Figured it out: IDE controller had reverted to PIO mode

Figured it out: IDE controller had reverted to PIO mode

Post by Marc Sabat » Mon, 11 Sep 2006 07:14:00

ell, lots of lots time over this, but it's all figured out now.
Further testing had shown the disk worked fine when accessed via USB
instead of IDE (I have a USB enclosure for the disk). So I had
concluded it must be an IDE problem. Took some convincing to get HP
support to buy into this. Before they'd let me send in my computer for
repair, they insistent on sending me a new disk, which I thought would
exhibit the same problem. Much to my surprise, the new disk worked fine
even via IDE. So it looked like some weird issue where my old drive
could talk fast via USB but not IDE (weird because the USB enclosure
uses the IDE connector internally). I resigned myself to reinstalling
everything on the new disk.

After getting the OS and drivers up and running from the original disks,
I went to the HP site to update drivers - I know I had needed a new card
reader driver to handle 2GB cards. Well, one of the available downloads
was a fix for a problem where an IDE controller, after a repeated access
failure (like the infinite loop my computer got into after the initial
crash), would revert to PIO mode instead of DMA. Bingo. I spend the
better part of a week debugging this, and the fix turned out to be
simply uninstall the IDE driver and reboot (an update is offered to
prevent the problem in the future). I'm a little disappointed no one at
HP support knew of this issue, since it is one of only a handful of
available driver updates for this model, but that's far outweighed by my
relief at having my computer back.

For reference, here is my original problem description:

I have a Compawq Presario V2000 notebook with a Hitachi hard drive.
Last week the system crashed with a blue screen that I think was
implying some sort of device driver might have caused a problem. My
first attempt to reboot failed - the system went into an infinite loop
displaying some other message I couldn't quite read before it erased and
tried booting itself again, over and over. So I tuend it off and left
it off the rest of the day.

The next day I booted and it came up OK - but it took around 15 minutes,
whereas the system had always previously booted in a minute or so. The
system works fine, and is actually quite usable in most respects, but
any disk-intensive activity (program startup, file copy, etc) is MUCH
slower than usual. I ran the disk diagnotics from Windows, and while it
was ridiculously slow (took all day to do the full scan), it reported no
problems. I tried calling HP support, and they walked me through a
couple of things to try (holding power button down for two minutes with
battery & AC disconnected, doing a Windows system restore) that had no
effect. Next they wanted me to reformat my disk and reinstall Windows,
but I'm not ready to go there - this seems much more likely to be a
hardware problem than a software one to me, and I don't really feel like
going through that pain just to verify that.

So I've done some poking around using utilities I've found online. The
PassMark performance test shows my system functioning more or less
comparably to other systems on all CPU, memory, and graphics benchmarks
(well, 3D graphics performance is not so good), but most significantly,
my system comes up about 20-50 times slower than other systems in disk
read and write. It is showing a throughput of only around 1 MB per
second. The SpeedFan utility says my hard disk is performin

Figured it out: IDE controller had reverted to PIO mode

Post by Gonz » Mon, 11 Sep 2006 10:38:43

P put a 40wire ide cable in mine instead of an 80wire one going to my DVDRW
drive. That knocked up the access mode one notch but this HP dvd640 drive
is still the slowest drive I ever owned.

They need to fix the firmeware in it or something.

"Marc Sabatella" < XXXX@XXXXX.COM > wrote in message
news: XXXX@XXXXX.COM ...