Posted by doener 2 days ago
Imagine if you had an Amiga Chipset and you had combined it with a RISC like chip. If you did that in late 70s with 3.5μm HMOS (like 68k). The resulting system would be insane, in terms of performance to cost. You could outperform minicomputers that cost 10-100x more.
The ARM2 like chip and the complete Amiga chipset seem to have less transistors then a single 68k, so the price off such a system would be very low. And we can see that with the Amiga, what really blows my mind is how cheap Amiga ended up being, an unbelievable achievment.
Its seem the issue really was the the companies that had the resources to do that amount of chip design knowlage and finances were not interest in making a home computers/workstation. Workstation ended up being made by startups who didn't have the resources to do so much costume work. Appollo was a split-off group DEC because DEC was not interested in workstations. IBM was just to slow and couldn't really do prodcut design, and we all know how the eventually got around that problem with the PC. Apple for the Mac did try to do one ambitious chip with VSLI but didn't end up using it.
The split between computer companies and chip design company was just to big to get the needed amount of integration, and there was clearly a lacking vision for what home computer could be. Jobs vision for the Macintosh went in the right direction, but really Jay Mine had the right vision, and he had it because he build a computer for himself. He wanted a home comptuer that was fast, had a proper operating system and enough media capability to run a flight simulator software. Sadly manamgent most of the time wanted him to develop a console and later when they allowed a home computer they didn't share his full vision.
But then also actually plulling this vision off, multi-chip costume design with very few resources is just an amazing achievment. And many of the people didn't even have that much knowlage in chip design, there was a lot of competition for chip design people. Getting into Commmodore where they had the actual semiconductor teams to get these designs over the line was lucky, many other companies who could have bought them might have messed this up.
In a perfect world you add ARM2-like RISC chip, a Sun-like costume MMU to something like Amiga Chipset and you move computing forward by 10+ years. In reality the exact opposite won, a 16-bit PC that had basically no costume design in it what so ever.
RISC and especially MMU with paging increase memory requirements. For comparison, the first Amiga was designed for 128K, got 256K. Linux/M68K with MMU on the Amiga required 4MB to be usable.
While on the other hand, RAM is easely equal to all other chips on the board combined.
So you need to desperatly do everything you can to limit RAM usage. That's why Amiga was smart to do shared RAM between the CPU and the Chipset. The Archimedes did the same thing as well. And ideally you get some SRAM on there not depend on the RAM to much, ARM3 chip did that and speed improved by a huge extent.
And you really need a good operating system that could give you advanced features without completely blowing up your RAM.
That is why it is sad that nobody with sufficent capital to pump into costume chip design took on a project like that, as it needs to be somebody with the volume the economics work out.
But nobody did and instead we got there threw an iterative process where PC manufactures pumped out PC and bought the cheapest chips, while Intel and chip companies integrated the board more and more. I remember learning about Southbridge and Northbridge, and that essentially the design you end up with.
The least they should have done was actually shipping Fast Ram controller (with unpopulated ram) in every single Amiga model since 1987 A500. Instead we got trapdoor expanding shared slow/chipset ram with very limited addressability.
Atari for what is worth quickly standardized on using SIMM sockets right on the motherboard and supported 4MB.
I think Amiga was fine at reading the room, at least the engineers. But the never did any signifciant chip develpement to improve things. Only the new chipset for the Amiga 1200 when it was already much to late.
https://forums.atariage.com/topic/207245-secret-atari-dram-r... https://web.archive.org/web/20180817061246/https://www.atari... TLDR up to $6Mil a month of smuggled ram from Japan by Atari to make money on a side :)
https://web.archive.org/web/20171220171525/https://www.atari...
"Source [an informant to the FBI who is apparently an Atari employee] remembers the first DRAM shipment arriving in 1988 was sold to Sun Micro Systems in Milpitas, CA. Source was told to deliver this shipment by [redacted]. [redacted] told Source to leave all [redacted] including his [redacted] and to take [redacted] which had an Atari logo on them. When Source delivered the Integrated Circuits, [redacted] cashier's check from Sun Microsystems to deliver to Atari, and the payee's name was left blank."
Considerably faster. I looked at both (and the ST) and bought an Archimedes.
ARM chips benchmarked from the ARM2 up to the RasPi 3B+:
https://stardot.org.uk/forums/viewtopic.php?t=20379
68000 benchmarks around that time:
http://www.faqs.org/faqs/motorola/68k-chips-faq/
ARM2:
Dhrystone/sec 5463
68000 @ 8MHz:
Dhrystones
68000 2100
MIPS
https://en.wikichip.org/wiki/acorn/microarchitectures/arm2
https://en.wikipedia.org/wiki/Instructions_per_second
ARM2: from 6 to 10 million instructions per second, depending on instruction mix
68000: 1.4 MIPS typical.
(For comparison: Intel 8086 at the same speed, something like 300 Whetstones, 0.5 MIPS. So either of them stomped all over a comparable x86 machine from that time.)
So, very roughly, ARM2 was between 2-3x faster in typical use.
Note:
. Neither CPU could do FP in hardware.
. Neither had cache memory.
. The Amiga had a lot of complex hardware acceleration for graphics; the original ARM2 machines from Acorn (Archimedes A2305, A310, A400) had essentially none.
So, Amiga games could do things that on the Arc required raw CPU, typically done careful hand-coded assembler.
To improve performance you need to start to add cache in or around your CPU/MMU and you need an increasingly clever memory chip to arbitrate memory access.
But you are right, you always going to run into limitations. As RAM is the most expensive part of the BOM, literally anything you can do to get maximum utilily of out of the memory bandwith is where you need to spend design time.
Like ARM did a 16 bit encoding for your 32 bit RISC would have been a great design feature to further reduce the how intensive it is to keep the CPU active. Of course nobody had thought of that combination, that was an early 90s thing.
RAM is just so expensive back then, you build your machine around sharing it as well as you can and living with the downsides.
There was never any ARM-based Amiga from Commodore or any Commodore partner.
Any CPU performance is 100% theoretical because there was no such hardware.
The 68000 performance numbers I cited are from contemporary benchmarks and they _favour_ the 68K. The real chip in real Amigas ran slower.
The Acorn Archimedes used the ARM2; the ARM was developed for the Archimedes range. Its display, sound, memory controller, etc. are all pretty much unaccelerated.
Now there are Arm-based Amigas but they run the Amiberry emulator on top of Linux.
The whole idea of ARM2 (that specific version, not ARM1, not ARM3) was that clever use of the RAM bus meant it could run at full DRAM speed.
But the other side of the coin comparing with the Amiga is that the whole idea of the Amiga was that ~5 years earlier, it was possible to do amazing unprecedented general stuff for gaming purposes in sound and video chipsets that no other computer at the time could do...
Leaving the CPU almost irrelevant. Which is in part why The Next Amiga, the Hombre, simply discarded the 68000 altogether and switched to PA-RISC.
Which infuriated the fans even back then because they wanted their backwards compatibility.
The core concept of the Amiga and Hombre is "in 1982, what amazing non-dedicated sound and video can we do in hardware for a mass market price"?
By the time it was a product, this was contaminated with "it's also a general-purpose computer" and "look at us, we have multitasking" and "look, it's a 680x0 machine, but with multimedia!" and "look, it can emulate a Mac!" and "look, it can run Unix!"
Different selling points selling in different directions.
Given the raw computer power of the Archimedes, it didn't need any fancy chipsets. They became irrelevant. Sure, yes, the Amiga could do stereo digital sound and multiple bit plane colour 2D graphics and still have enough power left over from a 7-and-a-bit-MHz 68K to run game logic.
The Archimedes looked at that, and went "how about we just do all that in software?" And bundled a solid shaded 3D demo for free. Never mind the bouncing ball, never mind the Juggler, this you could play.
The Mac was "in 1984, how much Lisa can we keep for how little money?"
The ST was "in 1985, how cheap can we make something that looks and works like a Mac but can do colour and sound, so it can play games?"
The Archimedes was "in 1986, can we do a 32-bit RISC machine so cheap and simple we can compete in the same market with Commodore, Atari and the PC compatibles?"
You are not saying "there was no uncontended RAM in the Amiga", which is what I understood you to mean. You are saying "the Amiga did have uncontended RAM but it did not ship with any as stock."
Is that right?
If so, yes, true, but the key things here IMHO are:
1. The Amiga was designed and built as a games console, one which happened to be able to do other things. It was primarily marketed and sold as a games machine.
I wrote about an Amiga-compatible OS recently:
https://www.theregister.com/2025/05/22/aros_live/
And the comments are from slightly baffled readers asking if this is any good for running their old Amiga games, or how this makes gaming any easier.
(It is not and it does not, but that is what people perceive the Amiga as being for in the 21st century.)
2. In the real 1980s market, the Amiga's glory was to a large extent stolen by the Atari ST. The ST was a much more limited machine but it was vastly better than any 8-bit machine, and an entry-level ST was more usable than an entry-level Amiga.
ST games were pretty good for the mid-1980s when a lot of people still had ZX Spectrum or Commodore 64 level kit.
Many games companies targeted the lower-end machine and ported to the higher-end one.
The Amiga was competing with the cheaper, simpler ST in the market, and keeping the costs down became imperative. That's why Commodore didn't add more hardware or more RAM to the base-level Amiga. In fact for the Amiga 600 took the spec of the A500 and cut it down.
I am not arguing that some Fast RAM as stock wouldn't have been good. It would. But probably irrelevant to most gamers, and probably would have hurt the machine's sales.
Or much better, the Agnus chip could have had this scratchpad added into it. It should have been feasible - it had about 20k transistors IIRC and a few hundred more should have been doable. I am fairly certain this would have worked but I'm not completely sure, it depends on if that addition would have complicated multiplexing and/or internal bandwidth demands, but the 68000 ran at 7MHz at the time, so it doesn't seem too difficult to me, armchair designer.
Thanks for the article! It's always nice to see AROS and Amiga get some love. :)
Achschully AROS m68k can be helpful for playing your old Amiga games in an otherwise open source way. You don't have to buy or pirate the official Kickstar ROM images if you use an AROS m68k ROM instead. But on the other hand, realistically you would probably pirate that old Amiga game to begin with, it's not like you have your old floppies lying around and if you did, you'd need a floppy drive to read them anyways...
Dieshot of Agnus:
https://upload.wikimedia.org/wikipedia/commons/9/9f/CBM_8370...
https://www.mobygames.com/game/attribute:268/include_dlc:fal...
On the other hand braindead lack of planning making early Amigas 1MB limited resulted in less than ~10 games ever using more than 1MB. Rare exceptions are for example Wing Commander loading some additional animations with >1MB available. There were also games that hardcoded check for exactly _1MB_ and refused to run with more :)
Yes Amiga was a gaming machine first. Chipset independent scratch SRAM would do wonders for games _IF_ it was introduced together with first A500 like trapdoor ram was.
Running code from 7MHz fast ram bumps amiga from 0,57 to 0,75 Mips, 30% speed bump.
Another Amiga missed opportunity was clocking CPU faster when its not accessing chipset/chip ram. This is possible at the staggering cost of one D flipflop and was figured out in 1989 https://aminet.net/package/docs/hard/14MhzA500 http://amigaga.chez-alice.fr/classic/bidouilles/hack/overclo... Works because 68000 is a horrible CPU with very long cycle count instructions.
Why bother? The 14MHz mod alone gives a diminutive but _free_ bump from 0,57 to 0,62 Mips, but marrying together 14MHz and running from SRAM fast ram bumps us into 1,51 Mips https://www.youtube.com/watch?v=2nlG8dGvq-U thats 20% faster than 5 years older Amiga 1200.
Even 8KB of dedicated fast ram would be ideal for small fast loops, and there is plenty of those in games.
Lets not mention Commodore being so incompetent they couldnt find/wouldnt pay for an ASIC designer to update Paula PLL in order to support HD floppies :| Plenty of low hanging fruits nobody at Commodore bothered or knew how to make happen.
But... how is that possible? The Amiga 1200 has a 32 bit wide bus and is also 14MHz?! Commodore did the stupid thing again and didn't put any fast-RAM on the machine.
* The Amiga 500 is basically a cost optimized Amiga 1000, released in 1985.
Hell, smart Commodore would design A500 with second trapdoor near the CPU. Ship unpopulated but offer official Commodore "turbo ram" expansion carts:
- $25 4KB version. $8 BOM, 2x $1.95 6116
- $40 8KB. $12 BOM, 2x $3.5 6264
- $100 64KB. $28 BOM, 2x $12 62256
- Extreme $400 256KB. $120 BOM, 8x $12 62256. Would stick out due to big PCB so make it an attractive piece of plastic with cool logo and "EXTREME" design.
Map it at D80000 (potential 256KB of space for activities) and let software vendors auto detect fitted option by running quick memory test. Easy speed boost and easy extra money for Commodore.
https://news.ycombinator.com/item?id=45745754
But I think in reality, less money for Commodore because of sales lost to the now-even-cheaper Atari ST.
Because that was basically the spirit behind them.
Also note that one of the original ideas for the Amiga, was to get into the UNIX workstation market, then ended up pivoting into a multimedia machine (thankfully) to what we now know as the Amiga.
> Also note that one of the original ideas for the Amiga, was to get into the UNIX workstation market
Could you source that for me? It might be true. I heard from interviews with Jay Minor that they were developing a game console, and then when the 1982 crash happened, he convinced them to pivot to a home computer. And specifically he wanted it to run games like flight simulation.
Commodore explicitly cancled its own Amiga workstation, the Z8000 based Commodore 900 Unix workstation.
When Amiga lauched they lauched with games and with pixel editors. It was endorsed by EA.
Doesn't sound like 'unix workstation' to me. It was later in the 1980s where they started to want to get into that market.
PS: BeBox much later, by then the PC ecosystem had already done most of that stuff.
This was the plan for the successor machine, codenamed HOMBRE and never released.
An Amiga-like chipset closely coupled with an HP PA-RISC CPU.
https://en.wikipedia.org/wiki/Amiga_Hombre_chipset
A little more info in German:
I always thought PA-RISC was a great RISC and it an interesting concept.
But even in that case that wouldn't have been the CPU. Commodore never dared to do that after the 6502 (of course that team left Commodore).
Commodore was buying 68000 CPUs from Motorola at $2 a pop thanks to owning a fab and knowing how much it cost to make such chip in house and using that knowledge in negotiations.
https://www.amiga-news.de/pics/L/lorraine/FB_IMG_17610203985...
Designing the chips in MSI, wire wrapping a prototype, and getting it all to work to the point where you get a functional chip run is the stuff of legends.
And we're obviously in the less cool branch. :-)
