Have you ever felt that documentation was as rare as hen's teeth? If you've never seen a hen, or searched for its teeth, I can tell you that you won't find many teeth in your hens.
How often do you read things, like this article, and figure that if somebody with a modicum of credibility writes it, it must be true. And certainly when you read something from "official" source, you absorbed it without question since it comes from the oracle [sic].
I'm that way. Like you, I don't have time to track down every fact, take it to my test bed, and confirm what I've read is the truth. It would be a maddening procedure, with a questionable rate of return. However, sometimes you need to do just that.
The Fast Drive That Wasn't
We had a legacy machine that barely met the requirements for installing Windows 2000. It had a 4.3 GB hard disk in it that was slower than the Iomega Help Desk. The machine's BIOS supported Ultra DMA 66 so we went all-out and got a Maxtor 30 GB UDMA-66 hard disk to put it in this not yet ready to be retired box.
Windows 2000 found the disk without any problem and we wrote the signature to the disk as you must do in order for the OS to work with a new disks. We then upgraded it to a Dynamic Disk before creating any partitions. Then the volumes were formatted.
Talk About Pigs That Are Slow
The old hard disk was one of those Bigfoot monstrosities with a blazing access time of around 13-14ms. The new disk had an access time of 8.5ms so I was looking forward to rocking and rolling with the new speed demon. However, the performance increase wasn't nearly as impressive as I had hoped. Something must be wrong.
After doing a little web research, I discovered that your UDMA stuff isn't going to work right unless you use an UDMA cable. No problem, I went to Fry's and got one of those and plugged the disks into it. There was some performance improvement, but again, not what was expected.
Is it Using UDMA?
I wondered if UDMA was actually working. I checked out www.ntfaq.com and found an article regarding how to confirm whether UDMA is enabled on a particular disk. To find this out, you need to open the device manager and look at the properties of you IDE controller. You should see something like the following:
Sure looks like Ultra DMA Mode has been enabled, doesn't it? I didn't have to do anything to configure these settings. I assume that Windows 2000 detected the hardware and determined that it supported UDMA 66 or 33 or maybe even 100? In any case, some type of UDMA support seems to be in place.
It's Not Over Until Its Over
Thinking that my disk was performing at its max capacity, I continued to use the computer like I always do. But from time to time I had the nagging feeling that something just wasn't right. I spend a wad on this disk, and it just didn't seem as fast as the UDMA 66 drives I've worked with in my Windows 98SE boxes.
Then something arrives in my mailbox, courtesy of Windows 2000 magazine. In their Windows 2000 Professional newsletter a tip regarding enabling UDMA 66 support was included. Huh? I thought UDMA was installed and working. Maybe it was UDMA 33?
Hit The Registry!
In order to get that UDMA support for you Windows 2000 machine, perform the following steps:
- From the Run command, open Regedit.
- Open the Edit menu, and select New | DWORD Value
- EnableUDMA66 is the name of the new value.
- Set the data value to 1.
After I rebooted, voila! My disk was faster than fast. The registry entry did the trick.
For More Information
I'd like to tell you that there was somewhere you could go for more information on this issue. If you search the www.winntmag.com website for UDMA or UltraDMA66, you'll come up empty. The same is true if you hit the Microsoft web site. So, if any brave souls out there have information on this, belly up to the bar and send me a note. I'll post your response and grant you five social credits to boot!