Is the Amiga architecture still relevant today?

[gizz72]

@stefcep2

It doesn't matter that the Amiga lacked development after '92 The topic of this thread askes if the amiga architecture is relevant today, what I (and others) point out is that the Amiga architecture wasn't even relevant in '92... As someone who was writing software for the amiga at the time, I remember the hoops one should have to jump through to do any that looked even remotely modern.

Yes the amiga has some cool hardware to do some specific effects, but this hardware was to compensate for the level of technology at the time... As chip technology improved the amiga architecture became a bottleneck, the Commodore engineers were working hard to rid themselves of it, and produce something at least comparable to off the shelf chips... But commodore folded.

Greetings,

...and 17 years later, we're still all discussing about it's relevance. A memory to some, yet it sticks like ROM. Oh, the Irony of it all! Any relevance of the Amiga's Technology today is all irrelevant. Without the support of the people who still uses it, is relevant. I say, use today's technology to develope a better or new Amiga architechture which takes advantage of faster machines of today. Like for example Windows 7.... yeah right! I mean emulation for backward compatibility, that's the reason why there's winuae, running on PSP, and so on.. Moving along... Sometimes it never hurts to look back once in a while.

Regards,

GiZz72

[Tension]

What Breathless had going for it was the fact it was one of the first, if not the first game of it's class to use C2P rather than chunky display emulation. In the long run, that gave it a significant advantage on faster machines.

So did it work better on a CD32 than an A1200?

Did Akiko work faster if you had an accelerator in the CD32? (ie would Breathless work better on a CD32/040 than an A1200/040?)

[Karlos]

Not sure if it used the akiko in any case.

The akiko, I believe, is able to take a block of 32 8-bit chunky pixels and spit them out as 32 words to separate bit planes at about the same speed as a 68040 is. By the time you get to 68040, the time it takes to physically push the data to chip ram has actually become the dominating factor. Of course, your 68040 could be doing something better with it's time, eg calculating the next frame.

It's probably fair to say that with an accelerator, the akiko is quite handy. Though it'll never be as quick and easy as just having a chunky display card.

[gertsy]

Touche @gizz72.  
I'm with you. Amiga's relevance is not in it's architecture. It is in it's indelible imprint on our psyche.

Gertsy

[gizz72]

Touche @gizz72.  
I'm with you. Amiga's relevance is not in it's architecture. It is in it's indelible imprint on our psyche.

Gertsy

@Gertsy
Exactly! Userbase makes or breaks an Architecture. That simply said.

[Hammer]

Hmm, not entirely sure, although I think it had 1Mb of graphics memory and was an ISA card.  I can definitely remember seeing dithering you wouldn't get with 24-bit colour.

At any rate I'd still say that, when it was released, a stock A1200 could better display still pictures (via HAM) than a PC with a mediocre graphics card.  Not that that's an immense achievement really - if anything it shows how much the Amiga's advantages over the PC had diminished by the time AGA came around.

i486 based PC should have come with VLB (VESA Local Bus)....

[Hammer]

Ofcourse the PC did. But answer this: what hardware development occurred, other than a CD ROM from the owners of Amiga after Commodore's demise?  Before Commodore died, PPC was where they were headed, new busses and new chipsets were planned and none of it ofcourse materialised..  What do you expect when all we got was bolt-ons to 7-10 year old architectures from third-parties.  A 7-year old planar chipset wasn't as fast as a new PCI/AGP card with built-in 3D functions and chunky screen-modes. Amiga had nothing as good as Direct 3D? What do you expect when the parent company has no in-house hardware and software developers, and it even takes several years for a relatively minor OS update to be written.

Its funny, as one of many active user at the time when all this was happening, we all knew the differences between a 486 and A1200 with an '030.  And we knew what was better and why.  Now we have people looking at the PC with rose-colored glasses, and telling us the Amiga would have been doomed even if Commodore wasn't.  Apple was shit-scared of Amiga, MS refused to write for it for the same reason. I know PC dealers today, who were also Amiga dealers, and unanimously they all agree that Amiga at its height was the superior home computing platform.  They tried telling Commodore's sales reps about Amiga's advantages of efficiency, multi-tasking, video capabilities, painting and animation in just 2-4 MB at time when the 486 needed 4 times that, when 32 meg simm cost over $1000, and a 1 gig HD was $1500, when PAL and NTSC output on a PC was an expensive waste of time, but no they wanted a games machine, they were never interested in pushing the productivity side.  Commodore, they say, didn't know what they had.

I think it was PA-RISC not PPC.

[Hammer]

Commodore's marketing was simple:
1.  A500, A600, A1200 cheapish games machine to play the 2D games that the were in vogue.
2. Anything with Zorro Slots: professional use, and 4 times the price for the privelage.

Not far off the Apple all-in-one Imac's Vs the expandable PPC towers that jobs introduced.  If only Apple had a multi-tasking OS to go with their PPC technology in the days of Win 95.

Apple would be in the same position i.e. it will lose against the cloned PC platform.

[Hammer]

I agree, though a straight quote from Darwin fits even better in my opinion. -"It is not the strongest of the species that survives, nor the most intelligent that survives. It is the one that is the most adaptable to change."

In other words the future looks bright for Linux...

Windows platform has ReactOS.

[the_leander]

So did it work better on a CD32 than an A1200?

Did Akiko work faster if you had an accelerator in the CD32? (ie would Breathless work better on a CD32/040 than an A1200/040?)

No it wouldn't - for one reason only: The CD32 only ever got an 030-50 accelerator, and they were as rare as hens teeth.  

One game it did make a difference on however was an Amiga only Bomberman clone that used c2p. I had an A1200 with an 040, and without using software c2p the game was unplayable, but the cd32, with its stock 020 (and an extra 8Mb ram) was able to play it without any such hacks.

[bloodline]

Akiko was a poor solution... Well, it was cheap anyway... A better solution would have be a nice chunky pixel mode added to Lisa... Perhaps only a single 800x600x8 mode but that would have kept the wolf from the door!

[Zac67]

Akiko's C2P converter just used some spare die area on a chip they needed anyway, probably it won the 'hey, what else could we do with xx spare transistors?' contest.

A somewhat decent solution would've been a converter sitting in Lisa's data path, converting the pixels on the fly. Next gen Lisa  could have included those functions. As mentioned above, it wouldn't have been too big a pain to have included chunky modes at least for AGA.

[bloodline]

Akiko's C2P converter just used some spare die area on a chip they needed anyway, probably it won the 'hey, what else could we do with xx spare transistors?' contest.

A somewhat decent solution would've been a converter sitting in Lisa's data path, converting the pixels on the fly. Next gen Lisa  could have included those functions. As mentioned above, it wouldn't have been too big a pain to have included chunky modes at least for AGA.

Thinking about it, a faster blitter and a separate framebuffer (physically separate from the chip ram) would also have been a good idea...