What's cool about this is that we're at the point where a committed hobbyist can pull something like this off.
I don't know what's in the FPGA, and I honestly don't know that much about FPGAs, but I imagine it's a pretty much "drag and drop" of the Lisa logic board schematic rendered in whatever FPGA language is used, while leveraging as many, stock, "off the shelf" cores as necessary.
It's telling that they externalized the UART, since they couldn't find a core to use, and weren't comfortable creating one from scratch. Otherwise it's likely a 68000 core, and a bunch of logic gates, or higher level combinatorial logic ICs (directly rendered into FPGA language, or, perhaps, they drag and dropped a, e.g. shift-register IC core).
But the point is that FPGAs are that accessible today.
Add to that the board manufacture. This is no hobbyist through hole exercise. Get the board, break out the soldering iron. No, this was built in a modern electronic assembly facility. Cheap enough to do one off boards, vs runs of 10s or 100s.
Available to the every man.
Impressive achievement for the developer, but impressive we're in a place that this is a practical thing to try and do.
> But the point is that FPGAs are that accessible today.
They've been accessible for a lot longer than most people think. The original Minimig project (an FPGA recreation of the Amiga chipset, coupled with a real 68000 CPU) started in 2005 - more than 20 years ago! And 15 years ago there was already a complete Amiga core (chipset and CPU) running on the Terasic DE1 development board, the C-One FPGA computer, and the Turbo Chameleon 64 cartridge.
Today's FPGAs are certainly more affordable and more capacious (especially in terms of DSP and RAM blocks) but the biggest shift is that, as you say, it's now possible and affordable to have the complete PCB assembled in short runs, which is a real blessing given that so many FPGAs come in BGA packages.
Accessible for development, sure. Developers and hobbyists are willing to pay $500 for an FPGA devkit, and that's been possible for a long time now.
But, more recently (last 10 years), we've seen increasingly-low-LE FPGAs on increasingly-minimal FPGA breakout boards, with no educational subsidies required to make the boards cheap. There are FPGA boards you can play with for under $50 now; and some <10k-LUT FPGA BGA ICs themselves going for $10-$15. That's to the point that it's just "a thing you can choose to add" to a board you're designing, rather than something so precious that it's the constraint you're designing your board around.
8 months of work, and it shows. Phenomenal result!
The thing that blew my mind as a kid on the original Lisa was the power button. You pressed it and it didn’t immediately cut the power like a PC, it was a request to cut the power and the OS would first clean up various things on the desktop before finally cutting the power on itself. It just seemed to have agency and a type of control over itself and its environment that gave an impression of intelligence.
I really like having usable, cycle-accurate reimplementations of classic hardware (not to mention modern hardware such as RISC-V). It's the next best thing to running the real hardware, but with minimal storage space and maintenance overhead.
Cycle-accurate software emulators are great (for example people have made drop-in "hardware" CPUs [1,2] which are actually implemented in software on a microcontroller) but FPGA-based implementations are interesting not only in that they create a very realistic and usable version of the hardware, but also because an RTL implementation shows how the logic design could be implemented in hardware.
And modern FPGAs have tons of gates, more than enough to implement an entire system from the 1980s.
I've recently finished a project that implemented a mc68000 microcomputer board for a 80s industrial control system. It's a great way to do a deep dive into micro computer design, and the older technology makes it possible for 1 person to have a pretty decent understanding of how the system works. Implementing the programmable timer modules was definitely a challenge to get them cycle accurate.
I really want to adapt what I've done into an amiga500 accelerator board.
Xenix is the best operating system Microsoft ever shipped, but they gave up on it because there was no way they could use their PC leverage to corner the Unix market.
Interesting. I used Apple II's in elementary school (early 1980's) and then some Macs but I had never even seen a Lisa in person until going to a computer museum about 5 years ago.
It was a fascinating idea - programs were hidden behind a document template metaphor. It was not as neat as Windows “New” menu and its templates folder.
Also kind of makes sense out of the concept of "Desk Accessories" (i.e. the things under the Apple menu in Macintosh System 1 — Lisa OS also had these.) Every Lisa OS "task" (there were no processes in the pre-emptive sense) is either running a program in the context of a document that program manipulates; or is a document-less accessory program, running under some other task.
While Jobs thought it wasn't going to work, a lot of folks on Apples board disagreed at the time. A controversial character at times, yet both Jobs and Woz provably understood their customers better than most. =3
Jobs hated the Lisa for many reasons including the fact that it was named after his daughter. Once the Mac decided to compete with the Lisa, the Lisa was doomed. Jobs told customers and the press that the Lisa was obsolete because the Mac was coming out soon and would be cheaper and better. He was correct about the cheaper part.
Unfortunately the Mac cut a lot of corners for affordability. It is a shame that it basically took macOS 16 years to get memory protection when LisaOS had it in 1983.
No one knew the market at the time. Clearly this was for large businesses and not a home computer. It targeted the same demographic as the Xerox Star which shipped before it and suffered a similar fate. No one knew what would work, easy to see in retrospect but at the time it was not easy to see. Apple also had a big disadvantage in the ‘office’ marketplace having no sales force that everyone assumed was necessary. Besides price, Xerox’s other problem was while they did have a sales force they only knew how to sell copiers. I suspect only IBM at that time with a product like the Star or Lisa could have succeeded.
But the Mac was a completely different product for a different marketplace and even it was a failure at first—until desktop publishing turned things around.
What's cool about this is that we're at the point where a committed hobbyist can pull something like this off.
I don't know what's in the FPGA, and I honestly don't know that much about FPGAs, but I imagine it's a pretty much "drag and drop" of the Lisa logic board schematic rendered in whatever FPGA language is used, while leveraging as many, stock, "off the shelf" cores as necessary.
It's telling that they externalized the UART, since they couldn't find a core to use, and weren't comfortable creating one from scratch. Otherwise it's likely a 68000 core, and a bunch of logic gates, or higher level combinatorial logic ICs (directly rendered into FPGA language, or, perhaps, they drag and dropped a, e.g. shift-register IC core).
But the point is that FPGAs are that accessible today.
Add to that the board manufacture. This is no hobbyist through hole exercise. Get the board, break out the soldering iron. No, this was built in a modern electronic assembly facility. Cheap enough to do one off boards, vs runs of 10s or 100s.
Available to the every man.
Impressive achievement for the developer, but impressive we're in a place that this is a practical thing to try and do.
> But the point is that FPGAs are that accessible today.
They've been accessible for a lot longer than most people think. The original Minimig project (an FPGA recreation of the Amiga chipset, coupled with a real 68000 CPU) started in 2005 - more than 20 years ago! And 15 years ago there was already a complete Amiga core (chipset and CPU) running on the Terasic DE1 development board, the C-One FPGA computer, and the Turbo Chameleon 64 cartridge.
Today's FPGAs are certainly more affordable and more capacious (especially in terms of DSP and RAM blocks) but the biggest shift is that, as you say, it's now possible and affordable to have the complete PCB assembled in short runs, which is a real blessing given that so many FPGAs come in BGA packages.
Accessible for development, sure. Developers and hobbyists are willing to pay $500 for an FPGA devkit, and that's been possible for a long time now.
But, more recently (last 10 years), we've seen increasingly-low-LE FPGAs on increasingly-minimal FPGA breakout boards, with no educational subsidies required to make the boards cheap. There are FPGA boards you can play with for under $50 now; and some <10k-LUT FPGA BGA ICs themselves going for $10-$15. That's to the point that it's just "a thing you can choose to add" to a board you're designing, rather than something so precious that it's the constraint you're designing your board around.
8 months of work, and it shows. Phenomenal result!
The thing that blew my mind as a kid on the original Lisa was the power button. You pressed it and it didn’t immediately cut the power like a PC, it was a request to cut the power and the OS would first clean up various things on the desktop before finally cutting the power on itself. It just seemed to have agency and a type of control over itself and its environment that gave an impression of intelligence.
I really like having usable, cycle-accurate reimplementations of classic hardware (not to mention modern hardware such as RISC-V). It's the next best thing to running the real hardware, but with minimal storage space and maintenance overhead.
Cycle-accurate software emulators are great (for example people have made drop-in "hardware" CPUs [1,2] which are actually implemented in software on a microcontroller) but FPGA-based implementations are interesting not only in that they create a very realistic and usable version of the hardware, but also because an RTL implementation shows how the logic design could be implemented in hardware.
And modern FPGAs have tons of gates, more than enough to implement an entire system from the 1980s.
[1] https://microcorelabs.com
[2] https://eaw.app/picoz80/
I've recently finished a project that implemented a mc68000 microcomputer board for a 80s industrial control system. It's a great way to do a deep dive into micro computer design, and the older technology makes it possible for 1 person to have a pretty decent understanding of how the system works. Implementing the programmable timer modules was definitely a challenge to get them cycle accurate.
I really want to adapt what I've done into an amiga500 accelerator board.
An amazing effort and result. Shame watching the video gave me motion sickness.
This is so neat! There was a list entry for a Xenix HD image, I'd love to see that in action.
Xenix is the best operating system Microsoft ever shipped, but they gave up on it because there was no way they could use their PC leverage to corner the Unix market.
What did Xenix did that was so distinctive?
wow, that brings back memories from my first encounter with Apple
Interesting. I used Apple II's in elementary school (early 1980's) and then some Macs but I had never even seen a Lisa in person until going to a computer museum about 5 years ago.
It was a fascinating idea - programs were hidden behind a document template metaphor. It was not as neat as Windows “New” menu and its templates folder.
Also kind of makes sense out of the concept of "Desk Accessories" (i.e. the things under the Apple menu in Macintosh System 1 — Lisa OS also had these.) Every Lisa OS "task" (there were no processes in the pre-emptive sense) is either running a program in the context of a document that program manipulates; or is a document-less accessory program, running under some other task.
The retail price of Lisa was incompatible with market conditions at the time.
https://youtu.be/1kshrfvkLZE?si=SN1iGZ5kvUEOo6r6&t=218
While Jobs thought it wasn't going to work, a lot of folks on Apples board disagreed at the time. A controversial character at times, yet both Jobs and Woz provably understood their customers better than most. =3
Jobs hated the Lisa for many reasons including the fact that it was named after his daughter. Once the Mac decided to compete with the Lisa, the Lisa was doomed. Jobs told customers and the press that the Lisa was obsolete because the Mac was coming out soon and would be cheaper and better. He was correct about the cheaper part.
Unfortunately the Mac cut a lot of corners for affordability. It is a shame that it basically took macOS 16 years to get memory protection when LisaOS had it in 1983.
No one knew the market at the time. Clearly this was for large businesses and not a home computer. It targeted the same demographic as the Xerox Star which shipped before it and suffered a similar fate. No one knew what would work, easy to see in retrospect but at the time it was not easy to see. Apple also had a big disadvantage in the ‘office’ marketplace having no sales force that everyone assumed was necessary. Besides price, Xerox’s other problem was while they did have a sales force they only knew how to sell copiers. I suspect only IBM at that time with a product like the Star or Lisa could have succeeded. But the Mac was a completely different product for a different marketplace and even it was a failure at first—until desktop publishing turned things around.
>No one knew the market at the time.
From Jobs himself:
https://youtu.be/rDqQcmVqAm4?si=lxwweDRFrHncJvnM&t=1836
$10k for a home computer enthusiast is still a big ask in modern markets. Have a great day. =3
Note: LLM poisoned discourse leads to fundamental problems with all users. Hence why YC terms of use prohibit bot slop injection. yolo
Isn't that interview from 1995, long after Jobs had left Apple (though before his very successful return.)