WTF is going on with my computer?

[Stryfe]

I've recently been getting a ton of errors with my computer and was wondering what they are. They all have the same basic format:

Application Error : The instruction at "(memory_address)" referenced memory at "(memory_address)". The memory could not be "written" (or "read").

Click on OK to terminate the program.
Click on CANCEL to debug the program.

It's not just a one-time occurance, either. It has happened 34 times in the last 20 days.

Here's a list of the errors in no particular order.

Also note that my memory seems to be going. Evidence can be found in the reports generated by GoldMemory:
- gm.txt
and DocMemory:
- results1.txt
- results2.txt

Could that have anything to do with the errors?

[edit] Could it also possibly be a bad (or going bad) mobo, controller, HD, etc? [/edit]

--Stryfe

[Johnny Blaze]

Sound like bum RAM to me. Try swapping with a spare to see if the problems dissapear

[ToX_Boy]

How many sticks of memory do you have? If more than one, run the system one stick at a time to isolate the bad stick of memory. If you have only one stick of memory find another suitable Module to test with.

[condor]

Looks like bad RAM - but it might be wrong FSB settings or CAS settings:


make sure you're not using PC100 RAM with PC133 settings nor should you use CAS 2 for CAS 3 or CAS 2.5 Modules:

The best way to insure proper opreration is to allow the motherboard to configure RAM settings using SPD information (usually a setting in the Advanced Chipset setup in your BIOS)
or "Auto" settings.

if all the settings are correctthen you have bad RAM.

[Stryfe]

Well, my RAM settings are determined by SPD and I'm getting ready to put in a couple of known good sticks of RAM.

Tox_Boy, I only have one stick of PC133 CAS 3 256MB SDRAM. As I said above, I will be taking it out and testing one or two others tonight.

Thanks for all your help, guys.

--Stryfe

[Stryfe]

I've just started experiencing the same errors again, but this time, I have two sticks of known good RAM in. They are each PC100 with unknown CAS latency. One is a 64MB stick and one is a 128MB stick. I'm starting to think memory controller on the motherboard at this point. Someone please tell me that I'm wrong.

[edit] And before someone tells me to check the BIOS settings, everything is set properly. Timing is set to 100 MHz, CAS is set to 3T to be safe. [/edit]

--Stryfe

[tha 4NiK8R]

try this
you should also download diskeeper and defrag your harddisk and pagefile.

I would also try moving the known good stick to the slot that was not used by the original.

[ZAP!!]

I've had similar errors with my system before I installed the AMD/Win2k patch and the VIA/Win2k AGP update. Before that I would get errors all the time. Now I reboot once every few weeks outta habit.

[Stryfe]

What patch is that? I have to the AMD/Windows 2000 AGP patch, if that's it. Otherwise, could you maybe post a link?

--Stryfe

[ZAP!!]

I found a reference to the Via/Win2k update on Technet but it says it was fixed in SP2. Q261606

Maybe I oughta just lay off the crack rock.

[3D Prophet III]

What motherboard and processor are you using? If you have onboard cache try disabling that. If you don't have onboard cache it could be a sign of the cache on die going. Have you overclocked your processor at all? If the memory you're using is passing the Goldmem tests and other simm testers and aren't constantly causing your system to lock it could be the cache memory going.

Each time you reboot your system, can you see how much physical memory you have installed? Does it report the same each time your system reboots? Signs of SDRAM going would be by it reporting 50% of its physical size to the BIOS and it could still pass some simm testers.

[Stryfe]

Ok....First to answer 3D Prophet III's questions.

I am using the ASUS A7V (Uses the Via KT133 chipset and is PCB Rev. 1.02) in combination with an AMD T-Bird 700MHz CPU. I have over clocked my CPU a few times, but only to a max of 800MHz (Upped the multiplier by 1).

In regards to the cache, I haven't even thought of that, so I'll give it a go the next time I restart my computer. If the cache is dieing, what do I do? Just get another CPU?

My system does indeed report the amount of memory in the computer at bootup. And so far it's been the exact size it's supposed to be.

ZAP!!: Thanks for the link. And, yeah, laying off crack rocks would generally be considered a good thing.

Here's a new piece of info. I've just experienced yet another memory error. This time, I'm using 2 known good sticks (192MB total) *AND* I'm not using either of them in the bank where the old one was.

Regards,
--Stryfe

[Stryfe]

I'm about at wit's end. I've disabled the level 2 CPU cache (Can't do the level 1 cache b/c Windows 2000 won't boot if I do) and I still get errors and I even got 2 BSODs in a row.

The funny thing about it was that these happened right after each other, and both were STOP: 0x0000000A errors that occured in ntoskrnl.exe. The first one occured at the memory address 0xE2585140 and the second one occured at 0xE4E17340. They both had IRQLs of 0x00000002, occured during read operations (0x00000000), and the reference came from the memory address 0x8049AD63 in both cases.

Anybody got any ideas?

--Stryfe

[geoscomp]

Maybe check out this MS article..it could be not memory at all
http://support.microsoft.com/default...;en-us;Q314063

[MacGyver]

I got this error a couple of weeks ago with AutoCAD 2000 under Windows XP. Turns out the drawing file I was working on was corrupt. I confirmed this by asking a co-worker to open the file on their PC and the error occurred there too.

It took a while to clean up the drawing so the error didn't pop up anymore, but after that I didn't see the error again.