Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!zaphod.mps.ohio-state.edu!samsung!noose.ecn.purdue.edu!en.ecn.purdue.edu!doctorj
From: doctorj@en.ecn.purdue.edu (Jeffrey W Davis)
Newsgroups: comp.sys.amiga.misc
Subject: Re: Workbench 2.0 for Amiga 1000
Message-ID: <1991May2.204253.12914@en.ecn.purdue.edu>
Date: 2 May 91 20:42:53 GMT
References: <834.281c071c@zodiac.rutgers.edu> <2662@public.BTR.COM>
Organization: Purdue University Engineering Computer Network
Lines: 59

In article <2662@public.BTR.COM> valentin@public.BTR.COM (Valentin Pepelea) writes:
>No magic is required to run 2.0 on an A1000. I do that quite often using
>ZKick and a kickstart image.

ZKick requires 512K of memory to be located at $200000.  Not all A1000's have
memory at this location.  Actually, ZKick requires the 512K at $200000 and
temporarily an additional 512K for loading the ROM.

This ROM image is a developers ONLY version and is addressed in a different
space than the actual 2.0 ROM image will reside ($F80000-FFFFFF).

>It might actually be a good idea for CBM to not only sell 2.0 ROMs, but
>also a disk with a 2.0 Kickstart image on it. You see, replacing the ROMs
>might cost an end user as much as $100, while 512K of ram, the amount stolen
>by the Kickstart image, costs only $25. Forthermore, that would allow
>customers to boot under 1.3 or 2.0, as they please.

I believe that there will be 3 forms of 2.0 ROM updates :

  1.  A 2.0 ROM (simple enough)

  2.  A Kickstart disk similar to the one shipped with A3000's without the
      A3000 specific code.  This could be loaded by any machine with an
      MMU.

  3.  An A1000 Kickstart disk that will split 2.0 into 2 pieces; 256K in
      the WCS and 256K elsewhere (possibly).

>In fact, it is irrelevant whether CBM is going to distribute 2.0 on disk or
>not. It is trivial to read a Kickstart image from ROMs and put it on disk.
>And ZKick is freely distributable. Either CBM shrink-wraps 2.0 disks along with
>with some nicely printed manuals, or people will pirate in droves.

As I mentioned above, a ROM image copy would do you no good.  Translating
the ROM to a different address range is no trivial task either.  Contrary
to popular belief, ZKick does not do any translation.  A special ROM file
located at a specific location was provided to developers; ZKick just loads
it and sets it up.

It is likely that 2.0 will be released on disk as well as in ROM due to the
existance of 030 and 020/MMU equipped machines.  They will be able to load
this ROM image with some MMU magic.  I expect that this ROM image will be
a verbatim copy of the 2.0 ROM addressed at $F80000-FFFFFF.

A reasonable solution to the 2.0 dilemma is to simply add 256K to the WCS
memory.  With 512K of WCS memory the 2.0 ROM could be loaded into ANY A1000
so equipped.  As soon as C= officially announces what forms the 2.0 Upgrade
will appear in, this will become a reality.  If 2.0 is released on disk,
someone or myself will introduce such a modification if it is necessary.

>Valentin

***********************************************************
* Jeff Davis                * Relax! And get into    ///  *
* doctorj@en.ecn.purdue.edu * the STRESS!!!         ///   *
*                           *                   \\\///030 *
*                           * -Gigahertz!-  Amiga\XX/ 882 *
***********************************************************
	    -=[ In Stereo Where Available ]=-