Path: utzoo!attcan!utgpu!jarvis.csri.toronto.edu!mailrus!uwm.edu!gem.mps.ohio-state.edu!ginosko!uunet!crdgw1!crdos1!davidsen
From: davidsen@crdos1.crd.ge.COM (Wm E Davidsen Jr)
Newsgroups: comp.sys.ibm.pc
Subject: Re: Don't buy from HDI
Keywords: yet another horror story
Message-ID: <1456@crdos1.crd.ge.COM>
Date: 25 Oct 89 13:57:11 GMT
References: <754@dekalb.UUCP> <11040@phoenix.Princeton.EDU> <902@gumby.cc.wmich.edu>
Reply-To: davidsen@crdos1.UUCP (bill davidsen)
Organization: GE Corp R&D Center
Lines: 25

In article <902@gumby.cc.wmich.edu>, bowden@gumby.cc.wmich.edu (Michael L. Bowden) writes:

|  I too bought a 296N from HDI, having been told that the 4096 had an
|  unacceptable return rate (which drove the cost up to $569 at HDI).  If
|  I had purchased the 4096 I wouldn't be getting more than about 230KB/sec.
|  With the 296N I'm getting about 450KB/sec, even with 2:1 interleave,
|  not to mention half the power draw and half of a drive bay free.

  I can't comment on most of what you said, but you comment about the
transfer rate of the 4096 is not correct. With a track buffered 1:1
controller the theoretical max for MFM is 510kb/s, with RLL 780kb/s. I
have seen a tad over 600kb/s using a wd1006VSR2 (RLL) controller, to a
program which was actually looking at the data rather than doing disk
tests. I therefore conclude that any 1:1 controller with real track
buffering (hardware not software) should be able to beat the value you
mentioned.

  If anyone cares, the theoretical max sustained transfer rate for more
than one cylinder (on a 4096) is 436 (MFM) and 668 (RLL). These are only
slightly higher than the coretest shows with appropriate controllers.
-- 
bill davidsen	(davidsen@crdos1.crd.GE.COM -or- uunet!crdgw1!crdos1!davidsen)
"The world is filled with fools. They blindly follow their so-called
'reason' in the face of the church and common sense. Any fool can see
that the world is flat!" - anon