The most detailed information about the Amiga A300?

[Daedalus]

The project was started with the intention of being a lower RRP than the A500+. As I said above, the initial target was an RRP of $300, with the 500+ having a target of under $500. This is according to Jeff Porter, who was heavily involved in the development of both.

[grond]

ECS in a wedge-type Amiga would only have added something notable over OCS if they had also carried over the 32 bit chipmem bus from the A3000 (and added a CPU to do 32 bit reads and writes). If they had also added Amber instead of the TV modulator, we would now be having neverending discussions whether the A600 or the A1200 were the better package...

[Megalomaniac]

I've always had the impression that the A300 was intended to be a cheaper companion to the A500+, not a direct replacement, from what I've read from Commodore staff of the time. Indeed the A1200's working title was the A800, which was changed because A1200 more accurately conveyed the advantage over the A600. I'd assume that removing the upgrade potential and most of the ports would be a good way to cut costs. Integrating some of the chips into one would probably be possible too, that was a common way of reducing costs of hardware over time. The integrated modulator might've cut £10-20 off the potential RRP too,a s well as the added convenience. At say £250 that might have been a worthwhile product, not that much more expensive than a console, and cheaper once you'd bought half a dozen games. And it's true that the keypad wasn't needed for the kind of games people who were considering a console would have wanted, at least initially.

Compared to the consoles, the Amiga also had the selling point that you could grow into it. Even if you bought it mainly for action games, during a 3-4 year lifespan you could discover games like Civilization or Eye of the Beholder, or learn programming or making music, or start using it for word processing or spreadsheeting. This was still largely true of an A600, and would have been partially true of an A300 with no upgrade potential whatsoever. And the price gap for games was even bigger between consoles and Amigas if you factor in budget titles, coverdisk demos (often entertainment in themselves, as well as the try-before-you-buy aspect) and PD games. On the technical side, Amiga games were still improving when the Plus and A600 launched, with the likes of Outrun a distant memory (though still available on budget, but then so was Lotus 1 by mid-1992)

[duga]

With changing mono composite to color composite on the A500 Plus they could've milked that another 6-12 months and skipped the A600.

A1200 = A1000 with 020.

[eXeler0]

Not seen it mentioned here yet, but Im pretty sure there was a plan to have a built in genlock in this ”el cheapo ” Amiga. Anyone remembers this?

Tbh, wouldnt it had made more sense to do an OCS Amiga based console without a keyboard instead of the 64GS in 1990 and let that be the cheap alternative instead.
Also, when they realized they would miss the price target, could they have chosen a 68000 @14MHz instead? That must have been peanut money in 1991?

[grond]

Originally Posted by eXeler0

Also, when they realized they would miss the price target, could they have chosen a 68000 @14MHz instead? That must have been peanut money in 1991?

I wonder how much a processor without instruction cache (doing instruction fetches from chipmem all the time) would be throttled by the 16bit chipmem bus. That would have been even worse than with the EC020 in the A1200.

[Megalomaniac]

What benefit would a faster processor, in itself, have been to an Amiga with limited upgrade potential and no keypad? Would developers have designed games for it knowing that they'd play much slower for the majority of Amiga owners? Console-style games won't have benefitted much from a faster processor, it's 3D stuff and maybe strategy games that would.

[eXeler0]

Originally Posted by grond

I wonder how much a processor without instruction cache (doing instruction fetches from chipmem all the time) would be throttled by the 16bit chipmem bus. That would have been even worse than with the EC020 in the A1200.

That's a legit concern but then again the "expansion port" could have been designed for more than 1MB FastRAM instead of another MB of ChipRAM. It would probably been more useful for ppl. Not much was designed to require 2MB ChipRAM at that point...

Btw, On the Wikipedia page for A600 blame points mostly at Bill Sydnes for changing the design goals of the A300 project ;-)

[Lineof7s]

An excerpt from "Commodore: The Inside Story" by David Pleasance (2021). A very good read. Highly recommended.

THE AMIGA 600 DEBACLE
At CBM UK Ltd, with some very clever marketing, we had managed to keep sales of the Commodore 64 buoyant for two years beyond its realistic sell-by date. But in February or March 1991, after analysing the results from the UK Christmas sales, it became obvious that the C64 as a saleable product was dying – in fact, for all intents and purposes, it was dead.

I went to New York for a meeting with Mehdi Ali during which I explained to him that we now desperately needed a very low-cost entry point Amiga to replace the C64. The idea was to attract that same demographic target audience, ideally with a similar retail price point of £200, although I believed we might get away with an entry-level price of a maximum of £249.

I explained that the way to achieve that price point was to design and manufacture a very basic, low-spec Amiga, but with the potential to be upgraded, as consumers funds allowed, with the addition of external hard drives, extra RAM and so on. In this way, we would attract and convert the previous C64 buyers, whose demographic profile showed they were less affluent than buyers of our Amiga range.

It was agreed per my suggestion to call the new model the Amiga 300, which, I explained in great detail, would easily identify to the Amiga community and the public in general that this was a ‘lesser’ model in the range compared with the existing Amiga 500. Remember, at that time the Amiga 500 was selling in very big volumes and was a profitable line for Commodore. My idea was to add the Amiga 300 to the range as a replacement for the now defunct C64 (which had not been selling anywhere in volume for years outside the UK), so this move would considerably boost our worldwide sales volumes and revenues.

In September that year, all the general managers met in Frankfurt to agree the final basic specifications of the A300, and we left the meeting confident that corporate management were probably – and for the first time ever – doing what we, and the market, had actually asked for.

Imagine our surprise when in May 1992 we took delivery of a large quantity of computers with the model number A600 and not A300!

It transpired that after our Frankfurt meeting the German subsidiary – who had clearly missed the whole point of the need and demand for the A300 – had underhandedly told Mehdi Ali that they could not and would not sell any Amiga that did not have a hard drive included, thereby sabotaging the whole well-thought-out plan to entice C64 buyers into an affordable upgrade path. Mehdi Ali simply rolled over and conceded, but never had the balls to talk to me or any other GM about the change of plan.

This A600 would become a catastrophic disaster for the following reasons:
• Though it was smaller and had fewer features than the A500, it cost more to manufacture.
• Being called the A600 gave consumers the impression it was a higher specification machine than the A500, resulting in sales of the A500 slowing down significantly, virtually destroying that existing (and profitable) market.
• We had prematurely killed off the A500 by introducing the A600 instead of the planned A300 – and on which we made considerably less money.
• Consumers who bought the A600 believing it to be a superior model to A500 were very upset, believing they had been misled (which of course they had).
• The press, with whom we had an excellent relationship, absolutely lambasted our misguided rationale, and justifiably so.

Does that help at all?

[Bruce Abbott]

Originally Posted by Lineof7s

An excerpt from "Commodore: The Inside Story" by David Pleasance (2021). A very good read. Highly recommended.

Does that help at all?

That explains why it was originally going to be called the A300.

But David Pleasance was a salesman not an engineer, and his demand for a 'very basic, low-spec Amiga' with potential to be expanded with 'extra RAM' and 'external hard drives etc.' at a much lower price than the A500+ was probably not realizable.

What he was basically asking for was a cost-reduced A500+. What were they going to cut to make it cheaper?

- Smaller keyboard? that turned out to be an embarrassing mistake because a few existing games used the numerical keypad (yes, Amiga fans were that picky).

- More compact case and PCB? - not much to gain there if the case had to be big enough to take a full-size keyboard.

- Remove the serial, parallel, RGB and/or external disk drive ports? ...and people would complain about that too.

- Use smd parts? But at the time they weren't much cheaper, so the only advantage was smaller size.

- Put the modulator circuit on the motherboard? They did that.

- Forgo sockets for large chips and solder them directly onto the board? They did that.

- Use a less expensive switching power supply? they did that.

I'm struggling to think of anything else they could have done to make it cheaper. For compatibility it needed everything the A500+ had in it. 1MB base RAM was essential, with possible expansion to 2MB. Anything less and it couldn't run all the same software as the A500+, the then-current 'base level' Amiga.

They tried smd, and a smaller keyboard and case (which Pleasance didn't complain about) and an expansion port small enough to fit in that case (PCMCIA) that could take extra RAM etc. (not a faster CPU, but Pleasance didn't ask for that). So where did they actually deviate from Pleasance's idea?

The only thing they added that he didn't ask for was an internal IDE hard drive. But this was a no brainer. The PC world was going IDE because it was cheaper, and the Amiga was missing out. The circuit required to implement IDE was dead simple and cheap. Their only 'mistake' was making it internal (and of course 2.5" because a 3.5" drive wouldn't fit in the smaller case and might need a bigger power supply).

People complained about the price of 2.5" hard drives, but all you had to buy was the drive and a short data cable. To put a hard drive on the A500+ you needed an external case with PCB, interface chips and bus connector, and probably an external or upgraded power supply. Commodore's own solution - the A590 with 20MB XTIDE drive - was selling for £285 in 1991. An A600 with 20MB 2.5" hard drive was only £90 more than one without a hard drive.

The PCMCIA port was cheap too. It may have been risky to expect this new 'industry standard' port to get a good variety of cards at reasonable prices, but creating yet another proprietary expansion port would also be risky. As it was a number of Amiga peripheral manufacturers did make a variety of affordable Amiga-specific PCMCIA devices including CDROM drives and sound cards, and network cards worked in it too.

The A600 didn't match Pleasance's wishes, but it probably couldn't have anyway because they couldn't have made it for the price he demanded. Instead they did the sensible thing, and made it better than just an A500+ in a smaller case (and cheaper too once you added a hard drive).

Pleasance didn't own an Amiga and didn't seem to know what people in the scene wanted. He was only thinking of some way to re-energize the market the C64 was in, not thinking about why people were leaving it.

[Bruce Abbott]

Also of interest is the original Gayle spec, which is titled "Gate Array for A300/A500+".

One of its features is EEPROM support, described as "a possible replacement for a floppy drive in an extremely low-end variant of the the A300".

It also mentions ARCNET using a COM20020 chip, and in the memory map they have 64kB reserved for "external IDE".

Finally in section 19.0 it says "Four registers are included that facilitate support of the cartridge slot...", suggesting that it might originally have been intended to have a proprietary cartridge slot rather than PCMCIA.

Seems the engineers were considering a range of features for the 'A300', and possibly improvements for the A500+ too.

They also tried to make a cut down A3000 which made it to prototype form - the A2200. This also had ECS, internal IDE and on-board Fast RAM, with either an 020 or 030 CPU and the same CPU slot as the A3000. It morphed into the A4000 when they realized that AA wasn't making progress and they had to get machines out with better graphics quickly because fans were getting tired of ECS.

[Matt_H]

I think Dave Haynie's Deathbed Vigil video gets into the history of the A600.

Commodore engineering intended the A300 as an entry-level machine to be positioned below the A500(+) in the product line, a very cheap way to get new customers into the Amiga ecosystem. With vastly fewer chips than the 500 and only 1 socketed chip compared to 8 on the 500 it should have been substantially cheaper to produce, despite the addition of PCMCIA and IDE.

For whatever reason, management/marketing decided to replace the 500 (which was still selling very well) with the 300 rather than complement it, increased the price, and then bumped up the model number to 600 to make it seem more valuable. Furthermore, they launched it only a month or two before the much more desirable 1200 but Commodore blew the most of their credit on parts for the 600. As such, they couldn't manufacture enough 1200s for the Christmas 1992 season and no one wanted the already-obsolete-but-still-fairly-expensive 600. Absolute financial disaster.

If Commodore had kept the price point below the 500, things could have looked much better for Commodore in 1993 and beyond.

[Matt_H]

Originally Posted by Bruce Abbott

Also of interest is the original Gayle spec, which is titled "Gate Array for A300/A500+".

One of its features is EEPROM support, described as "a possible replacement for a floppy drive in an extremely low-end variant of the the A300".

...



Finally in section 19.0 it says "Four registers are included that facilitate support of the cartridge slot...", suggesting that it might originally have been intended to have a proprietary cartridge slot rather than PCMCIA.

I suspect this is the groundwork for the never-used AmigaXIP mechanism, which was a design for running bootable software (i.e., games) from PCMCIA. I recall that it's at least partially documented on the Developer CDs.

[Bruce Abbott]

Originally Posted by grond

I believe most of the more useful PCMCIA expansions really were PC-Card expansions, i.e. relied on the later 32 bit variant of the same form factor and can't be used in Amigas.

Not only that, but they often needed a more powerful machine to make good use of them.

The problem with PCMCIA was that laptops were going for more integration, so things like sound systems and USB ports were being built onto the motherboard instead. Therefore the demand for those functions on a PCMCIA card disappeared. Modern laptops have ExpressCard which is totally different.

However for a while there were a fair number of PCMCIA PC cards that could have been used on the Amiga if drivers were written for them. As well as SRAM there were FlashROM cards, network cards, CDROM drives and MODEMs. Compact Flash needs nothing more than a passive adapter.

The real question is what would you want that would be appropriate for the A600? A sound card isn't really necessary since the Amiga already has pretty good sound built in. USB didn't exist until 1996. The standard serial port could do 9600 baud no sweat, which was the fastest MODEM speed on the market until the SupraFAXModem 14400 was introduced in 1992. An external MODEM was more flexible and could be used on other computers.

Many fans complained about the 'useless' PCMCIA slot that replaced the A500's expansion connector. But most A500 users didn't have anything plugged into it. When they did it was generally only a RAM and/or hard drive expansion. But the A600 took 2MB and a hard drive internally, so you only needed an SRAM card to match the vast majority of expanded A500s. You couldn't put a faster CPU in the A600 (yet), but how many A500 users did that?

IMO most of the objections were just a knee-jerk reaction to change. We had people complaining that they had just bought an A600 and couldn't immediately upgrade it to the max. Yet if those same people had bought an A500+ you can bet they didn't rush out and buy a fully loaded expansion box the very next day.

Sure it took a while for Amiga peripheral manufacturers to make A600 specific stuff, but that was the same for other Amiga models too (for obvious reasons). To expect otherwise was unreasonable. It was the usual Amiga fan's "I only want it because I can't get it" tantrum.

[Bruce Abbott]

Originally Posted by Matt_H

I suspect this is the groundwork for the never-used AmigaXIP mechanism, which was a design for running bootable software (i.e., games) from PCMCIA. I recall that it's at least partially documented on the Developer CDs.

Yes, I've read about it.

Technically it was a good idea, but had no application because to be successful a game cartridge has to be cheap. At that time PCMCIA cards were not cheap to make. Furthermore there wasn't a big enough user base to justify putting games on cards, and of course Amiga fans were used to only paying the price of a blank disk for their games.

The only way it might have worked is if a killer game was produced that could only be run from the card, perhaps with additional hardware like some console carts had. But the A600 was the wrong machine for that.

If game cartridges were to be practical on the Amiga it would need to have a cartridge slot like other machines had. Then the cart only needs a cheap edge connector on the PCB, and with a reasonably sized case you don't need ultra-thin smd components.

In some respects it's a pity Commodore didn't do this, because cartridges have several advantages. The ROM effectively gives the machine enough memory no matter how much is needed, and it works at full speed like FastRAM. Loading times are almost instantaneous, and there is less need to 'uncrunch' data before use. They were also much more difficult to pirate (though many Amiga fans might not have seen that as a plus).

But if they were going to that they had to do it from the beginning. When cartridges were introduced later in the life of a platform they always failed.

[grond]

Originally Posted by Megalomaniac

What benefit would a faster processor, in itself, have been to an Amiga with limited upgrade potential and no keypad? Would developers have designed games for it knowing that they'd play much slower for the majority of Amiga owners? Console-style games won't have benefitted much from a faster processor, it's 3D stuff and maybe strategy games that would.

This is also true. A faster 68000 wouldn't have helped much anyway and would have been another source of incompatibilities for hardcoded hardware banging games. It might have been of some use for harddisk based OS-friendly use.

Originally Posted by eXeler0

That's a legit concern but then again the "expansion port" could have been designed for more than 1MB FastRAM instead of another MB of ChipRAM.

Yes, certainly. A lot could have been better about the A600...

[grond]

Originally Posted by Lineof7s

An excerpt from "Commodore: The Inside Story" by David Pleasance (2021). A very good read. Highly recommended.


Does that help at all?

Well, I wonder what this A300 with less features than the A500 should have been. What would he remove from the A500 without breaking compatibility? The idea to replace the C=64 may have made sense from the business perspective but there simply wasn't any meaningful way of doing that other than producing computers identical to the A500 in technical features (to not break compatibility) in a cost-reduced manner. But it doesn't take a genius to figure out that producing the same product for less may be good for business.

[eXeler0]

Originally Posted by grond

Well, I wonder what this A300 with less features than the A500 should have been. What would he remove from the A500 without breaking compatibility? The idea to replace the C=64 may have made sense from the business perspective but there simply wasn't any meaningful way of doing that other than producing computers identical to the A500 in technical features (to not break compatibility) in a cost-reduced manner. But it doesn't take a genius to figure out that producing the same product for less may be good for business.

Small keyboard-less OCS A500 with 512k RAM but expandable to more.. Also with a keyboard connector.

[Megalomaniac]

I don't think removing the keypad was as trivial a matter as Bruce makes out, especially as the games that used the keypad were mostly of a particular genre for which the Amiga had a major lead over the consoles but was increasingly being encroached on by the PC. Owners of other Amigas were negatively impacted here.

Bruce also overlooks the inconvenience for hardware manufacturers, who had three brand new models to deal with inside a year. Anyone producing new or updated hardware for the Plus will have found it obsolete within 6 months, and in many cases the new product they made for the A600 will have needed a similar one for the A1200. And, from memory, there were no accelerators for the A600 until 1995.

All of this may have been acceptable for a budget companion product to the A500+, but definitely not for a direct replacement. David Pleasance's idea might have made sense from a business perspective, but it sounds like the engineers did try to produce a cheaper Amiga, they just couldn't find enough to significantly reduce the price.

I don't think the A600 was a better product than the A500+, and definitely not a worthwhile replacement. Perhaps if they'd released it alongside the A1200 instead, so as to give the A500+ a year on the market, it might have made sense, but not they unannounced haphazard way they did it.

[grond]

Originally Posted by Bruce Abbott

One of its features is EEPROM support, described as "a possible replacement for a floppy drive in an extremely low-end variant of the the A300".

Finally in section 19.0 it says "Four registers are included that facilitate support of the cartridge slot...", suggesting that it might originally have been intended to have a proprietary cartridge slot rather than PCMCIA.

Very interesting, thanks for bringing that up. I would assume that the cartridge slot was the intended replacement for the floppy drive and a completely separate idea than PCMCIA. There were games on cartridges for the C=64 and I can see how that idea was appealing to Commodore: save the floppy disk drive at the expense of a simple connection port, provide a pretty much copy-protected medium for games, instant boot, mechanically reliable. People could have connected an external FDD drive for compatibility. I have to say I like that idea.

Perhaps there is a link between cartridges and PCMCIA. The A600 can boot from a PCMCIA medium (which has always been rather surprising to me). This may be functionality left over from the cartridge slot idea.