Dave Haynie Talks About Developing The Commodore Amiga

[psxphill]

I also think it was the wrong path but that is easier to say now that PPC is becoming less competitive and new AmigaOS 4 hardware costs thousands of dollars and sells in the hundreds of units.

Custom hardware in small units is always going to be less competitive. If they were selling millions of units then PPC could be competitive, it worked for the PS3/360. The only way to be as cost effective as a PC is to use an off the shelf PC motherboard.
 
I think Dave didn't mention post commodore because he went off and lived his life and did other non-commodore things.
 
The talk was interesting, but like most of these talks a lot of the information has been talked about before.

[psxphill]

But the C= management would cancel any project that looked even slightly expensive...

They didn't cancel AAA fast enough. The way I see it please like Dave were supposed to be the caretakers who kept the old Amiga going until the "real heroes" delivered the AAA chipset. Top management wouldn't want to waste money on the old technology with AAA coming.
 
When you have someone at the top that doesn't understand the technology and yes men at every level down then things go wrong real fast.
 
Someone from engineering should have foreseen that AAA was going to be a disaster from day one.

[psxphill]

AA was waaaaaay late and so was AAA/Nyx.

I read that AA was finished earlier than expected, IIRC the project was started in late 1990 and the first working silicon was in early 1991.
 

If engineering had it's way (and the resources) the A3000 would have gotten AA from the start (or shortly after), about 2 years before it went into the market with the A4000.

When the A3000 was started AA didn't even exist on paper, in the talk Dave says they couldn't get any custom chip changes due to all the chip guys working on AAA. The A3000 was purely supposed to get money in to pay for AAA development & it couldn't be too good or it would compete with their coming soon AAA machines.
 
AA could probably have made it out a year earlier if the A3000+ had not been canned. But we're talking about late 1991 instead of 1992. That does seem to have been a management issue, however company politics are never simple and it wouldn't surprise me if there were people in engineering that were throwing in hand grenades because they didn't want AAA resources to be cut.
 
We know that engineering wasn't cohesive. The C65 project was one engineers dream to build the ultimate 8 bit computer, because they didn't like the Amiga. There were a few people who did all the stuff that got sold and there were a ton of others who spent more time creating things that were never going to happen. The view that management sucked and engineering were perfect is overly simplistic and not true.

[psxphill]

If you watch the Video, Dave says the AA (as it became known) was just a side project to extend the life of the Original Chipset (or more technically the "Enhanced Chipset"). they called it AA because that was one "A" less than AAA

The side project was only started years after AAA because they realised it was several years off (if ever) and they needed something to compete. The A500 had gone through it's most successful sales period by then and they couldn't rely on momentum any longer.
 
AA was called Pandora during development. I haven't read whether that name came from a person or the greek myth.
 
I don't know how long the AA was around for, it changed because of existing usage. Both AGA and AAA suffered from this though.

[psxphill]

Actually the processor being developed for Hombre was chosen because, though it didn't have 68k emulation modes like the PowerPC, its instructions were close enough to 68k to make porting easier.

The PowerPC didn't have any 68k emulation modes, it could run in big endian mode though.
 
They chose the HP RISC because they could buy modify the core and Ed Hepler was big on that type of design. It's the same reason the Arm took off.
 

The Amiga section was making sales. Commodore was losing money because they kept trying to compete in the PC sector and failing.

Not really. It was the parent company that went bankrupt first. The local sales companies only went under when they couldn't get any stock in. It was commodore Germany that was building the pc's.
 
The parent company had a couple of problems with lawsuits and tax bills and no cash flow. The cd32 that they had made weren't allowed to be imported until they paid their tax bills, they couldn't pay their tax bill because without the cd32s they couldn't make any money.

[psxphill]

The point of having "managment" is to weed out brainfart projects (C65,A600,Plus4...) and steer resources towards (potential) money cows (like AAA-Amigas).

The A600 and Plus 4 started out as good projects.
 
The Plus 4 started out as the C116 & C264, which would have been really cheap and sold a lot. I believe marketing killed that by heaping stuff on it.
 
The A600 started out as the A300, which would have been really cheap and sold a lot. Bil Sydnes was responsible for heaping stuff on it.
 
AAA however was not really the type of project that commodore ever did well, it was never going to work.
 
The C65 probably shouldn't have ever existed, but commodore were making a load of money by selling C64's so I can see how it kept going. I imagine it got killed when they admitted it wasn't really C64 compatible at all and so it would be hard to sell it.

[psxphill]

C=/Amiga in the early 90s did not only have good enough HW, but also a large userbase, big SW-library and all that running an OS that was superior to the mainstream of that time (Windows3.1/95,MacOS7-8).

Amiga was powerful hardware for the price in 1985, but by 1990 things had changed. By 1992 the consoles were beating it for games at the cheap end of the market and the pc was beating it for games and applications at the high end.
 
Windows 95 was ok compared to AmigaOS, it had a lot of things that we were missing.

[psxphill]

Not a developer so not sure how it does it. Just that when it was finally done, PowerPC could run 680x0 code reasonably.

The emulator that Apple wrote for the PowerMac to run 68000 code was just really good.
 

they should have made C128 with a 65816. Yes, it would be close to Amiga in some regards, but it would be able to tap into C64 market and software library, something that the A1000 couldn't and why it took a few years untill A500 for Amiga to take off.

The C64 was mostly selling in Europe to run games that mostly required cycle accuracy and illegal opcodes. A 65816 wouldn't be very useful, even the C128 wasn't compatible enough & it was expensive.
 
The A1000 was also expensive and there wasn't a lot of software. Once the A500 hit and European games developers started pumping out games then things improved.
 
The A500 was another project that was basically one persons dream, there were other people in engineering who weren't coming up with great ideas. It's unfortunate that as part of fat agnus project they didn't manage to increase the bandwidth and squeeze some more colours in. The only improvement other than cost was 512k of slow ram, if they'd gone straight to 1mb chip ram it would have been better.

[psxphill]

Allmost killed Apple and it took them to the release of the 1st PCI based Macs over a year later till they added some JIT-like speedups (and having more of the OS in actual PPC code) to really regain a level of performance comparable to the x86s.

Yeah, they hired someone in with experience to do the JIT IIRC.

[psxphill]

I wouldnt say it was "too late" already by 90-92. At that time I had Amiga only while a few friends got the PC, and it was much more userfriendly than what my buddies had.

By 1991 Id had started writing wolfenstein 3d, if he'd written it for the Amiga then things would have been different. The momentum that game created was enough to kill the Amiga.
 
The Amiga would have needed chunky 256 colours in 1990 to have built any kind of market lead. If it weren't for AAA then I believe that something like AGA but with chunky graphics could have been shipped in 1990. Even if 256 colour mode was low res only and lacked smooth horizontal scrolling (depending on how you could graft chunky pixel fetching on to the bitplane fetching might upset the horizontal scrolling).

[psxphill]

Both C= and Amiga were big sellers in Europe.
65816 was compatible enough for Apple.

Apple II didn't have VIC and SID with all the cycle accurate requirements that go with it.
 

a investment in evolving than Amiga platform would have made more sense.

Unfortunately their investment was AAA. The engineers that rode that train ought to be ashamed of themselves.