I hope you can shed some light on this. I have a friend (actually my wifes boss) who was having a problem with his computer (of his own doing). He called me to take a look at it. Long story short. We started a re-format and install of windows XP Pro. The format was taking a real long time (only 35% done after 45 min). I opened the box and found the following config.
1) Primary IDE had ATA-100 HD and a CD-ROM.
2) HD jumper was set to Master and CD jumper was set to slave.
3) The HD was connected to the Slave connector on the ribbon cable and the CD-ROM was connected to the Master connector.
I re-configured the HD to the primary IDE as master and connected to the master position on the ribbon cable and setup the CD-ROM on the secondary IDE bus. This improved performance dramatically. My question is -> Did moving the devices to seperate IDE buses fix this or was it the weird setup on MASTER / SLAVE settings / connector position or a combination of both.
Usually there's a jumper on the motherboard that enables "cable select" which will determine master and slave on a channel based on the cable connector in case the jumpers on the drives themselves aren't set. Usually though the jumpers on the drives are set, so cable select isn't needed. The fact that you had both enabled and conflicting is probably the source of the problem.
You DO get a performance benefit in copying from cd->harddrive or vice versa if the drives are on seperate channels, although for a format operation this would not come into play as no data is being copied between the disks.
So my answer is that the cable select and drive jumper conflict is what was the source of the performance issues.
Thanks for replying to the post. I did a little reading from some links off of this site and got a little more insight to the IDE interface. Your input confirms what I read and added a little more. My wifes boss is happy even though I think the cheap skate wanted me to change something for free! He's got the computer back and I hope I don't have to deal with him again though I know I will.