Is the Gigabyte Z68X-UD7-B3 the source of a world of hurt?

[meat juice]

Re: Is the Gigabyte Z68X-UD7-B3 the source of a world of hurt?

Hey Jim, thanks for responding. I'm not running any of the disks in an array. They're all JBOD. This problem persisted prior to the WD Blacks. I had a bunch of WD Greens awhile back. I noticed 4 out of the 12 Greens were in different states of failure. WD replaced them with these Blacks. The latency I described persisted. I replaced the PSU, GPU and MoBo and that did not make a difference either. I installed the OS on one of the Blacks and the problem still persisted. I did run the system with minimal hardware with the OS on both the SSD and a HDD. Any thoughts?

[profJim]

Re: Is the Gigabyte Z68X-UD7-B3 the source of a world of hurt?

Please copy your full hardware list from post #1 and add it to your forum signature.
This makes it much easier for us to quickly and easily help you.

Just to clarify with a minimal (1 or 2) hard drive setup, does every drive have a problem when connected to the Intel sata ports?
I would avoid using the Marvell sata ports until you get things straightened out.
Have you disabled all unused hardware in your bios?
Have you tested with your motherboard sitting on a cardboard box beside your computer case?
Did you start with loading Optimized Bios Defaults before you make any bios changes?
What bios version are you running?

I don't know if this will help, but maybe using a N.A.S box might help, but only after your system is performing properly with only a couple of hard drives.

[meat juice]

Re: Is the Gigabyte Z68X-UD7-B3 the source of a world of hurt?

To clarify,

As part of the trouble shooting process I had tried to run the SSD in both the Intel controller port 0 and port 1 with no other hardware attached except a stick of RAM. I found at times it would take forever to open the recycle bin (a problem that still persists). I also installed the OS on a Black and ran it in the same configuration and had the same problem. Accordingly I reverted the OS back to the SSD. The problems of latency seems to occur on all disks including ones attached via either eSATA, Intel, Marvel or Highpoint controllers.

I have disabled all unused hardware in the BIOS and recently C3/C6 as it would BSOD randomly when enabled. Thanks to Muchkin Sean for that tweak. :)

I have not run the board on a cardboard box. I did recently change cases. Problem is the same in either case. I'm upgrading my H2O loop tomorrow so if there's a need I could pull the mobo out and run it on a box. This would be a serious pain in the ass though.

I don't think I've used Optimized BIOS Defaults ever and not before any BIOS changes because if I'm not mistaken the machine would not boot using that setting. It's an issue I'm open to taking another look at though. I have certainly cleared CMOS a dozen times on both this board and the previous. All hardware has the latest firmware and drivers. I would like to hear more about Optimized BIOS Defaults as I see you've mentioned it in your signature. Can I do it now and if so do I clear CMOS or re-flash the BIOS? I could flash toF9 as I've abandoned the problematic Adaptec 2805 controller. Could write a book on that experience. Kid of excited to get some guidance on this as it's a fresh idea.

I tried a NAS solution via the eSATA back panel but Windows could never see the disks. I took it to a shop and it worked fine on their system. It's been awhile but I think I read there was a conflict issue with the Z68X-UD7-B3. Not sure about that though. Either way I returned the NAS. I could certainly try a different NAS. I've resisted this however as my current configuration should work and it's really the most attractive option. Accordingly, if I discovered it's an issue with the Z68 chipset or this particular manufacturer then I would change boards or whatever hardware is causing this.

As it stands the only part that hasn't been replaced to address this issue is the CPU and the RAM. RAM has been tested to death at both default BIOS and recommended manufacturer settings. Since I disabled the C3/C6 the system has been rock solid. Just this god forsaken latency issue.

[profJim]

Re: Is the Gigabyte Z68X-UD7-B3 the source of a world of hurt?

Flashing with the most recent bios might help.
I would wait to test your motherboard outside of the case until the last step before you RMA the motherboard.
You can load Optimized (or Fail-prone, uh, Fail Safe) defaults at any time, but you will have to manually set any required bios settings. If you do a clear CMOS, you should reload the Optimized Bios Defaults and then repeat any required bios changes.

A test that might be helpful is to boot into Safe Mode and see if you system still acts up. The next test could be to boot using window's Diagnostic Mode which will boot into windows but only with a minimal amount of drivers and services loaded. There might be some helpful information in the windows Event Logs by typing View Event Logs in the task bar's search box.

I would test with a bare minimum hardware setup, with only enough hardware to run windows; one memory module, a dumbed down old video card, no usb, no audio card, etc. It would be great if you could test a different cpu, just to rule that out. You could run windows resource monitor from the Task Manager main screen and watch the performance graphs for any irregularities. You could do a full system virus scan to see if there are any issues while it's running. It sounds like it doesn't matter if you're o/s is running with an HDD or SSD.

Have you verified that your temperatures are reasonable while you're in windows?
I couldn't see what type of cpu cooling setup you're using. Your hardware list is like one long run on sentence. Please spend a few minutes and make it more readable.

**edit** If you can't borrow another cpu to test with, you might want to spend $100 on a new cpu and turn around and sell the cpu after you're done with the testing. This would be a lot cheaper than paying a local shop to do this test.

[meat juice]

Re: Is the Gigabyte Z68X-UD7-B3 the source of a world of hurt?

I would wait to test your motherboard outside of the case until the last step before you RMA the motherboard.

I bought another Z68X-UD7-B3 a few days ago. The old one is on the table. I didn't want to wait weeks for an RMA.

You can load Optimized (or Fail-prone, uh, Fail Safe) defaults at any time, but you will have to manually set any required bios settings. If you do a clear CMOS, you should reload the Optimized Bios Defaults and then repeat any required bios changes.

Done. I'm afraid it made no difference. Given the Adaptec 2805 is not going to be used again I'll try vF9 BIOS tomorrow.

A test that might be helpful is to boot into Safe Mode and see if you system still acts up. The next test could be to boot using window's Diagnostic Mode which will boot into windows but only with a minimal amount of drivers and services loaded. There might be some helpful information in the windows Event Logs by typing View Event Logs in the task bar's search box.

I've tried booting in safe mode before and the issue does not occur although to be honest I haven't stayed in safe-mode long. Also tried diagnostic mode. No glory. I shall revisit this suggestion tomorrow and allow a reasonable amount of time as I can't recall specifically what I did.

I would test with a bare minimum hardware setup, with only enough hardware to run windows; one memory module, a dumbed down old video card, no usb, no audio card, etc. It would be great if you could test a different cpu, just to rule that out. You could run windows resource monitor from the Task Manager main screen and watch the performance graphs for any irregularities. You could do a full system virus scan to see if there are any issues while it's running. It sounds like it doesn't matter if you're o/s is running with an HDD or SSD.

Have tested with bare minimum hardware just as you've described including a lesser Radeon 6850 video card. Had to use USB for the keyboard and mouse though. The only way for me to test the CPU is to RMA it. Intel says they can have a replacement overnight. Might do that on payday if this isn't resolved by then and follow your instructions related to performance graphs. Ran without anti-virus installed and it was still symptomatic.

Have you verified that your temperatures are reasonable while you're in windows? I couldn't see what type of cpu cooling setup you're using. Your hardware list is like one long run on sentence. Please spend a few minutes and make it more readable.

Sorry for the illegibility. Under heavy load GPU/CPU temperatures have never exceeded 50c and 40c respectively.

[profJim]

Re: Is the Gigabyte Z68X-UD7-B3 the source of a world of hurt?

Even without doing a long virus scan, keep the Resource Meter graphs visible to see if there is any high activity, especially when your system seems to bog down.
Keep a simple time log of when windows started and what you did while watching the graphs. Note the time if/when the system bogs down. You might be able to tie your observations to warnings or errors in the event log.

Make sure that your network connection is disabled while testing.
Run HWmonitor just to verify that there aren't any temporary temperature spikes.
My preference is NOT to run EasyTune6 for monitoring, especially when you're trying to diagnose problems.