To be an Amiga fan during the dying days of the hardware platform back in the mid 1990s was to have a bleak existence indeed. Commodore had squandered what was to us the best computer ever with dismal marketing and a series of machines that were essentially just repackaged versions of the original. Where was a PCI Amiga with fast processors, we cried!
Now, thirty years too late, here’s [Jason Neus] with just the machine we wanted, in the shape of an ATX form factor Amiga motherboard with those all-important PCI slots and USB for keyboard and mouse.
What would have been unthinkable in the ’90s comes courtesy of an original or ECS Amiga chipset for the Amiga functions, and an FPGA and microcontroller for PCI and USB respectively. Meanwhile there’s also a PC floppy drive controller, based on work from [Ian Steadman]. The processor and RAM lives on a daughter card, and both 68040 and 68060 processors are supported.
Here in 2024 of course this is still a 1990s spec board, and misty-eyed speculation about what might have happened aside, it’s unlikely to become your daily driver. But that may not be the point, instead we should evaluate it for what it is. Implementing a PCI bus, even a 1990s one, is not without its challenges, and we’re impressed with the achievement.
If you’re interested in Amiga post-mortems, here’s a slightly different take.
If not for Commodore’s misManagement the Amiga would have taken the Apples place in history.
There is always one that says that, but then again, apple proved to be the better company by simply surviving.
Kind of funny by buying one failure (NeXTStep and Jobs) they did.
There is always one that forgets history. Apple failed its way to success.
Apple was floundering too when Commodore closed shop. They lost money in 1994, 95, 96, and 97.
Their OS was trash, They lacked time and talent to fix it.
Their business skills were so poor their CEO lost the deal with BeOS, Paid even MORE to acquire NeXT. Then despite wanting the deal to be cash, CEO Amelio pushed Jobs into taking 1.5 million shares, Which Jobs sold immediately, dunking the stock price AGAIN, which ousted Amelio and put Jobs in the big chair of a nearly rotten apple.
Then Gates pulled them out of impending bankruptcy to avoid Microsoft being declared a monopoly. Gates investment put Microsoft and Apple in bed together shutting down the two companies legal squabbles and giving Apple the time it needed to branch out into ipods and then Iphones.
If Gates had dumped cash into commodore to cover Microsoft with the FTC, or if the BeOS deal went through. We would likely be talking about Apple like Amiga, or lining up for new phones at the NeXT store.
Let’s quote Andrea Campanella:
“apple proved to be the better company by simply surviving.”
(And Microsoft had reasons to support Apple over Commodore; the link with Office.)
I personally don’t think the BeOS option was ever really going to happen. I think it served as a due diligence shopping around. With respect to BeOS, it was not nearly as mature as NeXTStep.
Though that may be true, Windows also wasn’t very mature at that time. I’d say BeOS was at least on par with Windows.
IIRC Apple made Be an offer, their counter was so insane, Apple simply bought Next.
Which was better in every way and included a talented bosshole.
At the time Windows was NT 4.0, which was a rock.
IIRC I was running a dual Pentium pro.
Better than 10 at everything but gaming.
Perhaps a slight overstatement.
Tools are better multithreaded now.
MacOS at the time was barely getting emulated 68K code out of the file system and network stack. Pre-emptive multitasking and memory management were dreams and had been dismissed as useless by the ‘chosen’.
PPC guys said ‘But look at the integer math benchmarks, that’s what matters!’ until the next gen then suddenly ‘look at the float benchmarks, that’s what matters!’.
While BeOS may not have been as mature as NeXTstep and Openstep. Both were far better than MacOS or DOS/Windows at the time. It wasn’t until NT5, or OSX that either were acceptable.
– there was no BeOS deal, there was just Jean-Louis Gassee dream of becoming Apple next CEO.
– With NeXT they got great OS and great CEO.
– STOP with that “Gates saved Apple” fairytale. Jobs forced Gates into settling a pile of lawsuits. Microsoft was caught stealing Apple source code and shipping it as part of Video for Windows, it was open and shut case that would probably lead to MS being broken up a year later when DOJ came pouncing down https://en.wikipedia.org/wiki/San_Francisco_Canyon_Company
The NeXT deal didnt include Jobs as a CEO, Jobs forced Amelio out AFTER the deal by dumping the stock Amelio pushed him to take in the deal. That was a coup not a negotiation. That was ALL Jobs, when Jobs hadnt been Apple for more than a decade.
All youve done with throwing the canyon company suit in the mix is illustrate how incompetent Apple was at business before Jobs took over again. Had Apple folded Microsoft would have been broken up as a monopoly. Gates and Jobs hammering out a settlement which MADE MICROSOFT BANK, saved Apple from Bankruptcy, and turned the Monopoly busting threat to an antitrust sanction.
It was a WIN for microsoft and a second chance for apple, Like I said.
If Gates had dumped cash into commodore to cover Microsoft with the FTC, or if the BeOS deal went through. We would likely be talking about Apple like Amiga, or lining up for new phones at the NeXT store.
Enjoy your own fanboi fairytale interpretation of history
>If Gates had dumped cash into commodore
What did Commodore had on offer at that point? GEOS?
Hahahhahah… hahhahahaahahahhaaha….. hahahahahaha.
I can’t stop laughing at the naive stupidity of someone who thinks that if Microsoft had given $100,000 to Commodore, that Irving Gould wouldn’t have extracted that cash nd pocketed it immediately.
Commodore died because the board and management started looting the company in its final years. Apple was lucky enough to have a board of directors and CEO (Amelio) who were determined to do everything possible keep it alive.
Commodore let the Amiga languish, barely investing the profits rolling in on improving the platform after acquiring it, instead they tried to milk it like their Commodore 64 cash cow, which was profitable for a decade with essentially zero investment.
Ars Technica has a fairly good rundown on what killed the Amiga, also David Pleasance is a great source of detailed and honest appraisal of Commodore during that time.
Pretty much. They had the worst CEO that cut to add profit to the balance sheet rather than invest in the technology and brand. The PlayStation showed up how far behind the amiga had fallen.
Apple survived by infecting schools with useless machines in the 80s and 90s.
Apple’s Survival can probably be attributed to a few products:
– The iMac starting in ’98, which injected enough cash to enable it to remain solvent for a few years.
– The iPod, which was hugely successful.
– The iPhone
The thing that probably save Apple’s computer business more than anything else was the popularity of the Web in the late 90s. Software was extremely platform biased, but the web didn’t care what kind of computer you were on. The iMac was visually attractive, and made it easy to get online. It was non-threatening.
I don’t think Apple was making enough money selling Performas to schools to survive.
It’s not always survival of the fittest, sometimes it’s survival of the luckiest, see mammals vs. dinosaurs 65 million years ago. (And yes I know about burds.)
The Atari and Commodore 8 and 16 bit machines were always smeared and hampered by the “it’s just a GAME machine” garbage and once the expansion slotted IBM PC clones began to get highly capable sound and graphics cards, it was all over for both Atari and Commodore.
The most sensible thing Commodore could have done, would have been to purchase Newtek and the Video Toaster, immediately develop a PAL format version, and offer a low-cost, very capable digital video effects system where systems like Quantel Paintbox cost hundreds of thousands of dollars, and then move into video editing as well.
I don’t want to think of the software that makes it all work together and still be called an “Amiga” rather than just another weird Linux computer.
There are already PCI busboard for some Amiga models, and drivers exist for some (admittedly, now vintage themselves) PCI cards such as graphics, sound and USB using the same frameworks as native Amiga cards. This project incorporates the hardware expansion into a standard sized mainboard, but still runs Amiga OS and still has the custom chipset so should be as much an Amiga as any other “recreation”.
Does it run Amiga OS with the original chipset AND all the additions, or is it like many of the expansion boards where you basically use the underlying Amiga as an expensive motherboard to bootstrap an entirely different system?
with this one weird trick…
Amiga didn’t have PCI because its Zorro bus using the Autoconfig protocol was way superior to PCI. Imagine having every add on board self configure itself (addresses, IRQ lines, everything) without jumpers almost a decade before PCI was introduced.. Adding PCI was made ossible through so called bridgeboards that allowed the use for example of PC graphics cards along with other Amiga (thus Zorro) cards. Unfortunately Commodore management did their best to destroy any potential non gaming use of the Amiga, and the rest is history.
I highly recommend “The Deathbed Vigil and other tales of digital angst” video by Dave Haynie, a then very well known Amiga engineer, showing the last days at the Commodore HQ in west Chester. Youtube has it.
Zorro is in no way superior to PCI. It may be earlier, it supported autoconfig, but superior to PCI? forget it.
PCI had approximately the same performance and was fully autoconfigured, supported 4GB, and multiple base address registers per card.
In what way do you find Zorro III superior to PCI?
Zorro III came 3 full years before PCI ad its *theoretical* speed still being marginally better. Despite being a one-man design was far ahead of its time and surely a great achievement.
Given that, here’s what its creator (Dave Hanye) said:
“There’s nothing, nothing at all, that Zorro III will do better than PCI. Take this from
THE Zorro III guy — I did every damn bit of Zorro III at Commodore, with no help”
The real advantage for Amiga users was the ZII backward compatibility and it’s why it has been kept on A4000.
This ^
Here’s some emails where Dave explains why PCI really is better
http://amiga.czex.com/local1998/7_9Z3vsPCI.html
The asynchronous nature of Zorro III is really painful and hurts performance. All control signals need synchronisation which really slows things down.
It has a “burst mode” which only works when your CPU is a 68030 and is impossible to actually implement on any synchronous designs due to the timing requirements.
Having designed a Zorro III board the whole thing makes me angry
Yep. A friend of mine used to make fun of the PC world’s configuration hassle by calling it “shrug and pray” in reference to the Amiga’s plug and play.
There isnt really anything breathtakingly great about Zorro Autoconfig compared to ISA Plug&Play kludge, and definitely not superior to PCI :-)
Oh wow, very cool! Makes a lot more sense (well, to me) than all the 1:1 PCBs for the different machines out there!
I wonder why it’s an OCS/ECS Version. If you go through that huge amount of work, wouldn’t an AGA version be not much more work?
Is it the availability of the AGA Chipset?
Ocs/ecs is socketed while aga is surface mount soldered.
What makes an Amiga an Amiga is already present in OCS and “just” enhanced in AGA.
Also the Amiga 500 sold well so there are alot of OCS/ECS chipsets available.
AGA chipset with exception of Gary are PLCC and can be socketed. The fact that they were soldered doesn’t really matter.
If they use an FPGA anyway, an AGA implementation should be possible.
But I guess if you have a motherboard with PCI you’d rather have a PCI graphics card with 3D acceleration and the AGA doesn’t have that huge of a software library.
I to wish this was AGA, just for the fact that it would be in line with the last iteration of the Amiga and support a all the AGA only games. But TBH the “AGA only” software library is quite small compared to OCS/ECS.
But with respect to the PCI bus, while it does open the door for a lot of cheap cards, there is not a lot you can get out of PCI that you cant achieve better with something like a pistorm. I desperately wanted PCI (for cheap, affordable sound and video cards) BACK IN THE DAY.. but now, its just a nostalgia item to lust after but which wont add much real benefit to my retro computing hobby.
I think a scan doubler, USB and networking are the only real “gotta have” items these days.
I wanna make clear that I respect the effort here though. You have to start somewhere, and OCS is the easiest Amiga architecture to implement. Even so, a full new Amiga design from scratch is a huge undertaking. Massive props to anyone who does this and then offers it up to the world as open source. And I still certainly want to build one ;)
As a teenager I could only dream of an Amiga (I had made the restrospective mistake of buying an Atari 1040 ST ;) ).
For my 30th birthday I bought a big A4000 T (Tower) as a dream-come-true, but it needed quite a bit of work to be bootable, and has been collecting dust in my attic since.
And, recently, at 52, I finally bought the Amiga Forever bundle, played a bit with the games and apps of my youth, and did not open it since.
It was a fantastic time, great memories of demo parties and whole nights spent playing Pinball Dreams with friends who had an Amiga, but time has gone.
I can understand that people want to keep this machine somewhat alive, but I must admit I like today’s computers (I work in software engineering).
Fellow colleague – I am a full time software developer (95% development, 5% coding at max) and I was an Amiga software developer in the 80s/90s myself. I loved the times. I am still fond of the fun, the experience, the learning, the challenges and some of the people (I knew some famous Amiga people I would rather forget to have ever met).
Those times are gone for good. We work differently today. We approach problems differently. We have other fish to fry and we can still have fun. Amiga was great, but it is not any more. Not even its OS was THAT good that we still need to have wet dreams about it.
I think all the retro computers and software have lessons to teach us even in these “modern” times.
If we could only re-do history where the best ideas in computing (and not marketing, or stifling thereof) won.
All our 8 bit computers would have a 6502 and a z80 inside them and be able to run each other’s software, as well CP/M applications like WordStar, SuperCalc and Dbase.
The PC done right would have been based on the 68000 CPU meaning no 640k limit, and the Amiga and Mac would have been like super PCs, able to run PC software but also with better graphics and sound and with the ability to run CP/M 68k and GEM 3.1 PC applications in a Window running at full speed without a bridgeboard.
Microsoft would be a small provider of old programming languages who were purchased by Borland in 1986, and Bill Gates would take accept a management job a Digital Research from his multimillionaire friend Gary Kildall.
Noob here born in 1985.
Self-taught in a lot of things and as someone who was intoended and all that other stuff as a outlier of all this stuff, amigos, known for its tracker by basic people and looking at it. Any system is only unique by its bottle next in the way it processes. If you have 10 computers I can do the same thing. Each computer. There’s gonna feel a little bit different because the way it processes and how it handles the electricity and data and software and blah blah. So if people want that feeling then yeah, go get the hardware but at the end of the day, the reason why that hardware was even special or important people was because of things like the tracker. And if you can just rebuild a tracker software wise on another piece of hardware than it doesn’t matter again, it’s very cool that someone is learning the skills to do this. But at the end of the day let’s get trackers into as many hands as possible
Motorola ’40 and ’60 chips must be hard to find these days. Even fast versions. So can this “new” improved Amiga accept the “Raspberry PI” driven CPU replacements work ? Since the CPU runs on a daughter board plugin. A raspberry pi only alternate daughter board (emulating Motorola) would be something.
the cpu is just another slot interface by the look of it so yes.
There’s always the FPGA “68080”, though I don’t think (sadly) that it is open source. That said, at this point, someone else could create one. http://www.apollo-core.com/
https://github.com/captain-amygdala/pistorm – PiStorm is where its at. The “problem” with these Pi based ones is they kind of don’t need any Bus support as it uses all the features of the Pi to supply Video, Storage, Networking and HID.
If you are going for a PiStorm based CPU it is unlikely that any of the supported PCI cards will give you much benefit so why not use the Denise board from Enterlogic instead?
It won’t be long before they add PCIE and run 4090 graphics cards just to prove that it can ;)
PCIe is largely just PCI with a different phy layer. so yes it can be done, but the real pain point is mapping the 4090 into the amiga’s limited memory model
Jay miner talked at a meeting I attended. When commodore bought the original company that designed the Amiga, Jay was tasked with firing the programmers as they completed their parts of the then ne amiga. Commodore as with so any corporations threw away te brilliant braintrust that created the Amiga and were never able to replace them. Hence any new versions 500, 1200 etc were just warmed over versions of the original vision.
What about those Atari falcons?
My bad i mean your PlayStation 1 and 2
I thought you meant the Atlanta Falcons…
B^)
This is so incredibly cool. Excellent job!