Performance Problem with Software RAID

I have two IDE hard drives and I made them as a strip set (RAID 0) with windows 2000. However, when I tested with Sissoft Sandra 2001, I found the hard drive performance is the same as the single one. I don't know what could be the problem. Any one can help to tell where is the problem and how to solve it? Thanks!
Here is the detail configuration:
Hard Drive 1: WD 60GB 7200rpm 2M
Hard Drive 2: WD 60GB 5400rpm 2M
Both hard drive connected to mother board IDE 0 with one hard drive cable.

Here is the result of File System Benchmark in Sisoftware Sandra (ver 2003.7.9.73)
Here is the result:
My Disk: 31483KB/S
ATA100 7200rpm 8M 120G Drive is 31400KB/S
ATA100 5400rpm 2M 30G Drive is 20000KB/S
I am expecting the my result should be at least 2x the 5400rpm dirve.
Anything I can do to improve the performance?
Thanks.

This is because of two things:
1. (Most important) different drive types... in RAID always use the same model otherwise you will get added latency and a big performance decrease, especially if you mix drives with different rotation speeds. The slowest drive will bottleneck the entire RAID array.
2. They're on the same cable and only one device can be accessed at a time on a IDE cable.

I love my Delta 60HP 7000 RPM fan that puts out more dB then CFM

I've got the same problem and I don't think any of the replies in this forum answer to that.
I got two Maxtor drives (80 and 60GB, both 5400RPM, U-DMA/100) and also set up a software RAID 0 array.
When I tested them before with Winbench both drives started with a transfer rate (TR) ~30MB/s and ended up with about 20MB/s (graphic looked like the one in the article).
After setting up the RAID, TRs where still 30MB/s at the beginning and stayed constant until the end (even going up to about 32MB/s).
I now you can't expect to perform them as well as when you use two identical drives (like 2*30=60), but at least somewhat around the 50s, I'd say. I mean, it's already a good thing that the TR stayed pretty steady at 30MB/s instead of going down, but that's also totally different from what I'd expect from the graphic in the article.

Someone PLEASE explain why that is, I simply cannot believe it's because they are "different". There should be at least SOME performance gain. All my results were worse than before except the TR at the end... :-(((

Thanks
Markus

*double post*
<P ID="edit"><FONT SIZE=-1><EM>Edited by Flinx on 09/14/03 07:49 PM.</EM></FONT></P>