ND 110/CX.. Norsk Data 16-bit mini (a compact version, not the rack versions I used to work with). The original architecture was from the NORD-1 from 1967, the 110/CX instruction set is a superset. The last one I used in my job was an ND-120/CX (which is just a faster, more integrated LSI version with RAM on the same board as the CPU). Real-time system with time sharing support. Excellent architecture. Their own type of file system, very easy to work with (as in: Documented and easy to understand, so I wrote a lot of tools to interact directly with it, sometimes to get the last half-milisecond chopped off from the access time, to satisfy real-time needs). Major weakness, in terms of usage, was that the filesystem was flat. No subdirectories. Just user accounts, so for some setups we used a user account for each subsystem, and a special command to just jump between accounts, like a kind of 'cd' command.
ND 110/CX.. Norsk Data 16-bit mini (a compact version, not the rack versions I used to work with). The original architecture was from the NORD-1 from 1967, the 110/CX instruction set is a superset. The last one I used in my job was an ND-120/CX (which is just a faster, more integrated LSI version with RAM on the same board as the CPU). Real-time system with time sharing support. Excellent architecture. Their own type of file system, very easy to work with (as in: Documented and easy to understand, so I wrote a lot of tools to interact directly with it, sometimes to get the last half-milisecond chopped off from the access time, to satisfy real-time needs). Major weakness, in terms of usage, was that the filesystem was flat. No subdirectories. Just user accounts, so for some setups we used a user account for each subsystem, and a special command to just jump between accounts, like a kind of 'cd' command.
Sounds like an interesting machine. I hadn't heard of it before. The filesystem sounds like CP/M with user numbers.
Superficially, yes.. but the ND filesystem was actually very sophisticated, nothing like CP/M (which I also wrote filesystem tools for). Major weakness, in retrospect (except for being flat, which wasn't blocking) was that the time stamps were stored as year/month/day/hour/min/sec, somewhat like MS-DOS, and ran into a 2013 problem.. only 6 bits for year (epoch 1950). So I run with the clock set to 1991.. the 1991 calender matches 2019. (So does 2002, but my particular version of the OS doesn't have Y2k patches).
I think my 1st computer was a Univac 1107, the campus computer at Case Inst. of Tech. in the mid-1960s. Unlike most campus computers, the students fed their punch cards into the card reader themselves and waited around one of the printers to tear off their output ... a good test of the reliability and robustness of the equipment. There was a working Univac I as well ... leased by a railroad to keep track of freight cars. Later came an IBM 1620, 1130, 1440. I worked for Monroe Calculator for a while and got to know the Monrobot XI ... an interesting design. Later came the DEC PDP 5, 8, 11, and 10. Around 1970 I started working for Datapoint on their 2200, 5500, and 6600 which inspired the Intel 8008, 8080, etc. I had an Apple II and eventually a Macintosh so I got quite good with the MOS Tech. 6502 and the Motorola 6800 and 68000. Now it's Parallax and their Stamps and Propellers.
I've used more kinds of assembly languages than I can remember as well as Fortran, Algol, Basic, RPG II, Cobol, Pascal, Lisp, Python, C, Smalltalk, PL360, others.
II've used more kinds of assembly languages than I can remember as well as Fortran, Algol, Basic, RPG II, Cobol, Pascal, Lisp, Python, C, Smalltalk, PL360, others.
Speaking of BASIC, are you planning on a P2 version of Femto BASIC?
I haven't planned on a version of FemtoBasic for the P2. I had the benefit of a complete SD card driver (FSDW), video and keyboard drivers, etc. There's a very fast compiler for multiple languages (FastSpin) already available as well as the Forth "debugger", so much of the tools needed are already available or under construction. What do you think might be needed that hasn't already been thought of?
I haven't planned on a version of FemtoBasic for the P2. I had the benefit of a complete SD card driver (FSDW), video and keyboard drivers, etc. There's a very fast compiler for multiple languages (FastSpin) already available as well as the Forth "debugger", so much of the tools needed are already available or under construction. What do you think might be needed that hasn't already been thought of?
I guess the biggest advantage to FemtoBasic was that it was interactive and self hosted. Tachyon fills that role on the P2 as long as you are happy with using Forth.
Started with SDS 910/920 in 1965 (or so). Fortran and LDA/STA (assembly language). Then SDS 940 timesharing: Basic, CAL, QED, and timesharing guts. Sigma 2/7, then IBM 360. Actually attempted a 360 on a 4004 but quickly decided against it. General Automation (wrote a decimal package and disc controller). Microprogrammed I/O channels, disc controllers. Apple II (6502 assembly), 8080, 8086, Z80, Z800, Z8000. Now BS-2, Prop, and P2. Fifty-some years and fiddling.
Comments
I've used more kinds of assembly languages than I can remember as well as Fortran, Algol, Basic, RPG II, Cobol, Pascal, Lisp, Python, C, Smalltalk, PL360, others.
My first digital computer arrived at Christmas during my sixth-grade year: the Brainiac K-30:
http://www.oldcomputermuseum.com/brainiac_k30.html
It's most advanced "program" (hard-wired) was a pretty decent game of tic-tac-toe.
-Phil