Path: utzoo!utgpu!water!watmath!clyde!bellcore!decvax!decwrl!ucbvax!ucsd!ucsdhub!esosun!seismo!uunet!wash08!txr98
From: txr98@wash08.UUCP (Timothy Reed)
Newsgroups: comp.unix.wizards
Subject: NCR 800 flame (was: Re: Cheaper winnies on an NCR tower)
Keywords: NCR TOWER 32/600 drive
Message-ID: <154@wash08.UUCP>
Date: 24 Jul 88 01:27:30 GMT
References: <531@prlhp1.prl.philips.co.uk> <105@cs-col.Columbia.NCR.COM>
Reply-To: txr98@wash08.UUCP (Timothy Reed)
Organization: American Chemical Society, Washington, DC
Lines: 23

(flame on)
That's interesting that NCR goes to such lengths to test those HD
cart tape drives, then how could they release the 800 with a virtually
incompatible drive, when compatibility with [246]00's is such a 
selling point.  I'll tell you, we weren't told about this by NCR - not
'til the thing was installed - I actually first found out about it 
through the net!  The 'dd' solution means we have to dd a tape out to
disk and read it on again, or take the 800 into account with every tape
we write - both methods slow and cumbersome, with the cause totally 
unecessary - the 800 should not have been released until the cart drive
was made compatible with the others n the NCR family.
(flame dying down)
How come 'cpio -B' doesn't swap bytes correctly?  And why does the
byte swap have to take place during the write - 'dd if=/dev/rstp/00yy
conv=swab' doesn't work - this kind of kludge would be much more 
palatable.
(flame off completely)
Glad to see the reset button moved out from 'under the hood' - you
can't get to the switch when you got a bunch of towers next to each
other.

# Tim Reed
# Usual disclaimers - I'm commenting entirely on my own behalf.