Contents | < Browse | Browse >

===========================================================================
==  Outlook on the Future                               By:  Aric Caley  ==
===========================================================================

  I'm going to open my big mouth once again and lay out what I believe is
the right plan of action for the Amiga.  I hope that this makes its way to
whoever (CEI or C= UK) actualy winds up winning the bid for the Amiga (my
personal preference would be CEI, and it looks like they might win).

I maintain that the Amiga market (and Commodore's market historicaly) is:

  Home computer users (basic WP, etc) and game players.

  Video and multimedia.

  Specialty applications (Kiosks, etc)

  Developers/techies.

Roughly in that order (in terms of unit sales)

  We don't, and likely won't ever, compete with high-end machines (like
SGI's and the like) except in special things like video, and even then only
because sometimes the Amiga is much cheaper (Video Toaster).

  We can't compete directly with IBM clones.

  Another market which I think we can, and should, make inroads to is
education.  Amiga's could be presented as a very attractive educational
computer.  First and possibly foremost, they are fun!  They are colourfull,
responsive, and just plain fun.  They have a high "fun to cost" ratio.
With emulation, they can run almost anything.  Multimedia perhaps has its
greatest potential as a learning tool.  Finaly, kids parents can actualy
afford to buy the same fun computer their kids use at school.

  This is all important to remember when you consider what kinds of changes
should be made to the Amiga.

Marketing:
----------

  What's that?  Oh yeah, that thing that Commodore didn't do very well.  
Actualy, I have several very fine examples of great brochures and marketing
devices from Commodore -- they are much more profesional looking than, say,
those from Atari.  Problem is, they never followed through on them.  Always
cutting off funding too soon.  And, not getting the materials to the people
who needed to see them.

  The new Amiga company needs to focus on the Amiga's strengths and its
market.  "Multimedia" and "Video" need to be stressed, as well as "Fun",
"Inexpensive" and "Powerfull".  Apple has been going hog wild with their
Multimedia campaign, and it's just rediculous because the Amiga can do it
so much better.

  We need a new image.  With Commodore gone, there is an oppurtunity to
completely change the Amiga's image.  I propose the introduction of the
Amiga V Series.  We won't tell anyone outright what the V stands for, but
just between you and me, it means "Video" and it's also the roman numeral
for 5 for "Fifth generation".  I see five stages in the Amiga's history:
A1000, A500/A2000, A3000, A1200/A4000, and finaly, the all new born-again
Amiga rising like a Pheonix in 1995 (note the 5 in that year).  Of course
there is also the law of fives for you Discordians out there (see
"Principia Discordia").  This is the kind of grand, unified, directed,
subtle marketing that we need.

  Apple keeps claiming that the 68040 used in their Performa's and Quadras
is a "66/33Mhz" chip.  Amiga should claim it too.  "80/40Mhz" would look
good when people are comparing to PowerPC's, Pentiums and 486DX2's and
DX3's.  If they can get away with that on the 040, it should apply to the
060 as well.  "100/50Mhz 060" and "132/66Mhz 060" sounds real tasty.

Use buzzwords: RISC, 3D, Multimedia.

Slogans:

"The Amiga V! Machine"
"Multimedia is more than pretty speakers and a CD-ROM drive"
"Realistic Multimedia at a realistic price"
"Only Amiga makes it Possible!" (hey, it's still a great one!)


V for Victory!
--------------

  The V machines should have CD-ROMs standard.  3x or 4x speed units on
high end machines would be nice.  We need SCSI-II on the A4000, and the DSP
(basicly, WE WANT THE DAMNED A3000+ THAT COMMODORE F*CKED UP).  Naturaly,
the CD-ROM should be CD32 compatible.

  Lets get realistic with the amount of HD and RAM included with systems. 
120MB 3.5" IDE drives are under $100, 250MB for $150, 560 for $219.  The
"V1200" should come standard with at least 120MB HD.  "V4000's" should
typicaly have one in the 500MB range, and options for 10MB of RAM or more.

  We need to lower the price of the 4000, and get at least one more mid
range model between the 4000 and the 1200.  We need some kind of basic 040
model for around $1000.

RISCey Issues:
--------------

  RISC is not absolutely essential in the short term, since most Amiga's
are used in low-end applications, and the 060 will provide a good amount of
power for low-end systems for quite some time yet.  I think even something
like an A1200 with a 40Mhz 030 would still do well this year.

  Long term, RISC is obviously important.  Therefore, it should be done
*right* -- make the best choice of CPU, make it work real well before
releasing it.  Let Apple and IBM cut their teeth with their adventures with
PPC and see how it turns out.  We might even want to use more than one kind
of RISC, and give people the choice -- perhaps PowerPC for low cost RISC,
and MIPS or Alpha for high end kick butt 3D rendering Amigas.   Who knows?


  As for AmigaOS under RISC.  Some say just ditch AmigaOS.  Why?  Apple is
keeping their MacOS on the PowerPC, and we all know "it sucks!" :).  It's
going to be a long time before they get it 100% ported to native PPC code
(if ever) -- right now it's nowhere near 100% PPC native.  Apparently a
good portion of Mac system code was written in assembly.  I may be wrong,
but I think most of AmigaOS is written in C.  Things like the exec kernel
(the multitasking core) are in assembly, but, it's pretty small and could
be redone in C and compiled native to the RISC chip we use.

  I think AmigaOS could be ported to RISC fairly easily, probably easier
than Apple porting MacOS, and it would run really fast -- most likely
faster than MacOS.  And in the process of porting, we could easily fix
things like lack of memory protection and virtual memory, etc.

  Look at "AppleOS" -- on the PowerMac it still does not have pre-emptive
multitasking or memory protection.  They are going to have a very difficult
time adding these things and they are not expected for another year.  And
when they do get it done, it will be big and inefficient.  Think of how
nice AmigaOS will be on a fast RISC processor!  And our software base,
though small, already supports Amiga-tized ideas like multitasking and IPC
and shared libraries and all that good stuff.

  I understand that, apparently, the engineers have/had decided on PA-RISC
for the Amiga, and that they plan on porting AmigaOS to it.  This sounds
like the right track to me and I hope whoever wins the buyout will properly
fund the development and completion of this project.


Bundle!
-------

  (not "bungle" or "bunghole", commonly associated with former C=
management).  Bundle products.  It's an easy way to add things, AND it's a
good incentive for developers to stay:

* Lets get CD-ROM on as many Amiga's as possible.

* Virtual Memory.  Not _real_ important to the typical Amiga user, but easy
enough to add to AmigaOS: just adopt one of the several already available
solutions.  In a "VM Preferences" application, you could switch it on and
off.

*  Put Envoy back in the OS.

* If necessary, bundle one of the third party 060 cards with the A4000 to
create the necessary A4000-060; do the same for 33 and 40Mhz 040 cards with
better memory systems.  The Cyberstorm board looks really nice....

* Bundle a regular graphics board?  Cybervision also looks damned nice...

* "V4000".  Video Toaster bundle: Includes card and basic software, CD-ROM,
500MB HD, 8MB RAM.  Make this as cheap as possible (I think an A4000 should
only be around $1000, so with a stripped down Toaster bundle a price of
$2000 seems reasonable).  Can Apple offer a system with this kind of video
power at any price?  Included software would include the basic stuff for
doing genlocking, titling, a simple switcher with effects, etc.  Perhaps
just an older version of the Toaster software.  Lightwave you must buy
separately (which is perfect since NewTek already sells it separately).

* Low end video bundle: V1200, genlock, software, 120MB HD, CD-ROM.  
Includes video titling software, maybe something like Scala, lots of cool
pictures and HAM8 animations on CD-ROM.  Lets see Apple top this for <$800

  Have you seen that insipid Apple commercial of the kid who does a video
about his vacation?  Give me a break.  The "video" was animation and it was
choppy.  What did that system cost?  It must have had a real time video
digitizer -- at least 15FPS (ha!).  Meanwhile you could have done better
with an A500 and a genlock in 1988.


Return to OS.
-------------

  The yellow brick road may need a lot of repair and work, but we can do
it.  OS upgrades should be more flexible.  There could be the "basic"
upgrade which is like what we have now, and then there is the upgrade that
includes the cool bundled stuff (perhaps this version comes only on
CD-ROM).

* CD-ROM "Multimedia" OS Upgrade.  Comes with new systems standard.  
Contains full AmigaOS with online AmigaGuide help.  Bundled VM system and
Envoy networking (throw in lots of PD goodies that use it).  Lots of the
best PD software (a term program, ToolManager, Trashmaster, etc).  And of
course, lots of cool pictures and HAM8 animations to show off with, maybe
some AGA games too.  Heck, why not throw in a PD C compiler too.  It would
be nice to have some multimedia authoring software -- maybe Amigavision or
Scala?

* RTG -- work on it, shouldn't be too hard to do.  Don't really need to
have it built in real quick since all the decent Gfx cards already have
solutions, might as well take the time to do it right.  Work with the
developers to make the official solution the best.

* Memory Protection.  I heard that Commodore had actualy implemented it,
but it caused some compatibility problems.  Well, I say, PUT IT BACK IN!!!
Sure, it WILL cause some problems.  So, you phase it in.  First, warn the
developers and tell them how to make their applications compatible.  Then
put it in.  Have a switch in preferences so you can turn it off.  Once it
is there, high end Amiga owners will simply demand that software be able to
use it.  Low end Amiga users (most of them!) won't particularly care (they
don't have MMU's anyway).  Applications that support it will be protected
from each other and from errant non-MP programs.  Hell, if Apple can phase
in MP and Pre-emptive multitasking (We just have to get a standard
abreviation for that one) into the Mac, surely we can phase in MP.

* Do more work on the Datatypes system (more drivers, make it FASTER!  Have
it use file notification -- a really easy way to get "Hot linked"
updating).  It should come with drivers for MPEG and JPEG, Postscript, GIF,
PhotoCD, etc.  Datatypes are potentialy a very powerful, but as yet
untapped, feature.

* Pre-emptive, prioritized multitasking.  Ooops!  I forgot, we already have
that!  Too bad Windows does not.  Nor does System 7.x on the Mac.  The Mac
will have to wait for TWO *major* upgrades before getting it (they are
code-named "Gershwin" and something else equaly stupid sounding).

Amiga Pheonix.
--------------

  I still say that the AAA chipset sounds pretty damn nice.  Now, I've
heard about this supposed 3D/RISC chipset.  But I don't know anything about
what it can do.  It may be really nice for a game console, but it is
suitable for a general purpose PC?  Could it be combined with AAA to get
the best of both?  Why not do both chipsets?  With RTG they won't have to
stick to just one chipset anymore.

  AAA still seems like the ticket for Multimedia Amiga's.  With totaly
programable video modes and both DRAM and VRAM configurations, support for
fast floppies (3.5MB Quad Density drives), fast serial ports, 16 bit sound,
sprites, blitter, copper, HAM modes, 15 and 24 bit true colour, easy
genlocking, chunky pixels, and lots more, it is *the* "Multimedia" chipset.

  Combined with a DSP and an 060 it would still be a pretty nice system,
even nicer with one of the RISC CPU's.  Call it the "Amiga Pheonix V5000"
and we have a winner on our hands.  Considering how close AAA was to being
completed (working silicon with a few bugs), it seems like an AAA machine
could be released this year.  And the "PA-RISC/3D-RISC chipset" Amiga could
be released in mid 1996.