Official GIGABYTE Forum
Questions about GIGABYTE products => Motherboards with Intel processors => Topic started by: Svarog on October 19, 2011, 08:33:34 pm
-
Hi all,
First all i'm new here so i couldn't really find a suitable forum section to put this in.
I have a strange problem which is going on for a couple of weeks/months now. Like every 1 or 2 weeks my PC decides to shut itself down. It simply shuts itself down and i have to press the power button again, it boots up normaly again without giving an error message after startup. I checked the Windows Log files a couple of times aswell at the time it shut down but it doesn't really tell anything that could have caused it.
Here is my PC Specs:
CPU:
Intel Core i7 950 @ 3.66 Ghz (166 x 22)
Voltage: 1.125 Normal, DVID +0.125 (Load-line Calibration Standard Intel Spec)
Cooler: Noctua NH-C14 Push/Pull 750 RPM
Temps: Idle 37C - Load (Gaming) 65C
Motherboard:
Gigabyte X58A-UD7 (Bios FC)
Memory:
GeIL DDR3 Ultra Tripple Channel 3x 2GB (7-7-7-24) (Voltage needed 1.65, set to 1.66)
(XMP 6-6-6-24 is unstable, BSOD Randomly)
Videocard:
Zotac GTX580 (Stock) (Arctic Accelero Xtreme Plus)
Installed in the first PCI-E 16x Slot
nVidia Forceware 285.38
Soundcard:
Creative X-Fi Fatal1ty Titanium Pro
Harddisks:
OCZ Vertex 120GB (Firmware 1.7) (Windows Drive)
OCZ Vertex 120GB (Firmware 1.7) (Games Drive)
Samsung EcoGreen F1 HD103SI 1Tb (Storage)
All connected to the Intel controller
Power Supply:
Seasonic X750 GoldPlus 750 Watt
OS:
Windows 7 Ultimate 64-Bit
What i tried so far:
- Reinstall of Windows.
- Different nVidia Drivers.
- Swapped around memory Modules.
- Tested each Memory Module with Memtest.
- Dropped the SSD Raid-0 and use them separately.
I probably tried some more but i kinda forgot. PC is behind a Back-UPS of 865 Watt.
I currently swapped my GTX580 from PCI-E 16X_1 to PCI-E 8X_1. Yes, it's a 8X slot but performance wise it doesn't matter. And the soundcard went from PCI-E 8X_2 to PCI-E 1X_2. I also disabled HPET. Since i cannot reproduce the problem it's a matter of waiting.
Last time it happend was today at 11:43 the time before that was 11 days ago. It's quite a rare issue and i hardly doubt it's a hardware defect. It never happens during heavy load.
Other things ive been thinking about is clearing the CMOS by removing the battery and perhaps even re-flashing the BIOS.
After that i'm pretty much out of ideas.
If anyone has any ideas of what it could be or what i could try to figure it out id be very thankful.
-Svarog
-
Welcome Svarog ....... not very good for raid this problem Huh!.......let's try and fix it so you can get your Raid back!
You have done a thorough job Thus far and a UPS rules out brown outs (Short black outs).......probably the first thing would be bring the BIOS up to date as you're two behind running FC (I'm assuming this is ver2 board) ......FD9 is the current.........FC was 2010!
When you said ....- Tested each Memory Module with Memtest.......(You're gunna hate this) did you do 10 full passes each module ?
Very prudent to check your firmware on the OCZ SSDs ...... Here's a handy link to a PDF about flashing to the latest firmware version.......
http://www.ocztechnology.com/resources/drivers/Vertex_Firmware_Flashing_Guide.pdf
In a nutshell........Flash both the mainboard and the SSDs and run them for a week or two........if it still throws a wobbly......come back and report your finding........we'll be here! ;)
Aussie Allan
-
I currently have the latest Firmware on my SSDs.
About the Motherboard BIOS. FC is the latest Non-Beta according to the website. There is a FD8 and a FD9 and they are both Betas. I guess i could flash them to that anyways.
And ye, i see i forgot that. It's Rev 2.0.
EDIT:
Just flashed to FD9 with @BIOS. I also checked the box "Clear DMI data pool" incase that helps.
I'm kinda wondering what the changes are because all it says is Beta BIOS.
And from now on it's a matter of waiting as i cannot reproduce the problem. So it might take another 2 weeks untill it happens, and ofcourse "IF" it happens again ... i hope not ofcourse ^^
-
Hi
I hope that your BIOS update has indeed worked alright as @BIOS is not to be recommended. It is much more likely to fail and it is much safer to use QFlash for updating when needed.
If you have got away with it then fine just leave well alone otherwise repeat the procedure using the QFlash utility and do as you previously did by disabling the Keep DMI Data option.
-
Beta Bios seem to be the norm now .... on my board the last 3 Bios updates are all Beta and have all been there for many months now.
No official explanation, just how they seem to do things at the moment (going back a year)
What is a BETA?
BETA describes a new version that is reliable yet may not include all the features of the final product. During this phase we are previewing new features and gathering customer input to insure our product provides the best experience possible.
I think that this is the statement at the bottom of the Bios update page that is making a lot of people uneasy and hesitant to install the latest updates.
My feelings are if Gigabyte have it up for loading....it's good enough for me! ...... you can always do a little experimentation by looking at the French site "Station drivers"......quite often they have gigabyte driver and Bios updates posted before Gigabyte :D
Aussie Allan
-
Hi
I hope that your BIOS update has indeed worked alright as @BIOS is not to be recommended. It is much more likely to fail and it is much safer to use QFlash for updating when needed.
If you have got away with it then fine just leave well alone otherwise repeat the procedure using the QFlash utility and do as you previously did by disabling the Keep DMI Data option.
Ive been using @BIOS when i Flashed to FC aswell. Aswell as on previous Boards, it always worked fine for me. But thanks for the heads up :)
And as i said before, it's now a matter of waiting. I really hope i get it sorted out. It's starting to get annoying that once every 10-12 days it just goes off like snapping with ur fingers.
-
Beta Bios seem to be the norm now .... on my board the last 3 Bios updates are all Beta and have all been there for many months now.
No official explanation, just how they seem to do things at the moment (going back a year)
What is a BETA?
BETA describes a new version that is reliable yet may not include all the features of the final product. During this phase we are previewing new features and gathering customer input to insure our product provides the best experience possible.
I think that this is the statement at the bottom of the Bios update page that is making a lot of people uneasy and hesitant to install the latest updates.
My feelings are if Gigabyte have it up for loading....it's good enough for me! ...... you can always do a little experimentation by looking at the French site "Station drivers"......quite often they have gigabyte driver and Bios updates posted before Gigabyte :D
Aussie Allan
I don't really have any issues with installing a BETA Bios or Drivers or anything. They just don't show a Change Log anywhere. So that kinda kept me from installing it :)
-
Change logs are now a pretty hit and miss affair nowadays with gigaboards......I agree......it would be nice to be kept up to date with the changes there tackling in the Bios updates.
Aussie Allan
-
To be honest they have never been very helpful as with many other makes. I don't know why though as it doesn't take much effort jkust to note down a few pointers.
-
Well, it just shut down again at 13:37 while i was gaming. I have no clue what's going on anymore, i'm out of options now.
Been 8 days now since last time. Can't be a hardware defect...
Can try to run CPU at stock, that's the only thing i can try atm.
-
Download and install a temp app that is active somewhere on the desk top and monitor to see if that could be related ?
Aussie Allan
-
Download and install a temp app that is active somewhere on the desk top and monitor to see if that could be related ?
Aussie Allan
Well it also happens during Idle operation. So it's not a heat issue. I do have CPUID HWMonitor installed and i just ran some DiRT 3 Benchmarks at Max Settings CPU Temp Max is 58C and Videocard 56C.
I did discover something strange tho.
The Seasonic X750 has two ways of connecting the CPU 8 Pin. One can be used simply on a seperate channel which is 8 Pin to 8 Pin (That's how i had it connected). The other way is on one of the two PCI-E connections which goes from 12 Pin to 8 Pin (Only 8 Pin wired ofcourse), i got it connected like this now. And i noticed that the CPU temps are like 5-8C lower when connected this way.
Apart from that i rewired the PCI-E aswell, basicly just swapped the 8 Pin wires around, maybe that makes a difference aswell.
Still i rule out a hardware defect, but perhaps the wiring is the issue. Again, it's a matter of waiting now, last time it took 8 days so it could be longer now.
Here is a picture of the PSU connections:
http://www.pcstats.com/articleimages/200911/seasonicX750_d2.jpg
As you can see there is PCI-E/CPU 8P which can also be used for 2 x PCI-E. And next to that is another CPU. So you can either connect 1x 8P and 2x PCI-E for SLI. Or 2 x 8P and 1 x PCI-E.
Now previously the CPU was connected to the CPU one, but now i plugged it into the CPU 8P one.
-
Something else which crossed my mind.
The PCI-E clock speed is at Auto which is 100 Mhz. Would it hurt to up it to 101 or 102?
-
This setting is generally only increased when two or more GPU cards are in a given system.......increasing this setting should be approached with caution......but can aid in stability
Aussie Allan
-
Another reason for not altering the PCIE clock speed is that it will be one more variable to take into account when you already have a problem. It will be hard to be sure if any further issues are due to the original problem or the change of settings.
-
Seeing the issue pretty much started happening when i swapped from HD5970 to GTX580, it's clear i have to search there. If the rewiring doesn't make any difference i got 2 more tricks up my sleeve.
- Up voltage from 1000 mV to 1025 mV (Perhaps a notch higher to 1038 mV).
- Force Constant voltage in MSI Afterburner so it doesn't go up and down while Adaptive Mode is enabled.
-
Seem to have happend again today at 15:50 and Windows Log was flooded with NTFS errors. PC still runs fine tho, but i got the feeling it's one of the SSDs after all.
Ive already been looking for some other SSD. Non-SandForce and Non-OCZ. I can't be bothered with RMA either and end up with another bad sample. Currently looking at a Crucial M4 256GB and go Single Drive aswell.
Posted on some forums about the Crucial, i basicly wanna know if they are reliable.
-
So far replaced Memory and SSD, no deal. This is getting annoying.
Right now i'm about to RMA my PSU, that's the only other Hardware that's highly likely to cause the problem. Basicly upgrading PC with parts 1 by 1 and t's very frustrating and expensive.
If anyone has other ideas let me know.
-
Well the problem is solved, seems it was the PSU.
Since i didn't have a replacement PSU i bought a new SeaSonic X760 and ever since then it's gone. I will RMA my SeaSonic X750 for a replacement and see if i can sell the new unit :)
Thanks everyone who offered help and suggestions. It been a long couple of months and i'm glad it's finally solved.
-
I am happy to hear that you have finally managed to get to the bottom of the problem and you would be surprised how often the PSU is at the bottom of the troubles. I suppose that because yours was a good make and a decent wattage it kind of got discounted as the cause. Just goes to show that anything can be fauty. :-\