Troubleshooting a Really Obscure Windows XP Problem

Lest I give the erroneous impression that Apple is the sole source of my computer woes, I spent all day (fairly literally) yesterday troubleshooting my main PC after what should have been a simple hardware upgrade went awry. I purchased a new graphics card and a new sound card. The graphics card went in without a hitch and dazzled me with both its performance and its nearly inaudible fan. After convincing myself that all was well so far, I put in the sound card.

I still don’t know exactly what I did wrong. It could have been that I connected the cables for the front-panel jacks and the CD drive incorrectly. It could have been that the new card collided in some manner with the motherboard’s native audio hardware or the serial-port card I use for microprocessor firmware development. (I’ve since removed both.) It could have been some weird fluke.

Whatever the cause, the machine didn’t run at all well after I put the card in, and taking it back out didn’t improve matters. The symptom was odd: the system took way too long to do anything that involved disk access. It would take a good ten minutes just to boot up and reach the Windows desktop. This made diagnosing the problem excrutiatingly slow, since changing any system settings meant waiting through lengthy shut-down and reboot processes.

My hunch was that something was badly awry with the hard drive’s data structure, but running CHKDSK didn’t seem to improve matters. After exhausting my not insignificant repertoire of PC troubleshooting tricks, I started looking around on Seagate’s website, still under the assumption that the hard drive was somehow confused. I tried their handy SeaTools diagnostic program; it gave both hard drives a clean bill of health.

Finally, though, in an article titled “How to troubleshoot poor or slow ATA hard drive performance” I found a clue. (I’d provide a link to the article but their database-driven Knowledge Base doesn’t seem to allow this.) One of the suggestions was to make sure that DMA mode was turned on. It turned out that DMA mode was not turned on on my PC. this led me to the article titled “My drive is using PIO mode instead of DMA in Windows XP?” which explains why DMA mode might not be turned on in Windows even though it’s turned on in BIOS, which was the case on my system.

I applied the fix described in the second article but it didn’t seem to improve matters, and further examination showed that the system was still stuck in non-DMA mode. Starting to fear that I had actually damaged the hardware somehow I decided that I needed a second opinion. I installed Vista on the second drive (which was empty, fortunately) and established that I could boot under that OS and it would happily run the drives in DMA mode. I even put the sound card back in and got it working under Vista, so I had established that the problem was not the hardware itself but something odd with the XP boot volume.

I rebooted with XP, thinking that I’d do something like trying Seagate’s fix again. Oddly enough the machine booted normally and the drives were again operating in DMA mode in XP. I don’t know why–theoretically I hadn’t done anything to the settings stored on the XP drive. Maybe the series of reboots and power cycles that I performed while installing and testing with Vista shook something loose, so to speak. In any case the problem was solved and the machine is running normally again.

As an aside, it’s common advice in computer-music circles to make sure that DMA is turned on to ensure optimum hard-drive performance. Given how slowly my system ran when DMA was turned off, it’s hard to imagine anyone even trying to run a music PC with DMA off.

Leave a Reply

Your email address will not be published.

*
To prove you're a person (not a spam script), type the security word shown in the picture. Click on the picture to hear an audio file of the word.
Anti-spam image