HOW TO: Disable IDE Error Checking

  • Thread starter Thread starter Dan
  • Start date Start date
D

Dan

I need to find out how to disable the error checking that
XP does on IDE hard drives. My system runs real fast
until I start working with large files or alot of small
files. Any kind of extensive movement of files whether
reading from or writing to, XP lowers the Transfer Rate
from UDMA5 to PIO mode which kills the performance of the
system. When this happens it makes it impossible to work
at all without rebooting. Any suggestions? I know my hard
drives are perfectly fine, as it does the same thing with
all of my drives including a new one. I am running XP Pro
with an Athlon XP 2400+ processor and 1GB of Memory. My
hard drives (5 of them, 4 are on the primary/secondary
ports on the mainboard, and the 5th is on a promise pci
controller card) are all Western Digital and vary in size
from 100GB to 200GB. I have a clean install of XP with
SP1 and all available patches and updates. If there is
any other information that I'm missing which would help,
please let me know. Thanks.

Dan

Note: The error checking i am referring to is when XP
detects 6 or more CRC errors or something like that and
downsteps the transfer rate of the ide port a given drive
is connected to until the errors go away. In my case, PIO
is as low as it goes, but is too slow.
 
I need to find out how to disable the error checking that
XP does on IDE hard drives. My system runs real fast
until I start working with large files or alot of small
files. Any kind of extensive movement of files whether
reading from or writing to, XP lowers the Transfer Rate
from UDMA5 to PIO mode which kills the performance of the
system. When this happens it makes it impossible to work
at all without rebooting. Any suggestions? I know my hard
drives are perfectly fine, as it does the same thing with
all of my drives including a new one. I am running XP Pro
with an Athlon XP 2400+ processor and 1GB of Memory. My
hard drives (5 of them, 4 are on the primary/secondary
ports on the mainboard, and the 5th is on a promise pci
controller card) are all Western Digital and vary in size
from 100GB to 200GB. I have a clean install of XP with
SP1 and all available patches and updates. If there is
any other information that I'm missing which would help,
please let me know. Thanks.

Dan

Note: The error checking i am referring to is when XP
detects 6 or more CRC errors or something like that and
downsteps the transfer rate of the ide port a given drive
is connected to until the errors go away. In my case, PIO
is as low as it goes, but is too slow.
Click to expand...

Where do yo see these errors? Event Viewer ?
 
Dan said:
I need to find out how to disable the error checking that
XP does on IDE hard drives. My system runs real fast
until I start working with large files or alot of small
files. Any kind of extensive movement of files whether
reading from or writing to, XP lowers the Transfer Rate
from UDMA5 to PIO mode which kills the performance of the
system. When this happens it makes it impossible to work
at all without rebooting. Any suggestions? I know my hard
drives are perfectly fine, as it does the same thing with
all of my drives including a new one. I am running XP Pro
with an Athlon XP 2400+ processor and 1GB of Memory. My
hard drives (5 of them, 4 are on the primary/secondary
ports on the mainboard, and the 5th is on a promise pci
controller card) are all Western Digital and vary in size
from 100GB to 200GB. I have a clean install of XP with
SP1 and all available patches and updates. If there is
any other information that I'm missing which would help,
please let me know. Thanks.

Dan

Note: The error checking i am referring to is when XP
detects 6 or more CRC errors or something like that and
downsteps the transfer rate of the ide port a given drive
is connected to until the errors go away. In my case, PIO
is as low as it goes, but is too slow.
Click to expand...

Instead of turning off the error handling stuff, I suggest that
you find out why you are getting so many errors, since CRC errors
should be *very* rare on a well-built PC. Some thoughts:

1. Firmly reseat all HD cables, both IDE and power. Free.

2. Find out which HDs are causing the problem, and replace the
cables to those HDs. Make sure the new IDE cables conform
to IDE standards, not >18", and not junk stuff. Cheap.

3. Replace the power supply with a better one. Not cheap, but
if you have 5 HDs, presumably in a single tower case, I am
suspicious that your PS does not have enough +12V juice to
handle concurrent seeks on 5 HDs.
 
Thanks for the reply. I already tried re-seating the IDE
cables and power connectors. It is a new power supply and
is 400 watts. That should be more than sufficient for 5
drives. Any and all of the drives experience the problem
when extensive data transfers occur. As far as the +12
goes, mine averages around 13 and 13.2 i believe
according to the Asus PROBE Utility. The BIOS recognizes
it in the mid-upper 12 range however, not 13 usually. My
old Athlon 1800+ system did not have this problem. Could
the speed of the processor be causing some kind of sync
issues somehow? Don't know how else to phrase that. The
event log shows ide errors often but i know the drives
themselves are not to blame. The IDE cables are the ones
that came with the motherboard and I tried switching them
out but to no avail. I've wondered if the processor is
causing the issue somehow or if the ide ports on the
mainboard is failing for some reason. Thanks again.

Dan
 
Correct, it is the System event log that shows the
errors. I figured after a clean load of the operating
system with all the latest updates, would resolve it if
it was a software based issue. It must be something
hardware wise that is causing the problem and I know that
isn't the hard drives. That leaves the motherboard (1
year old maybe, perhaps a few months older than that),
processor (about 1-2 months old now) or power supply (400
watt, only had 6-9 months or so). This all started
happening in January for the first time. It's so
frustrating. Thanks for the reply.

Dan
 
A bios update might silently update issues that the hard drive controller
may have (if you are using the one built into the motherboard).

Another culprit could be your cables.
 
Correct, it is the System event log that shows the
errors. I figured after a clean load of the operating
system with all the latest updates, would resolve it if
it was a software based issue. It must be something
hardware wise that is causing the problem and I know that
isn't the hard drives. That leaves the motherboard (1
year old maybe, perhaps a few months older than that),
processor (about 1-2 months old now) or power supply (400
watt, only had 6-9 months or so). This all started
happening in January for the first time. It's so
frustrating. Thanks for the reply.

Dan
Click to expand...

Error reporting is your friend. It's trying to tell you something.

It doesn't matter if the label is new, or whet the label says. The
first thing I'd do is pop in a generous name-brand power supply.

Try unplugging the power on everything but the minimal system, put a
digital volt meter on the 5v and 12v lines, and see what the
voltages are. Add disks one at a time and see how bad it gets.
I don't know what the voltage margines are, but post them. SOmeone
here does,

 
Back
Top