LN>Same here but with a big caveat. You have a bus master controller and
LN>so do
LN>I. The results always seem marginal on IDE/PIO systems that I have
LN>tested.
LN>I have not tested PIO SCSI so I don't really know about them. It
LN>could be
LN>that the skeptics are all using PIO or a much smaller cache.

RE>I was very skeptical when I read your post.  Not that I don't respect
RE>your opinions and knowledge but because I ran extensive tests on IDE
RE>and SCSI busmasters about 2 years ago.  The difference then between
RE>HPFS16 and HPFS386 was significant.

RE>However, after I read your post I did run a few tests using WARP on an
RE>IDE machine and have to agree with you.  There is no discernible
RE>difference between the two.

RE>Can you think of some obvious reasons why this would be the case?


I'm sure that Peter has been patient with us because he has been busy or he
is interested himself. Now to tie this last question up. Two years ago
would have been before Doug Azzarito and associates gave HPFS16 a through
going over correcting a few things while tracking down the elusive "CPS
extent not found" bug. One of the CSD's or private fixes speeded up HPFS16
a good bit. So 2.11 or Warp would not show as great an improvement.

--Lynn

 * SLMR 2.1a * To boldly code what no one has coded before....

--- DB 1.39/004485