Hello! I am having trouble with my computer, I think it might be my memory. Sorry if this is not the right section.

First, I have 2x 256MB of DDR 2700 RAM. I was getting all kinds of BSOD's, mostly suggesting CPU or RAM error. (BAD_POOL... IRQL_.....)

Then, I decided to buy some RAM, 2x 512MB DDR 3200. At first when I plugged it in, it was fine. I played a game of NBA Live and no troubles there. But, when I played Half Life 2, I alt-tabbed out of the game, and alt-tabbed back in. It took about 30 seconds to get back in the game. I think somewhere along the line something went wrong. And after 10 seconds of playing time, the computer suddently crashed making the looping scratching noise.

Does anyone know if alt-tabbing in and out of the game might have corrupted the memory chips? Thanks!

run goldmem it will detect if there is anything wrong with your memory. if it doesnt find anything wrong then its not your memory

Sorry, I forgot to add, I ran memtest86 and it gave me so many erros on tests 5 and 7. I just feel that I might have corrupted my memory by alt-tabbing out of HL2 and back in again. But then if memory was that easily corrupted...

There's no way you corrupted your memory by alt-tabbing out of a game, in fact no properly written software can really corrupt the physical memory. What probably happened is the memory was never good from the start and only when running memory intensive applications does the problem show up.

Run memtest86 on the new memory, if no problems show up the memory is fine. If you're still getting errors you should take the side off of your case and put a fan blowing on the system and try running the memory test again. Perhaps it's not bad memory but an overheating problem since it seems to only happen when the system is under load.

Thanks, I'll try that. If it doesn't work, I will probably send the RAM back since it has lifetime warranty. But then again, I bought it them off of eBay

I encounter same problem with my RAM when Alt+Tabbing Half-Life 2. However, after doing intensive RAM benchmarking, i found that is nothing wrong - it is the culprit -> Source engine (same problem with CS: Source, HL2: Deathmatch).

I just tested the memory on my friend's computer. The chips showed up as 504MB instead of 512MB, and only one of them failed tests 5 and 7. However, both of them showed up with erros on my system. I think my whole system is just too corrupt.

the reason that the chips prolly showed up less on your friends PC is because he has onboard video and the video is being used by the ram. But ya if you get errors again then i would think that its the ram. Once again like i said before try goldmem. Just so you can get a second option

Update: I think it might be my system. Everytime I do tests in Memtest86 on my old 256MB sticks, different errors occur on different tests at different sizes of the memory. Sometimes, there might be many errors on test 3, and other times on test 6. If the RAM is corrupt, wouldn't it have the same errors at the same place? Thanks!

Edit: I just did another test on the 256MB stick, and there wasn't any errors . Everything is just so random.

If you are really getting such random results from several sets of memory sticks I would hesitate to blame the memory. Almost ALL memory, top of the line to cheapo, is capable of giving average + performance. Even if the memory is not matched, in general, the worst result will be all memory performing at the least able memory sticks performance level.

It does not sound as though you are running your computer on the "bleeding" edge so perfect timings and such should not be an issue. The ambient temperature may be having an affect on the performance (reliability) of your memory and other parts of your system. Heat related troubles can be, and often are, random in their appearance.

If this is not a problem you might consider physical damage when changing the memory sticks. This is not so likely. As most physical damage would not show random errors. And most likely would be fatal at or near 100%.

You might consider that your OS may have become corrupt. Again this is more likely to be a problem on a heavily tweaked and/or hard worked computer. But Windows XP can become corrupted. Especially if you like to try out a lot of new applications and then uninstall those that you don't care for. The Windows Registry can really become a junkyard after a while. Problems in the registry can seem quite random, and can affect almost any component on your computer.

If that might be the case you could try some conservative usage of a utility like CCleaner or Tuneup 2004. If that helped you could then choose to risk more aggressive 'cleaning'. Or maybe a format/reinstall would be the solution.

In my 30+ years of experience with computers I've found hardware troubles to be random in their occurance as relates to temperature, humidity, vibration and even usage. But not as to what the failure is (error code etc.). Software problems can present in a baffling variety though.

Unless you have a large supply of spare parts or spare money, it might just pay for you to partition your drive (a secondary OS partition of 2-4 gigs) and install a second copy of XP just for testing. You can then use that partition for both testing of your hardware and the riskier games we play with offbeat software etc.

Thanks for your help, I think you're right, it may be the ambient temperature since it's getting hot here in NY.

I don't think it's my OS because I ran the test in Memtest86, which boots from a floppy disk before Windows. My system would just freeze, nothing moves respondes, no BSOD's either. Othertimes, a BSOD would pop up, mostly suggesting errors regarding memory, caches system RAM etc... These erros include IRQL_NOT_LESS_OR_EQUAL, BAD_POOL_CALLER, and some other random BSOD's. And when I do restart from the crashes, the system will not POST 85% of the time. I have to usually cut off the power, take out the RAM sticks and put them back in and start the computer again for it to POST. And ocasionally, upon the POST, it will say CMOS Checksum error...

My computer has become almost useless. After resource hungry applications are closed, the computer will likely freeze with no BSOD's, just hard freezing after about 5 minutes. This problem has gotten worse and worse.

This happened last year as well, mostly hard freezes. I switched a power supply, and it has gotten better, but the computer will still freeze. But now the problem the problem is reaching epidemic proportions.

I think it might also be connectivity problems. Sometimes when I replace the RAM after a crash, the computer will make a long beep code, indicating that RAM isn't detected. When I shut off the machine, and just move the RAM stick in just a bit tighter, although thinking it would make no difference, it worked.

BTW, My CPU is running at around 45 degrees C, when I checked in BIOS a few minutes after the crashes.

Thank in advance for your time on reading my post.

45 degrees C should not really be a problem. But if there is any physical damage to a component temperature CHANGE can be a problem.

You know, you might try blowing out your case & mboard/cards with some canned air. Dust can do some weird things too. Especially if humidity and solder 'creep' or growth (tiny crystals growing in joins) are also present.

You might try swapping you video card for another just to see if that is part of the problem. Video cards can also come up with some seemingly random crashes.

I wish you luck. Computers are about the most reliable electronics you can buy. But when the get 'sick' they can leave you about to pull your hair out in exasperation.

I have exactly the same problem. Have you or anyone found a fix? Thanks.

Incompatible RAM? Try flashing the bios (however this is risky).