>
> Version  1.01       ------Sequential Output------ --Sequential Input- --Random-
>                     -Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- --Seeks--
> Machine        Size K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP  /sec %CP
> hermes         300M 10718  24 10369   6  5119   3 24210  76 26282  11  83.2   0



it's quite slow writes anyway...



>                     ------Sequential Create------ --------Random Create--------
>                     -Create-- --Read--- -Delete-- -Create-- --Read--- -Delete--
>               files  /sec %CP  /sec %CP  /sec %CP  /sec %CP  /sec %CP  /sec %CP
>                  16   367  38 +++++ +++  1937  19   514  53   953  99  1093  41
> hermes,300M,10718,24,10369,6,5119,3,24210,76,26282,11,83.2,0,16,367,38,+++++,+++,1937,19,514,53,953,99,1093,41
>
> I thought that the seeks should be better than that, but my (completely
> different disk/controller/etc.) Gateway2000 machine gets about 70
> seeks/sec, so maybe the seeks aren't so bad.  It's certainly 10 times what
> I was getting.  (Note that the files are 300MB, so that may affect
> seeks...)
>
> Subjectively, the drive is *much* nicer than the old one.  X starts up in
> a couple of seconds, isntead of taking a minute, for example.
>
>
> I don't know if I'll keep the Maxtor controller or not.  It lets my drive
> do UDMA 5, I think (the VIA motherboard controller only supports UDMA 2),
> but I'm not sure how much real-world differnce that makes.  It feels a
> little faster on the Maxtor controller, but not vastly better.
>
>
>   ``I probably don't know what I'm talking about.'' --rauch@math.rice.edu
>

----------------------------------------------------------------------
Instead of asking why a piece of software is using "1970s technology,"
start asking why software is ignoring 30 years of accumulated wisdom.