Posted by cyrc 6 days ago
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.
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.
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.
It also appears to have been spearheaded by Taki Udon, one guy who wanted something so he built it. We do indeed seem to be entering the golden age of FPGA.
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.
Yeah, the soft power feels really magical compared to other systems of the time. And not only can it turn itself on/off in response to a user request from the power button, but you can also set an alarm within the microcontroller that handles soft power and the RTC and it'll automatically power itself on or off whenever the alarm goes off. Pretty neat for 1983.
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.
It’s neat that there’s such passionate Lisa fans out there. If anyone here is a current day Lisa enthusiast, I’m curious what makes you interested in it.
Then how was it originally developed? Did Apple also patch their own OS?
I really want to adapt what I've done into an amiga500 accelerator board.
I didn't correct for inflation but I wanted to buy the Lisa before it was released, it felt around 40000 Dutch guilders, maybe 80 times more expensive than this FPGA?
I did a few more back-of-the-envelope calculations of what I can do with these 2MB SRAMs:
Xerox Alto with Smaltalk-80 and Smalltalk-76 for $4. The Alto was the 1972 machine the Lisa tried to be the sucessor of.
Transputer T414/T800 for $50 but much faster than the original. You would make a supercomputer interconnecting hundreds of Transputers.
Vextrex without display but HDMI output for $50, $8 without the CRT/VGA/Oscilloscope, $100 with the cathode ray tube display built in.
200MB SRAM with 16000 cores 180nm WSI (Wafer Scale Integration) emulating most processors at $1000. It would outperform 2025 Blackwell NVDIA and Apple Silicon M3 Ultra Mac Studio because SRAM is faster than HBM or LPDDR5. It is much cheaper than the 2MB Sram on this Lisa FPGA (it costs around $25 per 2MB (16 Mbit) in batches of 1000 chips).
Exactly the first thought I had too. I know extremely little about FPGA development, but three things I noticed that came to mind re: difficulty:
- Alex used a Xilinx FPGA, the MiSTer uses an Altera Cyclone - dunno how portable code (if that’s even the right term for e.g. VHDL) is from one to the other. I know the MiSTer has a light framework for cores to plug into to get input handling, scalers, etc.; so maybe it’s more a matter of porting to the framework…?
- Alex mentioned the SCC didn’t have a pre-made FPGA core so they used a real one. I don’t think serial handling would be critical but I do suspect you’d at least need a dummy to get the OS to pass self-tests and boot properly. Possible that maybe the Mac core has already handled this, though.
- What little I know of RAM and the MiSTer would lead me to think the SDRAM card a MiSTer setup typically needs wouldn’t be a problem over the SRAM Alex used, and that either the framework or the wiring of the RAM card handles the details for you - but I definitely don’t know that.
On the plus side I suspect/hope maybe a bunch of stuff from the classic/original Mac core could be borrowed to get it up and running.
There’s definitely plenty of cores that haven’t yet been developed on the MiSTer… for instance there isn’t a color 68K Mac core, only recently have people started on 3D0 and CD-i and Apple IIgs cores, the Saturn core was pretty shaky until a recent overhaul, etc. I think what’s there is just a function of what was either already developed for an FPGA or what had the biggest demand from their respective communities.
Here's the source code: https://github.com/arcanebyte/uniplus
And here are the disk images, although I think you need to be logged in to download them: https://lisalist2.com/index.php/topic,103.0.html
1. Wow! That's so cool!
2. Why didn't someone this smart spend that time to build something that really matters?
Is there anything better than to spend your time doing things that matter to you rather than needing some sort of external utility validation?
Some don't like to sell themselves and compete.
Some have confidence issues.
Some have ADHD and an on-paper track record that looks terrible.
Some want nothing to do with corporate America.
My personal philosophy on this is that in the grand scheme of things, there is almost nothing that "really matters". So you might as well spend as much of your time as is reasonable doing something fun.
I'm typing here on HN this morning. I could've been doing something more useful.
I'm about to put this down and play Animal Crossing. That isn't something that matters.
You could reasonably argue that I'm not as smart as this person, but I'm 100% behind his desire to have hobbies he enjoys.