PDP2011: A re-creation of the PDP-11 computer systems in VHDL
DEC actually made a microprocessor implementation of the PDP-11 processor, called LSI-11. It has two chips, and is functionality equivalent to the large thing. Unfortunately, DEC was in a complete disregard of personal computers or workstations at that time, I don't think they made workstations with it. Otherwise it could be another legend in computing history: a PDP-11 workstation! Never happened.
Here's S100 computer board built around this microprocessor with a great introduction to the hardware details. http://s100computers.com/My%20System%20Pages/PDP11%20Board/P...
(In)famous for being 'middle-endian'
I used RSX-11M and I remember a few interesting things about it. First, it had an overlay linker: code would swap in if it used the same space as other unused code. Second, this extended to the operating system. There was a giant system build process where you linked all system programs in with the operating system. To run a command, it first had to be explicitly loaded into memory. So during this system generation process, you decide which commands should not interfere with other commands so that many could be loaded at once. Anyway, I remember it took many hours to run this process.
The 'M' means the system had virtual memory, but virtual address space was tiny- smaller than the physical memory space.
I'm so happy to see an upswing in PDP-11 projects in the past year. It's such a nice architecture- very extensively documented, and from a time where it was possible for a programmer to really understand what the machine is doing, top-to-bottom.
Another rathole to chase down. If I could emulate an 11/70 but connect a real decwriter paper terminal, I’d be doomed.
There is another VHDL PDP-11 called the W11a: https://wfjm.github.io/home/w11/
One eventual goal is for it to be a PDP-11/70mP, a PDP-11 "quadcore".
BSD 2.x is tbe most insane OS ever. It is a backport from Vax Unix 4.x -- which used real virtual memory, often "lots" of it (2MB or even more!), 32-bit registers -- back to the 16-bit PDP-11, which had two (count 'em) 8Kbyte mappable pages in its 64K address range, and could swap 8 other pages into each. It came with all the same utilities as Vax Unix, except the networking bits.
We were so happy when we got a big -- 3" high, 6-platter 5.25" 20M -- Shugart disk. Didn't know we needed to recompile the kernel so it wouldn't swap in the middle of the fs (which had been at one end of the old 5M disk).
This is really cool, though I have never played with the PDP-11 architecture. When I first took computer organization and architecture at my alma mater, Wittenberg University, my computer science professor (Brian Shelburne) made us learn the PDP-8 architecture. I will never forget learning machine code on the PDP-8. He was very popular among the PDP-8 fan community as he developed two different emulators for the PDP-8 and published a few articles on using the PDP-8 in teaching computer science. He made an emulator for Windows DOS and later on a different one in C++ for Windows 7/8/10.
I'm so drooling at the idea of building a pdp-11 laptop right now!
Supposedly, this project was going to include pinouts for the mini-replica front fascia provided by Pidp-11...a pretty cool project itself: https://obsolescence.wixsite.com/obsolescence/pidp-11
Cool! Now let's make a FPGA clone of Super Sprint.
I can't get onto this site - just times out - anyone else ?