[comp.unix.xenix] PS180 controller and DOS

davidsen@steinmetz.ge.com (William E. Davidsen Jr) (07/28/88)

  I have been looking for an RLL controller, since I am out of room and
money for hard disks. I have a Xenix/386 system which is about out of
storage, so I ordered one of the Perstor PS180 (31 sectors) controllers.
These give an 80% increase in capacity.

  When I got it I found that it is not a drop-in replacement for the AT
controller. It does not have a floppy controller included (there is an
XT version, also). I got a WD1002-FOX AT style floppy controller, and
started installation.

  The documentation is purely cookbook, no info on how it works. It
reminds me of UNIX documentation, in that there is not a wasted word.
The gist of it is that you set your CMOS for "no disks" (type zero) and
the controller gets the types from jumpers on the controller.

  I was able to do the low level format and surface check, and installed
DOS3.3. I didn't try any disk products like Speedstor, since DOS3.3 will
partition a disk (up to about 120MB) and I wouldnot need files > 32MB. I
used a 2:1 format with a 16MHz 386, and the performance was very good. I
beat it up and had no problems with reliability.

  When I tried to install Xenix, I found that the boot floppy didn't
want to hear about the controller, and hung silently. After several
phone calls I found that the Xenix version has a floppy controller, will
cost even more, and will be available Real Soon Now. The current version
also does not include the track buffering.

  I haven't decided if I will wait for the new version or go with
something which gives less capacity but is available, such as the
Adaptec 2372.

  The dealer has been very helpful and has offered a full refund, and
was very sorry that the leterature for the new version got out before
the controller. The people at Perstor were helpful and competent. I have
good feelings about the company, there was just a mixup in
commumications which left me with a unit I can't use.

  I would recommend this controller for DOS applications, it seems very
fast and was solid when I beat on it. DOS had no trouble using the
disks, and reducing the number of seeks cut back the measured access
time by quite a bit.
-- 
	bill davidsen		(wedu@ge-crd.arpa)
  {uunet | philabs | seismo}!steinmetz!crdos1!davidsen
"Stupidity, like virtue, is its own reward" -me