Jump to content

Welcome to Geeks to Go - Register now for FREE

Geeks To Go is a helpful hub, where thousands of volunteer geeks quickly serve friendly answers and support. Check out the forums and get free advice from the experts. Register now to gain access to all of our features, it's FREE and only takes one minute. Once registered and logged in, you will be able to create topics, post replies to existing threads, give reputation to your fellow members, get your own private messenger, post status updates, manage your profile and so much more.

Create Account How it Works
Photo

Gigabyte Blues


  • Please log in to reply

#1
DataJoe

DataJoe

    New Member

  • Member
  • Pip
  • 5 posts
Perplexing Gigabyte Mainboard

For anyone who likes to troubleshoot mainboards, here is a good one. I have spent days trying to resolve the problem, but to no avail. Any insight or help will be appreciated. BTW, my brother gave me this board as he said that it became unstable. I would have to agree with him. My plan is/was to put the board into my secondary desktop system.

It is my conclusion that the board hates Windows XP, however loves Windows 98/Me, but the chipset is not even designed for the old O.Sís. For what itís worth, there are no problems, no instability at all if I connect a drive running Win 98 or Win Me to the board.

Mainboard - Gigabyte GA-K8NS Rev 2.0, AMD Socket 754, nVidia Chipset (circa 2005)

Note that the actions listed below are not in any particular order.

The first thought was that because the Gigabyte had several domed top (one leaking) capacitors (3300uf/6.3v), that it was the cause of the instability problem. So with a solder tool in hand, I went to work and replaced them all. The surgery was successful in a sense that the board is still fully functional. And the same problem that existed before, still exists afterwards. Gigabyte boards (from 2002-2006) were apparently plagued with many bad capacitors.

Used 4 different DDR memory modules of different sizes in different slots - all had passed using MemTest86.

Flashed the BIOS chip to the latest 21a version (10-2008).

Must have changed the BIOS settings a hundred different times.

Replaced the 3-volt CR2032 mainboard battery with a new one.

Used 3 different graphics cards, including one old PCI S3 to bypass AGP.

No other cards were installed on the board.

The processor is an AMD 754 - 3200 - pulled and inspected, but had no replacements to use.

No shorts on the board as it was mounted on a heavy piece of cardboard.

Tried 3 different Power Supply Units.

Most of the time it would boot using Safe Mode, but always failed if a Normal boot. And after 5 or 6 attempts of Normal booting which would fail at the Desktop screen just after the icons were displayed, it would blank the screen and reboot itself, and eventually it would corrupt the MBR which I had to restore - last count 4 times.

Did a complete, fresh reinstall of Windows XP Pro and no problems were encountered during the installation, but no change in behavior. The ntbtlog.txt reveals little information at least to me. Left the board running overnight in Safe Mode, still fine the next day.

Downloaded and installed the latest nVidia chipset drivers for the board - didnít expect any change and didnít get any.

Oh, used two different hard drives. Chkdsk gave them a clean bill of health. Then I took one of the drives with Win XP and reconnected it to my other system, booted just fine and Iím using it now to compose this text.

A guess would be a memory address error.

Thatís about it for the highlights. Any assistance will much appreciated.

Thanks, DataJoe
  • 0

Advertisements


#2
Digerati

Digerati

    Grumpy Ol' MSgt (Ret.)

  • Retired Staff
  • 3,997 posts
  • MVP

Gigabyte boards (from 2002-2006) were apparently plagued with many bad capacitors.

It was industry wide, not just Gigabyte.

So after replacing the caps, the board works fault-free with Win98? I note that Win98 uses FAT32 while XP uses NTFS (normally). I wonder if there is not something wrong with the controller that does not like NTFS?
  • 0

#3
DataJoe

DataJoe

    New Member

  • Topic Starter
  • Member
  • Pip
  • 5 posts
Interesting concept about FAT32 vs NTFS partitions, though the probability is not very high. However, I will add the item to my list of future testing considerations. Thanks for the tip. Any other thoughts?

I still lean toward a memory address problem, or secondly, a Windows installed driver conflict. Oh, I previously neglected to mention a point. During a couple of the boot ups, I chose for Windows not to automatically restart after a fatal error, but there was nothing much on the blue screen **STOP: 0x0000008E. I also saved some of the Page Dump Files, although not much can be read in Notepad. Is there a program that can be used to fully read the Windows .dmp files?

One other point to mention - I had a beta version of Windows 7 on another hard drive and I temporarily connected it to the Gigabyte board. Same type of problem and/or in addition to; however, on the blue screen there was a description of the error: Page_Fault_In_Nonpaged_Area. A quick Google search revealed most common occurrence was due to driver memory conflict/incompatible software. This maybe true since the Gigabyte board, chipset, graphics card are nearly five years old. However, I will not be using either Windows 7 or Vista with this board.
  • 0

#4
DataJoe

DataJoe

    New Member

  • Topic Starter
  • Member
  • Pip
  • 5 posts
Okay, I installed the MS dumpchk.exe and had it to analyze the minidump file. Here is the last part of it:

Unloaded modules:
f09b9000 f09c4000 imapi.sys Timestamp: unavailable (00000000)
f09c9000 f09d8000 redbook.sys Timestamp: unavailable (00000000)
f09d9000 f09e3000 amdk7.sys Timestamp: unavailable (00000000)
f5062000 f5067000 Cdaudio.SYS Timestamp: unavailable (00000000)
f09f9000 f0a09000 cdrom.sys Timestamp: unavailable (00000000)
f12a8000 f12ab000 Sfloppy.SYS Timestamp: unavailable (00000000)

Finished dump check

Why is Windows XP attempting to load the amdk7.sys file? The AMD 754 processor is classified as a K-8. I did a search, but could not find anything conclusive. It would seem that any attempt by the O.S. to load a processor driver that is not compatible would indeed cause the problem as outlined above. And insights would be appreciated.
  • 0

#5
rshaffer61

rshaffer61

    Moderator

  • Moderator
  • 34,114 posts
Your stop code has several reason for it's existence as shown HERE. Also if you would like to test that memory theory then I suggest you do the following.

If you have more than one RAM module installed, try starting computer with one RAM stick at a time.

NOTE Keep in mind, the manual check listed above is always superior to the software check, listed below. DO NOT proceed with memtest, if you can go with option A

B. If you have only one RAM stick installed...
...run memtest...

1. Download - Pre-Compiled Bootable ISO (.zip)
2. Unzip downloaded memtest86+-2.11.iso.zip file.
3. Inside, you'll find memtest86+-2.11.iso file.
4. Download, and install ImgBurn: http://www.imgburn.com/
5. Insert blank CD into your CD drive.
6. Open ImgBurn, and click on Write image file to disc
7. Click on Browse for a file... icon:

Posted Image

8. Locate memtest86+-2.11.iso file, and click Open button.
9. Click on ImgBurn green arrow to start burning bootable memtest86 CD:

Posted Image

10. Once the CD is created, boot from it, and memtest will automatically start to run.

The running program will look something like this depending on the size and number of ram modules installed:


Posted Image

It's recommended to run 5-6 passes. Each pass contains very same 8 tests.

This will show the progress of the test. It can take a while. Be patient, or leave it running overnight.

Posted Image

The following image is the test results area:

Posted Image

The most important item here is the ďerrorsĒ line. If you see ANY errors, even one, most likely, you have bad RAM.
  • 0

#6
DataJoe

DataJoe

    New Member

  • Topic Starter
  • Member
  • Pip
  • 5 posts
Kudos to Digerati! :) The assumption that the mainboard controller might not like a NTFS partition appears to be correct. I repartitioned and reformatted the drive using FAT32, loaded the basics and with the same configuration Voila! it now boots in Normal mode and is running fine. Go figure! I obviously never experienced a similar problem. Why would Windows XP fail to fully boot because of a partition type issue and especially with the one that was designed for its kernel? Not to mention that the Gigabyte board, controller and drivers were designed for Windows XP too. I realize that there maybe more issues behind the scene than the NTFS vs. FAT32 one, i.e., Windows XP attempting to load the amdk7.sys file and not the k8 one. So, if anyone can shed some light on this point, it would be appreciated.

Oh, I agree with rshaffer61 that MemTest86 is a superior and a very useful tool to test the intergrity of memory modules and which I had previously ran on all modules used.
  • 0

#7
Digerati

Digerati

    Grumpy Ol' MSgt (Ret.)

  • Retired Staff
  • 3,997 posts
  • MVP
Good glad you got it figured out. As to why you had problems, my guess is a faulty device. Until Man can create perfection, there will be faulty devices.
  • 0

#8
DataJoe

DataJoe

    New Member

  • Topic Starter
  • Member
  • Pip
  • 5 posts
An update - I repartitioned and reformatted my secondary hard drive in NTFS then installed Windows XP in progression, Original, SP1, SP2 and after each installation performed various testing. The results: No Problems! I'm holding off installing SP3 because of the following (if I only knew this earlier):

When installing Windows XP SP3, some AMD64 systems shipped by OEMs with XP preloaded, primarily from Hewlett Packard but also with Mesh Computers, boot up problems have also been confirmed, where the preloaded software image was originally created on an Intel based computer then prepped for use on AMD systems by a method that is unsupported by Microsoft. This can affect any AMD64 motherboard used by the OEM, not just ASUS.

There is a different issue that appears unique to the ASUS A8N32-SLI motherboard, resulting in a critical stop error and BSOD with 'BIOS is not fully ACPI compliant' message. This is completely unrelated to the Intel processor driver (IntelPPM.sys) implicated in the OEM 'preload' issue above. The OEM preload related issue does NOT implicate 'BIOS ACPI compliance' as the problem. It reports a different 'general' stop error due to some unspecified hardware problem, usually 0x0000007E.

I would conclude that my Gigabyte board is experiencing a similar type of problem with SP3. I may need to spend a little time researching this matter further to find a final solution. Also, I have discovered that Stop: 0x0000008E problem can occur because of a third-party driver that may prevent the opening of files successfully on a NTFS file system volume. The NTFS file system volume expects a non-null value for stream control block (SCB). However, this value is null.
  • 0

#9
Digerati

Digerati

    Grumpy Ol' MSgt (Ret.)

  • Retired Staff
  • 3,997 posts
  • MVP
That SP3 problem was caused by HP (and a couple other computer makers) making unsupported changes to the image file they used to install (or recover) Windows on factory installed hard drives. Are you saying that Gigabyte motherboard came installed in an HP or Mesh computer and that you do not have an genuine Microsoft installation disk for Windows? In other words, that problem is only a problem if the installed version of Windows was installed by the factory, or by using an HP (or Mesh) "recovery disk" or "recovery partition". If that is not the case, then I recommend you install SP3.

For more information, and a fix for that problem, see http://msinfluential...ing-xp-sp3.aspx
  • 0






Similar Topics

0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users

As Featured On:

Microsoft Yahoo BBC MSN PC Magazine Washington Post HP