I found that yesterday, the link it provides goes nowhere.

EDIT: (Looks like the post I replied to was removed by its author...)

I found a version of data life guard tools that will enable or disable ata 66 https://www.driverguide.com/driver/detail.php?driverid=64596

ya i know i deleted it but i found it somewhere else.

Wow, great find, that looks like the right one, only named something different. I'm on it and will report back. Many thanks!

dlg66 -d0 -o would be the command. if your drive is C:\

Well, darn. Unfortunately, it didn't make any difference.
In Normal mode, in Device Manager, the Primary and Secondary IDE controllers still have the yellow asterisks, and neither the HDD or CDROM show up. Which means I can't set them to DMA. They are still in Compatibility mode.

Restarting into Safe Mode, it displays the familiar "MBR has been modified" warning. In Device Manager, the HDD shows as "Generic IDE Disk Tyoe 47" and the CDROM shows correctly as a Toshiba 6202B. So even in Safe mode the WD disk is not ID'ing correctly.

I'm stumped. Any ideas?

take the CD rom off the cable. You should put the CD rom on the other ide channel anyways and not have it as a slave on the master channel becasue it effects performance.

Try changing the ribbon cable. Use a 40 pin cable instead of a 80 pin. check you cables. check your jumpers.

if all that doesn't help try updating the bios.

this honestly doesn't sound like a problem with the hard driver becasue the 2ndary IDE controller doesn't have anything connected to it so it shouldn't be an astrix.

The CDROM is already on the 2nd IDE, so is not sharing the HDD cable.
Have changed IDE cables; triple-checked jumpers; BIOS is the last one made for this mobo. BIOS settings for bus mastering on or off, make no difference.

What's freakin' weird is, all I have to do is replace the WD with an identically-imaged Maxtor and everything works fine.
I have two of the WD's and they both cause the same thing.

Bizarre.

The plot thickens. Or, sickens...

Recall, I have two of these 313000 drives. One was barely used, the other I used for a year or so without issue in a different computer.

I took both drives, removed everything with FDISK. Ghosted the known-good image onto them both. Double-checked that ATA66 was disabled on them both. Put the more-used one in the system, and whadya know, everything came up fine. No errors. CDROM and HDD show in Dev Mgr.
Put in the other one, which I've had issues with, and no change. Still the same errors.

Made copies of the MBR's from the good and bad ones, and compared. They were not the same. So I copied the good drive's MBR to the bad drive. Now they compared fine. So the MBR isn't defective. Put the bad drive with good MBR in, and still no luck.

Made copies of the boot sectors of the two, and compared. They were identical.

I'm at wit's end. It works fine in DOS mode but won't talk to Windows. Everything I can check is identical to the good one. These drives were even made just two days apart. I have come to the conclusion that the bad drive must have bad firmware.

Is there such a thing as a firmware utility for these things?

Guess you're probably beyond needing it now, but you can find the original standalone WDATA66 utility and associated howto here

https://web.archive.org/web/19990508044800/ht … ss/wdata66.html

precaud wrote:

In Normal mode, in Device Manager, the Primary and Secondary IDE controllers still have the yellow asterisks, and neither the HDD or CDROM show up. Which means I can't set them to DMA. They are still in Compatibility mode.

That seems odd to me.. I haven't' ever seen a hard drive cause the actual IDE controller to have a yellow asterisk. Sure, if the hard drive is having errors, it will cause the disk section in Device Manager to show a asterisk, but not the controller. Are you 100% sure you have the right ATA driver in windows installed for the motherboard? Have you tried a PCI IDE controller?

Yes, I tried all available controllers, no difference with this one HDD. I've never seen this either. I was pretty optimistic that writing the MBR from the good one onto the bad one would solve this mystery. But no...

So I'll just back up data to this drive and the other one will be the daily driver. I wanted it to be the other way around...

Well I just put the supposedly "good" AC313000 in to button up the system, and it too is now running in Compatibility mode, with the yellow asterisks by the IDE controllers, and no CDROM showing in DM. This is ridiculous. Nothing changed since yesterday. I'm tired of screwing around with these things. They'll get used for data backup only. I'll live with the Maxtors, though they haven't been as reliable for me as WD"s.

I've just wasted four days trying to track down a phantom issue.

i only use SSDs now. Have several IDE drives that work fine but IDE is just slow and my standards have evolved.

Yeah, well then you get to worry about data retention, a whole 'nuther can of worms...

Speed is not a major concern for this system, its an engineering workstation. But stable data retention is a must.

By the way, many thanks for your help here.

Perserverence furthers. I figgered it out. I was looking for one problem, and I had two.
First was the ATA-66 issue, which WDATA66 fixed.
The other has to do with Ghost. I was using it in clone mode, and using the -svee switch, which tells it to create partitions of equal size as on the source drive. I had assumed that meant it would make a bye-by-byte exact duplicate of the entire source drive on the destination drive. But it turns out, it doesn't. Most important, It doesn't copy the MBR. And probably not the boot sector, either. That's why Windows was throwing up "MBR modified" message on start-up.

So after making the "clone", I have to do SYS C: from a Win98 floppy to write the system onto the cloned drive, and do it on the same computer that it will be installed in.

I remember seeing something similar to this years ago, when I partitioned and formatted a disk on a VIA-chipset machine, and when installing it into a 440BX machine, it threw up this same "MBR modified" message.

So many little gotchas. Now I can proceed with the XP install (dual boot).