Was anyone else disappointed with the A1200?

[Karlos]

Let's not forget that there were dubious CPU socket upgrades for many older PCs that put a faster CPU on a board to plug into the original socket. I don't specifically recall any that had their own local memory though.

I don't think they were a success, the constantly falling prices for PC hardware over time just made it more cost effective to upgrade the whole system after some years.

[Karlos]

Originally Posted by TCD

All this energy spent on discussing what-ifs... Amazing

I honestly think the only "what if" that makes sense in the context of the historical development of the A1200, based on what commodore had in 1991/92 is the question of including fast ram in the design by default. Even if it was only 1MB that would be up redundant if a trapdoor expansion installed more (much like the PCMCIA being unusable with 8MB). The downwards trend of memory pricing in by the end of the 1980s shouldn't have required a crystal ball to project as a reasonable additional base cost by 1992. Having to use a trapdoor slot expansion to hold the memory in the first place was a significant increase in the end cost to the user for a given amount of memory.

Having the CPU able to run at its full potential and less dependent on chip ram cycles helps AGA run at its full potential too. This was a missed opportunity and in my view the only mistake in the A1200 given the intended price point. While all the talk about chunky pixels, etc. is interesting, AGA is what they had in 1992 and anything else is a redesign too far if it was to be released that year.

[Promilus]

386SX was upgradable the same way A600 is... I don't get it why Bruce seems to be oblivious to that. Those upgrades weren't all that popular in PC world (exactly because you could always just buy new and faster PC) but they were there. Now then... while ISA was obviously rather underperforming (real vs peak bandwidth) it was there for almost any PC out there with a range od expansions. A1200 had basically only turbo slot for cpu, rtc and ram initially (and scsi sometimes). It took several years to get hands on anything else allowing RTG when PC had already PCI dominance and was moving to AGP. So that expansion-friendly nature of A1200 seems to be overstatement.

[Bruce Abbott]

Originally Posted by Gorf

We had this already - it depends of course, where you lived on this planet in these years.
I do not doubt your experience in New Zealand as a shop owner. But things were different in Germany (own experience) and USA (conclusion).

I remember very well quite a few friends getting VLB equipped PCs in that period of time - despite me showing them my A3000

Anecdotes aside, sales figures for different CPUs show that in 1992 most people were buying 386 machines. This did change rapidly, but let's not forget that Commodore also released a 486 equivalent in 1992.

Oh wait, of course we will ignore the A4000 because it cost about the same as a name-brand 486 PC. Unacceptable! If the Amiga isn't both more powerful and much cheaper than an equivalent PC then I don't want one! Bad Commodore for not breaking the laws of physics!

The 386SX machines reached their very end of shelf live in 92 - you really can not argue that these were somehow the lates technology! Even down under...

Of course not, but the A1200 was not supposed to be the 'latest technology' either. It was an 'entry level' machine, the replacement for the A500.

However I can say that in 1992 'down under' 386SX's were selling like hot cakes to people who wanted the functionality of a 386 without the price. That is what it was designed for, and companies like Microsoft, Creative Labs, Dell, and Gateway agreed. The Multimedia PC standard released in 1991 specified a 16 MHz 386SX CPU with 2 MB RAM.

But the AGA machines were the latest and greatest and very best Commodore had to offer at the end of this year - don't you see the mismatch?

Not really. I never expected Commodore to have the 'latest and greatest' computer ever, and they never did. The A1000 had some great stuff in it, but by 1985 the 'latest and greatest' PC was in many ways more powerful.

In 1986 Compaq released their 32 bit Deskpro 386 running at 16MHz with up to 10MB RAM and 130 MB ESDI hard drive. In the graphics department IBM had their PGA card with 640x480 in 256 colors in 1984. It also had an onboard processor that did line drawing and area filling. Clearly this was not designed for games, but high-end CAD work - with a price to match. But if you were looking for the 'latest and greatest' in professional hardware the PC had it.

The Amiga was a budget machine for people who wanted better gaming hardware than what was in typical low-end 8 bit home computers. Its entire design revolved around this, including the floppy drive 'controller', shared memory and TV video standard. It was never about being the best possible.

By now you really should know that this argument does not hold any ground with me. I invested heavily in the Amiga platform and was (still am) the proud owner of an A3000 at this time.
And I continued to use this machine (with upgrades) exclusively until 2001.

Good for you. I also invested heavily in the Amiga platform, pre-ordering an A3000 and paying NZ$7200 for it. Over the years it got a bigger SCSI hard drive, an 060 CPU with 64MB RAM, an ethernet card and RTG card. This lasted me until 2003, more than a decade. Before that I had an A1000 upgraded with 2MB FastRAM and a 20MB hard drive. Later on I bought an A600 that I upgraded to a 33MHz 030, an A1200 upgraded to 50MHz 030, and a CD32.

However unlike you I didn't pine over what the Amiga could have been if Commodore had acted differently, rather I thanked them for what I got (it could have been so much less).

Yes: an increase of 400% in R&D for the Amiga sounds about right.
Maybe a little bit more, considering the meager budget it had in the beginning ...
As I mentioned before: after the purchase of Amiga it should have been "all hand on deck" at Commodore for the new platform.

Right. I think perhaps you don't appreciate the state Commodore was in at the time. They certainly pulled out all the stops to get the A1000 out, and continued to improve the Amiga as best they could. But they couldn't just drop all the other stuff which they had invested a lot in and were hoping for a return on (the C128 was introduced in 1985 and sold 5.7 million units by 1989, far more than the Amiga). We should also bear in mind that Commodore was not a monolithic organization. 'All hands' may not have helped that much.

I'm not arguing that Commodore couldn't have done better - of course they could. So could IBM, and Intel, and Microsoft, and... But this argument that the Amiga 1200 was screwed up by Commodore's incompetence or whatever is silly. Without Commodore it wouldn't exist at all! Other companies with purportedly better designed machines and more competent management did worse!

Look at Commodore's history from the beginning and you see them lurching from one disaster to another, taking uninformed risks and pushing stuff out despite dodgy engineering and yet managing to come out on top enough to keep going. It was messy, but this is how private enterprise delivers the progress and diversity we otherwise wouldn't get.

This is how we got the machines we love, that nobody would have planned out from the beginning. Jay Miner wanted a big boring box that copied the IBM PC. Gould hated that. Somehow it all came together to produce the A1200, a unique and wonderful design - not technically the best possible, but a realization of what the Amiga was originally intended to be - a great home computer that we still enjoy today.

[sandruzzo]

Just, with all chip sets at 32bit 14mhz and better ram timing we would talk another history. On Aga with blitter you can do a lot more, I know con I did a lot of tests.

[sandruzzo]

Originally Posted by Gorf

I can not argue with your personal perception.
"Results may vary".

Isnt' perception, its true tests. I'm talking about same blit size and planres. I'm able to blit more..

[Dunny]

I owned an A1000 until I upgraded to the 1200 in about '96? Basically everyone I knew had gone up to the 1200/4000 and I was stuck with a 1.2 machine, so when I had managed to cobble together the necessary cash I bought the pack that had Brian the Lion and Zool 2 in it.

God it was amazing. The sound was clearer, I had more RAM, it was faster for my AMOS stuff. The following year I added a 360MB HDD and then the year after that a Blizzard 1230/IV 50Mhz and that's how it was until I moved to PC a few years later.

Definitely not disappointed.

[Bruce Abbott]

Originally Posted by Promilus

386SX was upgradable the same way A600 is... I don't get it why Bruce seems to be oblivious to that. Those upgrades weren't all that popular in PC world (exactly because you could always just buy new and faster PC) but they were there.

This is ridiculous. The A600 was not supposed to be upgradable in the CPU department. It's only by fluke that the surface-mounted 68000 could be contacted by an upside-down PLCC socket (shaved to clear a capacitor on the motherboard). If Commodore had put the 68000 in a socket it would have been much easier, but they didn't because that reduces reliability and the A600 wasn't intended to have an upgradable CPU! (just like most 386SX motherboards weren't). I'm not going to fault them for that, I'm just happy we found a simple way around it.

Recently I bought (from the the UK) a 63C09 CPU to upgrade my Tandy Color Computer 3. Not looking forward to unsoldering the original 40 pin CPU, but that's what I will have to do because Tandy didn't put the chip in a socket. And why should they have? It already had the fastest 6809 that Motorola ever produced, and sockets are unreliable. I really shouldn't even do it, because it will just encourage me to develop software that doesn't work on a stock machine.

So why am I doing it? The thought of replacing the CPU with one that 'fixes' some of the inefficiencies of the 6809 is appealing, though not very useful. There are inefficiencies in the A1200 that bug people too, but we aren't stuck with a CPU that normally runs at only 0.89 MHz. To appreciate how good the A1200 is you have to go back to those under-powerd 8 bitters with horribly crude chipsets that caused us to pine for the Amiga.

[grond]

Originally Posted by Bruce Abbott

Once PC gamers got on that treadmill they never stopped.

You make it sound as if this was anything bad. It certainly was good for the industry and it would have been good for Commodore, if Amiga users had also got on the treadmill.

[BigD]

@Bruce Abbott

The C128 was introduced in 1985 and sold 5.7 million units by 1989, far more than the Amiga

Yes, and maybe the fact there were too many competing and incompatible models further eroded the ability of C= to make a success of the Amiga PLATFORM! The C128 was a dead end, no further C64 compatible products were planned and yet the chance to market and sell an ongoing Amiga platform was lost. It does not matter how many widgets are sold in the computer industry if you drop compatibility every one or possibly two iterations! You can get away with it with consoles but not if you are developing computers with expensive applications and hardware peripherals!

[Karlos]

Condemning the A1200 for being too little too late seems to be vogue but people are becoming completely unrealistic about how much more it should've been at the target peice point. Bruce is absolutely on the money about that.

[Promilus]

This is ridiculous. The A600 was not supposed to be upgradable in the CPU department

Neither were 386SX PCs.

It's only by fluke that the surface-mounted 68000 could be contacted by an upside-down PLCC socket

That's no fluke. That's creativity. And that kind of creativity lead to snap-in 386SX upgrades.

I'm not going to fault them for that, I'm just happy we found a simple way around it

Not so long ago your were talking about how amiga was so much better to expand than 386 going with a600+vampire as a prime example.

And why should they have

Well C64 still had many chips in sockets and why should they? Those weren't meant to be easily replaceable or upgradeable... silly them.

To appreciate how good the A1200 is you have to go back to those under-powerd 8 bitters with horribly crude chipsets

Well, yeah, but... should one had A500 first then spent his earnings on A1200, was forced to downgrade kick for some older titles and didn't receive any immediate reward by vastly improved aga titles... I can easily see that as leading to disappointment.

[sokolovic]

Originally Posted by Karlos

Condemning the A1200 for being too little too late seems to be vogue but people are becoming completely unrealistic about how much more it should've been at the target peice point. Bruce is absolutely on the money about that.

Agree with that.
Now people are blaming the stock A1200 (released at £399 in november 1992...) for not being able to run Doom smoothly even before the game was released.
That's becoming utterly ridiculous.

[grond]

Originally Posted by Bruce Abbott

Not really. I never expected Commodore to have the 'latest and greatest' computer ever, and they never did. The A1000 had some great stuff in it, but by 1985 the 'latest and greatest' PC was in many ways more powerful.

The A1000 wasn't the greatest technology but it had such a clever design that it achieved a much better overall user experience. Commodore surely couldn't have kept up with the pace of technology and most people here agree that the Playstation for low-cost gaming and the PC for high-end gaming and productivity would have killed the Amiga only a few years later than it actually happened (and most probably before any new iteration of the technology). It would have required the sort of miracle that saved Apple to make Commodore live to see the year 2000 (or actually an even bigger miracle).

What people criticise is the failure to do things that would have been possible even for Commodore and that would have made the A1200 or AGA quite a bit better than what they ended up being: chunky pixels in an 8 bit palette mode, 32 bit blitter, more sound channels, more modular computer (CPU off-board, fastmem configurations), perhaps DMA-driven IDE. Mid-priced computers more capable than the low-end computers they kept producing until 1992. These things would have been possible without requiring lots of investments, just a bit more forward-thinking (which for us is easy because we are doing all this in retrospect).

[dabba]

Originally Posted by Gorf

If the market share drops below a certain threshold, software developers will ignore your system, which will eventually lead to the death of your platform.

That is also the reason why the A1200 never really took off. Its market share was too low. The A1200's market share did justify specific AGA versions of Amiga games in 1993 and 1994. However a lot of AGA versions of Amiga games were de facto A500 version's code with the 256 color graphics from the PC version slapped onto them. The more powerful CPU of the A1200 was not used by these versions.

So in the end the A1200 could not always show its true potential (which was not overwhelming, to be fair). Most A1200 games were just the same in bit more colorful.

[grond]

Originally Posted by dabba

That is also the reason why the A1200 never really took off. Its market share was too low.

Every new technology starts with a market share of exactly 0. The A1200 didn't inspire people to believe its market share would become significant within a reasonable time. It basically was a facelift to an existing product.

[Karlos]

@quote

What people criticise is the failure to do things that would have been possible even for Commodore and that would have made the A1200 or AGA quite a bit better than what they ended up being: chunky pixels in an 8 bit palette mode, 32 bit blitter, more sound channels, more modular computer (CPU off-board, fastmem configurations), perhaps DMA-driven IDE

As much as I'd like to jump on the bandwagon, I don't think I can agree with this. One thing hindsight shows us is just what a bad shape commodore was already in by the time ECS was out and it wasn't getting any better. This approach you suggest would have to have been the roadmap as soon as the A500 was out and selling.

All of these "could haves" and "should haves" are somewhat moot by 1991-1992. Frankly, given hindsight, we were lucky to see a new chipset at all.

[ImmortalA1000]

Between the A1000 chipset Commodore bought off the shelf and the A1200/4000 chipset there was no improvement worth mentioning in that 7-8 year period. 7-8 years is Apple 1 and Commodore PET of 77 to Amiga 1000 working prototype!

Nobody told Gould to buy Amiga just to spite Jack. He bought it, had no idea what to do with it, they disbanded the Los Gatos engineering talent that designed it, who went on to design Lynx and 3DO chipsets and chose not to go with Ranger but the German A2000 turkey of a successor. Nobody told Gould to cut himself million dollar bonus cheques each year as Commodore was in deep trouble etc etc.

You can also blame Dave Needle who illegally sabotaged the working chipset designs Atari had contractually paid for before sending them to Atari. Imagine how screwed Commodore would have been if Dave Needle had done the right thing and sent Atari the correct schematics they had already signed a contract to do so, imagine if Irving Gould's Commodore had to fight Jack Tramiel with the same custom chip technology in their back pocket.

Still, even in 1992 and 1994 there are things you couldn't do with a PC at any price and at £399.99 respectively (1280x512xHAM8 static displays in 1992, Super Stardust needs a Pentium 133 in 1994 to do 50fps). But the sort of talent that made amazing OCS games that pushed the system to breaking point was also gone, Lotus is still the only zero frame drop racing game, Shadow of the Beast and Turrican III OCS games the best in their class, hell the shmup section of Turrican II is closer to console quality than all those pathetic AGA only shmups that looked like A500 graphics. You can't blame the hardware for lack of effort on the development side. Maybe only 2 arcade games really pushed the A1200 and CD32, poor showing for any machine. Can't think of many computers since 1977 that had less than 2 decent games justifying the expense of the hardware, not many at all, even in those very early days.

[grond]

Originally Posted by Karlos

This approach you suggest would have to have been the roadmap as soon as the A500 was out and selling.

All of these "could haves" and "should haves" are somewhat moot by 1991-1992. Frankly, given hindsight, we were lucky to see a new chipset at all.

But that is exactly how high-tech companies operate: once you have completed development of a product, you start work on the next generation. Sure, you also have a team that does product maintenance/perfection but you start work on the next gen the moment you fixed the worst bugs with the current generation.

Commodore dropped the ball long before work on ECS started. The moment OCS was in the A500/A2000, they should have had the team work on AGA or rather AGA+. Then they could have provided a meaningful upgrade to the chip architecture in time for 1990 or 1991. After all, what else can microchip developers do except develop new microchips? Or you can just fire them and wonder why you are left without a product a few years later.

[sokolovic]

Originally Posted by grond

Every new technology starts with a market share of exactly 0. The A1200 didn't inspire people to believe its market share would become significant within a reasonable time. It basically was a facelift to an existing product.

Again, an unfair and oriented trial. A1200, who didn't start with a market share of zero by the way, was quite well supported even AFTER the Commodore demise. Go tell to the wonderful, superior and impressive Falcon buyers (Falcon you probably didn't bought) that the A1200 support wasn't significant.