDD may need a jumper if it is to remain to be used in a SATA II port? Blurays in either remaining available SATA II or SATA III ports should both work the same.

 

[sccarlso]

Excellent!  Thank you for the frank and disarming post as well as tips for posting specifications      Got your PMs, but I'm not allowed to send PM's apparently, so I couldn't reply.

earlier this morning, after reading posts from other boards, links between different forums, it was back to the bios settings, and it was still misdetecing the timing in auto mode.  So I set it to "link" from auto and changed the timing from 9-9-9 to 9-10-9 which the memory timing is for this Gskill memory.  Also I manually set the memory to DDR3-PC1600 even though it detects this correctly as DDR3-1866 when in the auto setting, as I bought higher spec memory for overclocking stability than what was required.

And so far, no more lockups (fingers crossed).

After I made this change I did an uncompressed 1080p capture with a USB 3.0 capture box from my sat receiver and it captured 2 1/2 hours of video at a whopping filesize of 1.5TB without dropping a single packet.  Then I compressed it to AVCHD which put all the cores to 98% utilization for 5 or 6 hours, which took the filesize for the video down to 15GB.  But the big part...no lockup, and I can RDP without those strange errors.

Appreciate the info!  And something to pass on to anyone getting lockups and instability when not overclocking on this board, check the memory timing!  I don't know why the board fails to autodetect it, but whether it's Gskill or MSI, hopefully another bios will be released so these folks, including myself don't need to spend days trying to isolate why their brand new machine is locking up randomly.

Thanks again!
Scott

 

[HU16E]

Appreciate the info!  And something to pass on to anyone getting lockups and instability when not overclocking on this board, check the memory timing!  I don't know why the board fails to autodetect it, but whether it's Gskill or MSI, hopefully another bios will be released so these folks, including myself don't need to spend days trying to isolate why their brand new machine is locking up randomly.

Many of us tend to mostly blame the RAM manufacturers coming up with non standard 'off the wall' SPD profiles. :lol_anim: But glad to hear your rig is up & running strong now.

 

[sccarlso]

Ya, I would agree.  It's more likely to be the RAM manufacturers than MSI.

When I first got the board I didn't have lockups, but it had just one HD and a bluray and I had not put any kind of stress on it other than loading the OS.  But once the OS was loaded up, when i enabled the OC genie to "see what it could do", it would still boot, but if I ran any kind of CPU performance test or full CPU load, it would crash, even at 4Ghz. 

So I'll do a few more video captures and encoding/compression (which is what this machine is for) to ensure it's stable at standard speeds, then I might try to OC a bit and see if I can at least get to 4Ghz and stay there.  If I had no intention of OCing, I wouldn't have bought a 2600k eh?

BTW, I've never seen so many possible emoticons anywhere else  lol

Thanks again!
Scott