Interesting results. A few questions for clarification:
- Were the SEEKs random?
- What were the SQL statements you used?
- Sizes of hardware buffers on HDs? Although, the link I posted would have you believe they don't make *that* big a difference to overall HD performance
- Were you able to eliminate the effects of OS disk caching in your tests?
On a minor note, I assume you meant "NTFS" in those places where you wrote "HPFS" as the HD file system.
Regards. Al
"Violence is the last refuge of the incompetent." -- Isaac Asimov
"Never let your sense of morals prevent you from doing what is right." -- Isaac Asimov
Neither a despot, nor a doormat, be
Every app wants to be a database app when it grows up