PE> I personally had an application (rescan of my messagebase), which
PE> an HPFS386 4 meg cache made take 2 seconds instead of 25 seconds
PE> for a rescan, because 4 meg allowed all the indexes to be stored
PE> in RAM.  It is very silly IMO to say

PF> And,  in reality,  how often are you required to do this rescan twice in a 
PF> row?

I would do it quite a bit, basically after each call I noticed,
to read new mail sent to my BBS.

PF> This is the type of silly bench test I was talking about...

Not only that, but the exact same indexes being in memory helped
with the tossing of QWK replies.  One fellow posted about 30
messages/day, and called long distance to do it.  With HPFS386,
it would take about 2 seconds to toss his replies.  With HPFS,
it took about 20 seconds, or something similar, a dramatic
difference whatever it was.  Not a silly bench test by any
stretch of the imagination.  BFN.  Paul.
@EOT:

---