Wally Dug's Historical Guide to the Amiga

Wally Dug's Historical Guide to the Amiga

Sit back and enjoy this journey back through time with that ever-ageing historian, Wally Dug

Before I start this month's Guide, I would like to thank all of you for the many wonderful letters which you have sent, especially to C Sinclair of Cambridge. Thank you for your kind comments, Sir!

Ten years ago, I was the first ever person outwith Commodore to sign the infamous Non-Disclosure Agreement (this thing makes the Official Secrets Act look like a copy of The Beano) which was later signed by other "trusted" people such as Jeff Walker. This Agreement was essentially a bribe - if I kept my mouth shut for ten years, I could see Commodore's latest developments. Well, the ten years are over now, so JAM can exclusively reveal not only the development of the current Amiga, but also the development of the future Amiga!

Just before the Commodore 64 was launched in 1982, the development team at Commodore were already working on its replacement - the Commodore M68K. The M68K was based and built around the Plus/4 (at this time, production was just starting on this computer) and, in fact, the Plus/4 has many features of the current Amiga (for example, a screen resolution of 320 by 200 pixels, screen windows and interactive software).

At the heart of the M68K was the Motorola 68000 chip that we all know and love, hence its name, and a massive 256Kb of memory (remember - this was 1982) and could produce 256 colours and four channels of sound (two "music" channels and two "white noise" channels). In fact, it is amazing how true to the M68K the Amiga has stayed, despite being completely different. Of the 200 prototype M68Ks built, only 132 ever worked and, at the last count, there were only nine left in working order - five which are held by Commodore, three others in the States and the sole survivor in Europe is owned by me.

Ask anybody which part of the Amiga is the most important and they will probably say the set of four custom chips. Yet, believe it or not, there were no custom chips in the M68K at all. (Incidentally, the Manager of Research during the Amiga's development was called Gary; the three female developers in the team were called Agnus, Denise and Paula.)

There was no Workbench either. Instead, we had the Workhorse which was a crude variation of Workbench, but still far advanced of all the "latest" machines on offer at the time. When I first saw Workhorse, I was disgusted: "That's for wimps!", I said, and the term stuck! Admittedly, Workbench has progressed to magnificence (Workbench 3.0 is absolutely brilliant and so easy to use).

There was also a version of today's AmigaDOS known as DOSSER (Disk Operating System Structured Enhancer Routine) which is much faster than AmigaDOS, but larger in byte size and smaller in command size. For once, Commodore listened to the criticisms directed at one of their products and made the necessary adjustments. Two years later, in 1985, Commodore launched their new "baby" - the Amiga 1000 - to high critical acclaim.

Not ones to rest on their laurels, Commodore continually developed the machine to the 500 and 1500/2000 that we all know and love and to the 3000 that we all know and lust. And the rest, as they say, is History.

So what becomes of the Amiga now? Will Commodore continue to support the family or will they dump it for new systems?

There has been much hype and speculation in the media recently about the Amiga 4000, the "last" edition to the Amiga family, and while much of this is true - it will probably be released in September/October in the States and January 1993 in Europe - it is by no means the last in the line of Amigas, for the development team are currently working on the Amiga Master.

The Master (not the Amiga 5000 in case of confusion with the 500 - nice move, Commodore) is a treasure-trove of beauties, although some details are still very sketchy as some components are still in the development stage. It will be based around either the Motorola 68050 processor (with built-in 68883 maths co-processor and memory management unit, of course) or the 68055 (68050 plus built-in 68000 for total compatibility) if it is ready. It will definitely include SCSI (Super Chip Set Mark I) which will give at least 4Mb of chip RAM.

The Master will come with 16Mb of RAM on board, expandible to 64Mb, and will sport both Kickstart and Workbench Version 4.0 (or above). Extra memory will come via a 21ms access time 200Mb hard drive which will be able to be used as virtual memory, in other words, you could have 264Mb of RAM to play about with...

Storage comes from a CD-Rom drive (CDTV and PC-CD compatible) and from a single floppy drive. This floppy unit will be able to accept the current Fast File System disks as well as the proposed Super-D System in much the same way as Workbench 2.0 supports both the Old and Fast File Systems. This super density system will give 3.52Mb to a single 3½" floppy. However, if Super-D is not ready in time, Commodore will revert to a 1.76Mb High Density System.

Hardware emulators may soon disappear as the Amiga Master will offer complete IBM PC-AT, Apple Macintosh and Unix V Point 4 compatibility via a new custom chip (the PRIM chip - Personality Routine Interchange Module) currently in the final stages of debugging.

Other features include 24-bit colour as standard, 8 music channels with the possibility of a derivative of Roland's new Sound Space system being added to the 16-bit stereo sound, four ZorroIII and six ZorroIV expansion slots, two video slots and double parallel and serial ports.

The most exciting aspect about the Master, however, is its pricing policy. To quote from the Master's design specification "...an affordable, yet powerful supermedia machine". One source from the design team suggested to me recently a price of "about $7500" - approximately £4000!

 

The M68K

M68K Boot ScreenThis is what you see when you first switch on the M68K. Not a lot, is it? In fact, the orange system information (more on this later) would not have appeared on a production model.  The black border around the screen signifies that the Workhorse is required.

 

Workhorse V1.02

M68K Workhorse ScreenAfter we type RUN WORKHORSE, this is what we see. Note not only the lack of memory on the memory meter, but also the clock.

As mentioned earlier, Workhorse was rather crude by comparison to Workbench, a fact that can be seen by the large icons. It was not very colourful either. Apart from the blue border signifying that Workhorse is loaded, all of the icons are black and white.

However, windows on Workhorse are remarkably similar to Workbench - note the "fuel gauge" and close gadget. There is also a single back-to-front gadget which works in exactly the same way as the back-to-front gadget in Workbench 2.0 and above. Although the window cannot be resized, it can be moved around the screen using the drag bar.

Unlike today, you cannot double-click on an icon to run that particular program. Instead, by clicking on an icon using the left mouse button, we highlight that icon. To run that program, we click with the right button to activate the entire menu, find RUN (in the APPLICATION menu) and highlight that by again clicking on the left mouse button. (To load a directory, we would highlight the WINDOW option.)

M68K Workhorse ScreenUnder the SYSTEM menu, we have HELP which attempts to help the user find his way around Workhorse. Unfortunately, it is not very good. VERSION tells us the Stable version number, Workhorse version number and creation dates. SYSTEM resets the system in much the same way as the famous Three-Fingered Salute, Amigan Death Grip, etc.

With the APPLICATION menu, we can load directories, run programs and KILL programs, i.e. stop them.

If we SORT the GARBAGE, we are simply bringing up the garbage window to view all programs/directories that we have placed there. If we REMOVE the GARBAGE, this deletes the contents of garbage.

 

Dosser V1.05

Unlike today's Workbench which runs "on top" of the CLI, DOSSER runs "on top" of Workhorse, but by shutting down as much of Workhorse as possible (e.g. the actual Workhorse screen), it is able to get much more memory. You can tell that DOSSER has loaded by the orange border around the screen.

M68K DOSSER screen showing commandsDespite it's size (those 25 commands are all of DOSSER's commands), there are many useful features. BLANKER is a screen blanker which runs in the background and turns the screen off after a definable time. By typing BLANKER 1 5, BLANKER would show a black screen after five minutes of non-use; BLANKER 2 5 would cycle slowly and smoothly through all 256 colours after five minutes.

Many commands have extra commands built-in, such as HELP. HELP INFO would tell you that INFO was a command that gave you full screen-based (INFO S PROGRAM) or printer-based (INFO P PROGRAM) instructions on how to use the program specified after INFO. If HELP INFO FULL was used, the user would get a detailed description as well as a step-by-step guide to INFO and its uses. Note also the MORE command which would not only give a more verbose description of a command, but also various examples.

M68K DOSSER screen showing the SYSTEMINFO commandAnd now to SYSTEMINFO. This command can tell you everything about your system (if SYSTEMINFO FULL is used, it will tell you absolutely everything - seven screens full of information!):

Stable is the equivalent of Kickstart and this is the version number of your Rom;
Motorola 68000 is the CPU;
Free memory in bytes;
All programs loaded - in this case, Workhorse V1.02 is the main program loaded with DOSSER V1.05 being a sub-program;
Workhorse1.02: is the name of the system disk and, as you can see, it was created at 2:45 pm on Friday, 25 March, 1983. There are only 85 bytes of free space left on this disk, which is 0.4%.

The next screen gives further disk information - how many directories and programs are on the disk; largest continuous free space; the status of the disk (i.e. write-enabled or write-protected) - along with other information which is nice to look at, but not very practical (number of MIPS; frequency of clock; how long you've been on the system; etc.).

 

Workbench 4.0

Workbench 4.0 is not going to be that much different from Workbench 3.0 (which is itself only slightly different from Workbench 2.0 - the real changes are "under the bonnet"). There will be new screen modes which take full advantage of the SCSI and the Topaz bitmap font will finally be killed off in favour of the Topaz outline font (!). The main difference in Workbench 4.0 is that all programs will have ARexx ports and will use the ARexxII language currently under development. Based around ARexx, this new implementation will not only provide the "update" feature which Soft Logik's HotLinks provides, but also other goodies such as a high speed, small size and simplicity in use (it says here...).

Kickstart 4.0 continues the Commodore policy of getting as many of the AmigaDOS commands off the floppy and into the ROM as possible. The main change with System 4.0 is AmigaDOS 4.0. Gone are the infamous CLI and Shells to be replaced with a SID-like interface, again fully ARexxII compatible. This will be similar to EOS (Enhancement for the Operating System) supplied with Workbench 3.0, but Commodore, as usual, have promised a major upgrade. I hope this is true - EOS is so slow and buggy that I'm glad that a Shell is supplied with Workbench 3.0.

 

EOS V0.9

EOS is the replacement for the Shell which is shipped with Workbench 3.0. It has a SID-like environment so should be familiar to many users. However, that is where any similarities end.

EOS - Enhancement for the Operating SystemEOS is very slow and buggy. The commands which you see in the boxes cannot be amended from the default. This gives rise to commands such as SetPatch taking up valuable space as it is usually the very first command to be executed in the startup-sequence and once executed cannot be run again. Another waste of space is the Echo command. EOS works on the principle that the commands in a box can only be executed and not inserted into a document. In other words, when you click on Echo, a new line is printed on the screen in a box similar to the Execute Command... item with Workbench 2.x instead of the user being able to specify what and to where output should be.

Another gripe is the screen format. EOS uses an interlace screen for increased information. Unfortunately, it also uses the new double-height font facility of Workbench 3.0 which cancels out the double-height screen benefit. Again, this cannot be changed.

 

Next month, we have another Build-It! feature when I'll be bringing you "Wally Dug's Build-It! Guide To Decoding Those Bleeping Noises Which We Sometimes Hear On The Radio And Which Are, In Fact, Satellites In Space Sending A Signal To Mission Control". Until then, keep sending in the letters.

Now, do you want to go back to the Article Index or go to the next article?

Copyright © Wally Dug Multimedia 1998