odd memtest 86 results

2233 / 3344
December 1, 2009 at 17:54:38
Specs: Windows XP, 4gig
i have had trouble diagnosing my computer because many problems may have occurred at once.
it started a few months ago when my moniter would randomly change colors and shapes after warming up. you could see the mouse move, but it did not look like a white arrow, it was a large square with many changing colors inside.
then eventually the computer would not boot, nothing would appear on the screen when the power button was pushed.
i described the problem to many experts and they all came to the same conclusion, that it either had to be the system memory or the graphics card memory.
i ordered a new ati 5750 graphics card and it arrived today. i slipped it in and the computer appeared to boot properly.
This means that the graphics card was deffinately bad.
But there is at least one more problem.
I began installing the drivers.
i had read that there were problems with this driver on windows xp, but that they had recently solved the problem with the release of catalyst 9.11, and i could not find anyone on the many many sites i visited who could say that they still had problems with windows xp after installing catalyst 9.11.
but for me, the installation halted after the file extract every time. i would click install, and the catalyst window would disappear, and when i looked up the installer process in the task manager, it was consuming 0% of total processing power. i waited patiently but eventually rebooted and tried installing over and over again, both from the xfx cd included with the graphics card and from an installer downloaded from ati's site with the same results every time.
after 5 tries a window appeared displaying the message "windows hard error" i kept xing it out and it appeared again and again. i tried to reboot and the computer would not. the computer now refuses to boot from the hard disk, it just continually says "system error, insert boot disk and press enter."
so i ran memtest86, thinking that perhaps my memory was also bad
with both of my sticks in, memtest 86 completes the walking ones test, but then reboots at the 66% marker of the "own address test"
with one stick in, it completes all tests with no errors.
i have tried all combination of stick placement, the system refuses to boot from the hard disk.

See More: odd memtest 86 results

Report •

December 1, 2009 at 17:56:17
mobo: gigabite ep35 ds3l
memory: 2x2sticks ddr2 1066 ocz
graphix: xfx ati 5750
power: x-power 585 watts

Report •

December 1, 2009 at 18:23:20
What do you have the RAM voltage set at?

Report •

December 1, 2009 at 19:55:07
You have to test the memory in dual channel to get correct results. My take on memory is if it fails at all it is bad as long as you have bios set to default or failsafe.

As to boot from hard drive it is not a good test anymore. Bad memory could have messed up the install. Use a live cd. Use memtest off cd or floppy or usb. Don't use Memtest86+ V4 as it doesn't properly test dual channel.

Get ultimate boot cd to test system maybe. Memory has to work for 24 hours or so then go on to other tests.

If all goes well you may have to do a repair on the xp install then work on the new adapter.

Playing to the angels
Les Paul (1915-2009)

Report •

Related Solutions

December 1, 2009 at 20:55:40
if i can get no errors testing 1 stick at a time then how can i get errors with both sticks in?
i will get ubcd and try other tests but as of now i dont see how it is possible that the memory is bad

Report •

December 2, 2009 at 14:12:53

Known problems

There are two versions (or "streams") of Memtest86. One is simply known as Memtest86 (current version 3.5) and the other is known as Memtest86+ (current version 4.0). They both have almost identical screen appearance. The 3.5 version is known to fail on some (or many, or most) computers with 4 GB of installed memory (the failure results in a spontaneous system reboot soon after the program starts running). The Memtest86+ 4.0 version does not exhibit this behavior. Also, there is a single-core and multi-core version of the Memtest86 3.5 version, and the multi-core version has been observed to not function on systems with AMD nor Intel Quad-core CPUs.

ok so that explains the repeated reboots. i have 9 passes on one stick without errors and 27 passes on the other without errors. that must mean my memory is good right? and amd wouldn't release software that would corrupt my OS all by itself right? so then whats wrong with my computer????

Report •

December 3, 2009 at 01:45:50
im beginning to think that ati WOULD release a driver that would corrupt my OS :(

Report •

Ask Question