Contents | < Browse | Browse >

===========================================================================
                   AN OPEN LETTER TO AMIGA TECHNOLOGIES
  Jake Hamby                                      hamby@aris.jpl.nasa.gov
===========================================================================

As a former Amiga developer, I was very interested to read about Escom's
plans for the Amiga and your intention to provide good support for
developers, and more importantly, support and marketing to encourage users
to buy Amiga computers.  I think you've done an excellent job so far, but
you've concentrated very little on your plans for the vast United States
market.  Selling the Amiga in the U.S.  will definitely be a challenge,
because the majority of computer buyers primarily care about Microsoft
DOS/Windows compatibility, but the North American market is too big to
ignore, or handle poorly.  I've given this area a great deal of thought
through the years, and I have 10 fundamental suggestions which will enable
Escom to sell at least 10,000 (and potentially many more than that) Amiga
computers in the U.S.  and Canada next year. 

1) Recognize the Amiga's advantages.  If you only compare raw price to
performance ratios, the PC appears to come out ahead, and in terms of raw
graphics and CPU performance, SVGA and Pentium CPUs have passed up the
Amiga.  However, the Amiga is still a force to be reckoned with, even in
1995.  Features such as "Plug and Play" of peripherals (which the Amiga
pioneered with their unique AutoConfig system, and which the PC still has
severe problems achieving), and a tradition of innovative software are
strong points the Amiga can advertise.  Another advantage, shared by the
Macintosh and graphic workstations, but not with the PC market, is the high
use of SCSI peripherals, which, although the initial cost is higher than
IDE, are infinitely more expandable and versatile, as well as higher
performing.  In fact, the only major disadvantage of the Amiga in 1995,
ironically, is the lack of DOS compatibility.  This issue is covered in
suggestion #5. 

2) Find a target audience.  While there is a market in Europe for the Amiga
as a game machine (a niche the A1200 was designed to fill), Americans
wouldn't even think to buy a machine with a keyboard to use as a game
machine.  So don't try to sell the idea of a game machine that can also be
used as a computer, or you will fail miserably.  Instead, when Americans do
want to buy a computer, they expect something reasonably powerful and
expandable.  Also, don't expect to target the first-time computer buyer;
they would be disappointed and probably confused to buy a computer that
wasn't a PC or a Macintosh.  Instead, I have three specific types of users
who would be interested in buying an Amiga:

Developers of vertical applications: Many custom software solutions do not
depend on DOS or any other specific software compatibility concerns.  A
programmer developing a data collection program for a factory or an
inventory system for a retail store only wants a solid, maintainable
development platform that they can quickly deploy within their company for
their custom application.  For example, many such programs have been
written on the PC using Linux, a free Unix clone, because the developers
were impressed by its built-in networking, development system, and robust
performance.  The Amiga has similar advantages, including extremely low
memory requirements, high performance multitasking, and an easy-to-program
GUI.  To position the Amiga in this market, you simply need to advertise in
magazines read by developers (such as Dr.  Dobbs Journal or even Linux
Journal), describing the Amiga platform's technical advantages, and provide
a C/C++ compiler (most likely SAS/C) for a low price, such as US$100.  You
may even consider bundling a C compiler with the operating system. 

Internet Service Providers: As I'm sure you're aware, the Internet is the
fastest growing and most visible segment of the computer industry right
now, and the world-wide demand for Internet service providers is at an
all-time high.  An ISP's primary requirement for hardware is a system with
fast Ethernet and hard drive performance, and a solid TCP/IP system.  The
Amiga provides all of these things, thanks to SCSI and AmiTCP.  Providers
requiring the highest performance can run either NetBSD/Amiga or
Linux/Amiga, two popular free Unix clones with good Amiga versions.  To
succeed in this market, start producing Amiga Ethernet cards for a low
price, such as US$150, bundle AmiTCP with AmigaDOS, and advertise the
Amiga's advantages in Internet publications and on the World Wide Web. 

Unix users: Millions of people, from hobbyists to professional developers,
have discovered the benefits of running Linux, FreeBSD, or other Unix
operating systems on their PC.  In addition, Unix (on PC's and
workstations) is widely considered to be a robust operating system suitable
for almost any mission critical application.  The Amiga has the advantage
of two free Unix clones, in addition to AmigaDOS, which is very similar to
Unix in many ways.  To position the Amiga to Unix users, who would be very
interested in its innovative solution, simply follow the same steps as the
other two markets: Sell inexpensive Ethernet cards and C++ developers kits,
and advertise in Unix publications describing the Amiga's technical
advantages for running Unix (such as SCSI hard drives and accelerated
graphics performance for X Windows). 

3) Satisfy the Potential market with Innovative hardware: I know you have
hinted you plan to port the Amiga to RISC and continue on the AAA chipset,
but there are more cost-effective innovations you can introduce today.  A
perfect example of this is the Iomega Zip drive: This drive is extremely
impressive because it combines low price (US$200 for the drive, US$20 for
each disk), high capacity and performance (100MB per disk, 1MB/sec transfer
rate), rugged construction, and quiet operation.  Power Computing, a maker
of PowerMac clones will soon be bundling Zip drives with their computers;
Amiga should do the same.  After all, the Amiga 4000 has a built-in SCSI
port, why not take advantage of it?

If you include an internal Zip drive with every system, then you can
distribute the OS and bundled software on a single disk, which would allow
users to run the software directly from the disk, or install it on their
hard drive if they desire.  This kind of simple, innovative solution is
something the Amiga really needs.  I would strongly discourage you from
spending too much effort on long-term projects, such as RISC, when you can
introduce this kind of new technology today.  On the other hand, I would
also discourage you from trying to introduce low-end PC-clone technology,
such as IDE hard drives, which lack the kind of open expandability that the
Amiga has always had, and which would make the Amiga no different from the
ordinary PC clone. 

4) Build an affordable, powerful machine: The A1200 is far too underpowered
to sell in the U.S.  market, as I've already mentioned, so the A4000 should
be the system of choice.  You should build an economical A4000 model,
perhaps with a fast 68030 instead of a 68040, 4MB of RAM instead of 6MB, an
800MB hard drive instead of 1GB, and/or a desktop case instead of a tower.
However, do NOT try to cut the price by removing critical features such as
SCSI, the floating point coprocessor or memory-management unit (because a
machine without both an FPU and MMU can not run Unix!), or other important
features that give the Amiga a competitve edge.

Bundle the computer with the operating system and basic applications (the
software you are currently bundling is perfect), plus an Iomega Zip drive,
and sell the whole thing for US$1700 (without monitor).  Make sure that the
user can use an inexpensive VGA monitor (this will probably require
modifying the motherboard with something like the A3000's flicker fixer),
and people will be knocking down your door to buy one (assuming you follow
the next six suggestions too)! 

5) Advertise to your potential market: I can not stress this enough!  The
primary reason Commodore failed in the marketplace was that they weren't
spending their advertising dollars effectively.  Escom should not try to
advertise in the United States through expensive television campaigns, nor
should they rely on word-of-mouth advertising (i.e.  Amiga fanatics
preaching to their friends to buy Amiga).  The solution is to target each
of the three audiences I've mentioned through trade magazines and the
World-Wide Web. 

6) Provide adequate sales channels: Another area where Commodore failed was
trying to sell the Amiga through mass-market retail channels (department
stores such as Montgomery Wards and discount stores such as Kmart, for
example).  After all, the Amiga will not sell to the average American who
only wants PC compatibility, so why try to target this unreceptive
audience?  Similarly, retail clerks would require extensive retraining,
because many of them have never heard of the computer, and so would have
difficulty selling it (I hear that you have this problem in your European
Escom stores).

Instead of this, you should primarily sell the Amiga through mail order and
through a dealer network of Amiga or Amiga/Mac computer stores (I'm sure
many of these stores have remained in business after Commodore's bankruptcy
by selling Mac hardware, and many of them would be interested in selling
Amiga again, if it was a viable option).  I suggest that you start a
toll-free number, such as +1-800-AMIGA-96 for people to call and ask for
their nearest Amiga dealer or mail order catalog. 

7) Even though the target audience for the Amiga in the U.S.  may not
require DOS compatibility every day, the PC is pervasive enough that nearly
every user will require the ability to run DOS programs when necessary. 
Commodore has already developed the 486 PC Bridgeboard, all you need to do
is update it to use AMD's 486DX4 100MHz CPU, which is high performance,
100% compatible with Intel, and very affordable.  In fact, if you can
provide a 100MHz 486 Bridgeboard for US$300 (assuming it can run Windows
95!) you will easily make 50,000 extra sales a year.  Simply put, if you
can give the buyer two computers for the price of one, they will be a
customer for life.  And since Shapeshifter, an excellent shareware
Macintosh emulator, is available, you are actual giving the users THREE
computers for the price of one!

8) Don't forget the Developers: As I've mentioned above, providing a cost-
effective C++ compiler and development environment is critical to the
success of the Amiga.  Developers should also be provided with the
documentation for programming Workbench 3.1 and other specifications that
have not been available since Commodore's bankruptcy.  To save expenses,
you should make this information available on the World Wide Web, perhaps
requiring developers to pay a small membership fee (such as US$40 per year)
to receive access to this information. 

9) Don't forget existing Amiga users: Our experience and confidence has
helped Escom's position immensely, and you will need to rely on the
existing user base to make your sales (in the United States as well as
worldwide) successful.  One way you can say "thank you" to existing users
is by providing the current system software (Workbench 3.1) to all Amiga
users for no more than US$100, which would more than cover the cost of
documentation, disks, and ROMs.

10) Provide a great system for a reasonable price, stand behind your
products, and the United States market will be a great success for Escom. 
Don't try to be the next Macintosh, but instead concentrate on market
niches that will most appreciate the Amiga.  Similarly, don't try to turn
the Amiga into a PC, because they are two very different markets.  In
summary, here are the kind of components you should be selling, and the
price I believe customers will be willing to pay (all prices in U.S.  
dollars):

$1700-$2500 - A4000 system (range from low-end with 68030/4MB RAM/850MB HD
    to high-end 68060 system.  All systems include bundled software,
    keyboard, mouse, floppy drive, and Iomega Zip drive.
$200-$1000 - Monitor (ranging from 14" standard VGA monitor to 17"
    multiscan)

$300        - 486DX4 100MHz PC compatibility card
$250        - Quad-speed SCSI CD-ROM drive (price if bundled with the system,
            otherwise users can add their own internal/external CD later)

$150        - Amiga Ethernet card
$100        - C/C++ compiler/developers kit