Hypothetical Amiga evolution...

[brett71]

I posted this over on the Natami forums, but thought I'd also bring it up here.

This was just something I was pondering the other day and curious to see what other people in the know would have to say about it...

Back in the day, we Amiga users always used to knock the PC about their commoditized parts versus the Amiga's custom-built chipset. (Well, at least I did.) The Amiga used custom chips for specific functions like video and sound, but nowadays PC's aren't really so different. True, at first video and sound were added to a PC through plug-in cards, but more so in the last 4 or 5 years, PC mainboards have video and sound and even network built-in. I was trying to think about where the Amiga might be today if Commodore-Amiga, in their final last gasp after the 1200 and 4000 were released, had chosen to take advantage of the commodity chips and parts from the PC world and abandon the custom chipset design.


For example, the next machine after the 1200/4000/CD32 could have looked like this in 1995 or 1996, and I'm referencing parts that were available around that time, but then continued to evolve over time until present day:


CPU: The machine would come with a 68060 stock, but instead of dropping the 68000 line for PowerPC, Motorola continued to evolve it, similar to the evolution of the Pentium processor.


Memory: the 4000 had standard SIMM slots, and this would continue to evolve along with the newer memory types that came out after 1995/1996.


Video: Replace the custom Angus/Denise and Alice/Lisa with an early 2D/3D chip like a Rendition V1000 or nVidia Riva128 chip.


Audio: Replace Paula with a Soundblaster 32.


Hard disk support: The 1200/4000 already had IDE, but assume that developments there mirrored the PC, including the eventual move to SATA. CD drive would be a standard component and support improved in AmigaOS.


Floppy disk support: Scrap the CIA chips used all the way back to the Amiga 1000 and adopt standard PC HD floppy drives. (I think I read that it was the CIA which was the bottleneck for floppy I/O which resulted in the specially-made half-speed Amiga HD floppy drives, correct me if I'm wrong.)


Expansion bus: Replace Zorro with PCI, which would likely result in AUTOCONFIG working more like the PC's Plug & Play (i.e. the driver for the device is no longer baked into the card and automatically loads itself into the OS, but instead needs a device driver being installed from disk), but would also open up the machine to utilizing network cards and other specialized expansion devices. Also eventually the machine would also include USB, Firewire, etc.


So, the main question given all this is: Would such a machine, if it could've even been constructed and made to function, have kept Commodore and the Amiga alive and a player in the consumer computer market? Or would this combination of hardware just never have worked because of incompatibilities? Or would it not have worked because Kickstart or AmigaOS couldn't have possibly supported it? Or would it not have worked for any other technical reasons?


I'm just curious about people's thoughts on if Commodore-Amiga had opted for going this route. I look forward to your responses.

[mc6809e]

Originally Posted by brett71

Video: Replace the custom Angus/Denise and Alice/Lisa with an early 2D/3D chip like a Rendition V1000 or nVidia Riva128 chip.


Audio: Replace Paula with a Soundblaster 32.

What a horrible suggestion.

Commodore had the basis of a modern GPU in the Amiga's blitter, and wavetable synthesis with Paula.

Had they simply continued development on more and more powerful versions of Agnus/Paula/Denise, they could have owned the graphics accelerator/audio synth market. They had a 7 year head start! There would have been no NVidia or Creative Labs and we'd all have Amiga hardware sitting in an PCI Express slot.

There might even have been no Microsoft. On our computer's POST screens, we'd instead see a message like "hit F12 for DOS or wait for Amiga OS to boot".

Commodore blew it big time.

[desiv]

My fear tho, is that Commodore would have continued that path.
If you do the math, you get a souped up 68k machine with PCI for video and sound cards..

It probably wouldn't be too long before the swap the NT over AmigaOS.
(Dave Haynie mentioned that, I think, in an interview somewhere as a possible future..)
And NT had HALs for other CPUs back then. Alpha...

Then, it's not too tough to swap the CPU and you've got a PC...

Maybe it's running Windows. Maybe it's running some theoretical AmigaOS for X86 variant...
But considering the work it would take to add memory protection to AmigaOS and the fact that Commodore was cheap... I don't see that happening...

To be honest, the best possible outcome would have been if someone else (No idea who) would have "saved" Amiga from Commodore (or bought Commodore) and hired back some of the original designers...
(Maybe adding Matt Dillon to add the memory protection. He had some interesting ideas back then. Now he's working on BSD.)

I don't see how it could have happened, but I think we got the best possible Amiga Commodore was able to deliver. It was going to go downhill, one way of the other..

desiv

[johnim]

havent we got that with blizzard ppc and a mediator system plus fast ata with sata converter or a scsi to ide/sata adapter


or any of the next gen systems eg x1000

addons are still being developed even tho not by amiga/commodore were getting a quicker machine from todays tech

[s2325]

As gaming machine it was perfect without any changes, I think fresh gamers were responsible for fall of Amiga, not hardware.

[lesta_smsc]

Originally Posted by s2325

As gaming machine it was perfect without any changes, I think fresh gamers were responsible for fall of Amiga, not hardware.

Agree - on basis of what was available 'at the time' - the Amiga was well ahead of the PC/MAC/Atari - unfortunately, it was not utilised to its full potential due to many areas of downfall - marketing, pirated games etc.

[mrbob2]

I think the simple answer is the amiga would simply have become a PC.

What made the Amiga apart from the OS is the components.

That leads me to the next part,

windows was/is developed by a multibillion $$$ company dedicated to the OS

Commodore made the OS and the machines with a budget of $100

All that would have happened is windows would have been better than any OS commodore would have made and the machine would just have become a windows device in the end.

its like saying what if the spectrum 48k had 2gb of ram and a nvidia GTX 280 gpu....

These machines had a finite lifespan from the start and nothing would have changed that

[brett71]

It's interesting to read your responses and views. I certainly did not want to start a flame war....not saying it's become one, don't get me wrong!

I suppose what I was trying to imagine, strictly from a technology point of view, is what Commodore-Amiga's last ditch effort to keep the Amiga relevent might have been, alongside PC and Mac, even if it was a distant #3. We all know that the Amiga's future was crippled by short-sighted marketing and horrible management, but if marketing finally "got it" and management straightened their act out (i.e. a "best-case scenario" for Commodore-Amiga), what would the Amiga after the 1200/4000 generation have looked like as far as specs go?

I suggested replacing the custom chips with Rendition or nVidia for graphics and Soundblaster for audio simply because I read that one of the biggest costs of developing and building the Amiga were the custom chips. So, I thought what if C-A took advantage of components available from other sources. One thing I did not take into account was changing out the custom chips for commodity chips would just about break every game or demo ever written up to that point, unless an abstraction or translation layer could be developed to make the commodity chips work like the custom ones.

[jimbob]

I guess something like you describe, replacing the custom chips with commodity hardware, could have happened but that would also have been the end of the road for classic amiga. It sure must have been costly to develop custom chips but I think commodore needed to be more ambitious in this regard. Then their next generation chipset could have become the commodity hardware instead of being overtaken whilst they stretched the ECS to the A600, underpowered AGA and cancelled AAA.

From a moores law perspective. How does it go, 18months=double performance for half the cost. So there were about 4.5 "moore periods" between the launch of the A1000 and the A600 yet they are more or less the same machine. Commodore should have seen that this was a failing strategy and made more advanced hardware themselves.

[stevsurv]

Interesting but... if the Amiga did go the way of integration of PC type components, e.g. plug-in graphic cards, sound cards, network cards etc, then it would of indeed been a kinda PC clone.

If let to evolve like this then it would of certainly become just another PC with the name of AMIGA. <--- Sound familiar?

At least when Commodore shut down their operations in the early 90's the Amiga was a unique computer that stood on its own. Historically it still does.

[Mequa]

Amiga Hombre chipset

This was originally designed for a new CD32-style games console, including full 3D acceleration, but may have found its way into a new A1200/A4000-like machine.

If Hombre had materalised after the shelved AAA chipset, it would have also meant the end of the "classic" Amiga line, because:

Originally Posted by Wikipedia.org

Hombre was designed as a clean break from traditional Amiga chipset architecture with no planar mode support. Commodore also decided to drop support of the original Amiga eight sprites because at the time sprites became less attractive to developers for its limitations compared to fast blitters.

So such a next-gen "Amiga" would have no compatibility with "classic" Amigas. It would have essentially been a replacement for Amiga.

[jimbob]

Interesting, those wiki articles on aaa and hombre suggest that commodore did have ambition to stay ahead with custom hardware but just couldn't get a complete system together in time to stay ahead of many others generally working on smaller but inter-compatible subsystems.

I was a bit surprised to read the note about hombre breaking with amiga OS as well as earlier hardware. Strikes me that a large back catalog of OS friendly software would be a big advantage for a new hardware platform.

I wonder if hombre had been successful and commodore still in the custom chip business, we would have seen maybe an optional extra custom chip, (codename Lorraine?) with 100% classic compatibility. Integrated minimig circa 1997.

Also interesting to think that if hombre had succeeded as a PCI multimedia card and the CD64 console, gaming hardware could be united across console and "PC" Anyway, it would be different if there was still another big player in desktop computers, (or not so different, "ha you fools running commodoze amigaO$, you should use Abuntu").