Path: utzoo!attcan!uunet!lll-winken!ncis.llnl.gov!helios.ee.lbl.gov!pasteur!ucbvax!decwrl!labrea!rutgers!bellcore!texbell!uhnix1!sugar!ssd
From: ssd@sugar.uu.net (Scott Denham)
Newsgroups: comp.sys.amiga
Subject: Re: External 3.5 info wanted
Summary: External drive fix for the 2000
Message-ID: <3320@sugar.uu.net>
Date: 23 Jan 89 19:46:24 GMT
References: <620004@hpnmdla.HP.COM> <5160020@hplsla.HP.COM>
Organization: Sugar Land Unix - Houston, TX
Lines: 41

In article <5160020@hplsla.HP.COM>, tomb@hplsla.HP.COM (Tom Bruhns) writes:
> >In article <620004@hpnmdla.HP.COM> glenb@hpnmdla.HP.COM (Glen Baker) writes:
  
> >On a related question: Is it possible to take any old half height 5.25 inch
> >360K disk, put it in the A2000, and have it work with programs such as
> >DOS-2-DOS without this ID signal?  The external 5.25 disk for the A1000 is
> >too expensive for me (and getting difficult to find) and I don't want to have
> >to buy the BridgeBoard just to read and write MS-DOS 5.25 disks.
> Can't you just do the same hack for the 2000 to use the drive externally?
> That is, won't an external 1000 drive work on the 2000?  (I'm interested
> in the answer to this, since I am considering getting a 2000 to replace
> my 1000...)
 
 As you correctly surmised, it is possible to use the same hardware hack 
to hang an extra 5 1/4 or 3.5" disk on a 2000.  I built such a gadget 
when I had a 1000 and had no problem with the hardware when I traded up
to a 2000.   
 
  HOWEVER.........
 
Note that above I said no "hardware" problems.  There is a software problem
that may get you in the above scenario.  There is a BUG in Kickstart 1.2/1.3.3
that causes the trackdisk device to reject perfectly valid raw I/O packets
if you have a drive at df0: and one at df2: (the first external on a 2000)
but no drive at df1: (the second 2000 internal drive).  This renders the 
external 5 1/4 drive useless for transferring MS-DOS files with the 
PCUtilites packages that came with 1.2.  I don't know if DOS-2-DOS will 
have the same problem - if they go through trackdisk, the almost certainly
will.  If they go straight to the hardware there's no problem.  I was told
that the error is that trackdisk is checking the wrong location on a raw
I/O request, and if there's not a DF0, that location will probably contain
values such that the I/O request will fail.  I confirmed by borrowing a 
second internal drive that the problem does not occur on a 2-drive 2000,
though.  SO... if you want to be able to do this sort of thing before 1.4
comes out (which is supposed to have this fixed), keep your 1000 or get
a 2-drive 2000!
 
> >-- 
  
  Scott Denham 
   Houston TX