Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!mnetor!uunet!husc6!rutgers!cbmvax!daveh
From: daveh@cbmvax.UUCP (Dave Haynie)
Newsgroups: comp.sys.amiga
Subject: Re: 1.6 meg drives and the Amiga
Message-ID: <2707@cbmvax.UUCP>
Date: Mon, 9-Nov-87 16:58:29 EST
Article-I.D.: cbmvax.2707
Posted: Mon Nov  9 16:58:29 1987
Date-Received: Wed, 11-Nov-87 21:29:13 EST
References: <942@sugar.UUCP>
Organization: Commodore Technology, West Chester, PA
Lines: 37

in article <942@sugar.UUCP>, peter@sugar.UUCP (Peter da Silva) says:
> In article <2615@cbmvax.UUCP>, daveh@cbmvax.UUCP (Dave Haynie) writes:
>> > I want the 1.6 meg high-density microfloppies. ... [How about it?]
>> Well, maybe.  The 2 meg (unformatted) floppies weren't designed into 
>> Paula, so it's not as simple as dropping in a new device driver....
> Now, I might be sucking swamp water on this, but...
> I have heard that the encoding used on the Amiga 1000 diskettes can be
> modified to pack nearly 1.3 megabytes on the floopies. It's somewhat
> akin to the idea of using 16 bit nybbles in 18 bit GCR on the Apple-II
> instead of 6 bit nybbles in 8 bit GCR. Could you comment on this?

There was lots of talk on this awhile ago.  The MFM format used on the
current Amiga floppies was used basically because it's very easy to encode
and decode using the blitter.  As I recall (it's been awhile), 3 passes
encode it, 1 pass decodes it.

The end effect of the encode/decode is that for every bit encoded, two bits
get written to the disk (actually, flux changes, but you know what I mean).
There are certain rules that have to be followed, like no two "1"s in a row,
and no more than three "0"s in a row, or something like that.  Now, if you
sat down and enumerated all of the legal combinations in any given word written
to the disk, you'd certainly find more than 256 combinations are permissible.
The 1:2 encoding give you a valid set of encodings, but not the best set
(density-wise).  However, if you wanted to write a more compact encoding
scheme, you'd probably have to do it via table lookup, rather than blitter,
so you'd incur more of a speed penalty.  Whether or not this penalty would
be significant in comparison to DOS overhead remains to be seen, as does the
actual increase in density you could get with a practical system.  But it is
possible, at least in theory.

> -- 
> -- Peter da Silva  `-_-'  ...!hoptoad!academ!uhnix1!sugar!peter
> -- Disclaimer: These U aren't mere opinions... these are *values*.
-- 
Dave Haynie     Commodore-Amiga    Usenet: {ihnp4|caip|rutgers}!cbmvax!daveh
   "The B2000 Guy"              PLINK : D-DAVE H             BIX   : hazy
    "Computers are what happen when you give up sleeping" - Iggy the Cat