5 ms·

So was mine, although a little more indirectly. With AmigaBASIC 1.2 from MicroSoft but probably without any input from Bill Gates. My first programming sessio

by bhaak 10y ago

So was mine, although a little more indirectly. With AmigaBASIC 1.2 from MicroSoft but probably without any input from Bill Gates. My first programming session involved changing the parameters of a ellipse drawing routine. Which was AFAIR from a programming tutorial that was part of the manual.

vidarh 10y ago
And AmigaBASIC was part of cementing Microsoft-hate amongst Amiga users early on. It's famously buggy and slow. One of the bugs involved hacky use of the top 8-bit of pointers to store flags - this worked on 68000 CPU's as thy only had a 24 bit address bus and ignored the rest, but would cause crashes on machines with faster CPUs/accelerators (there are patches, but AmigaBASIC was dropped for later AmigaOS releases anyway) > My first programming session involved changing the parameters of a ellipse drawing routine. Which was AFAIR from a programming tutorial that was part of the manual. This is one of the Commodore traditions I really miss: The manuals tended to start with programming, and also included fairly detailed schematics. I know many other home computers did this, but I think Commodore with the Amiga is one of the ones that persisted with it the longest, and were most systematic about it. The VIC-20 manual I think is one of the best early examples of this approach [1], where the manual starters with programming right after showing how to unpack and connect the computer. [1] https://archive.org/details/Personal_Computing_On_The_VIC-20_1982_Commodore https://archive.org/details/Personal_Computing_On_The_VIC-20...
- cmrdporcupine 10y ago
  Yeah it's interesting that both Atari and Commodore dropped the ball on the BASIC options for their 16-bit machines introduced at the same time. ST BASIC was also crappy, and it wasn't until later that better third party options became available (GFA BASIC was really good, as was HiSoft BASIC, but both came later). In large part the C64 and Apple II had their success because it was easy for owners to make the transition from hobbyist to programmer by starting with the built-in BASIC and just working up. BASIC was right there in front of you, always accessible, and you basically _had_ to learn it to use the machine. So it created a ready supply of new content producers. To program the ST effectively you really needed a C compiler, but of course back in those days that was not cheap at all or even readily available (or very fun on a machine with no hard drive, which was the common situation.) I can't help but wonder if those platforms could have had better success if they'd just shipped out of the box with a decent BASIC, an assembler, and maybe even a C compiler. Or at least had that available on launch day for a price accessible to hobbyists.
  - tired_man 10y ago
    But that was half the fun of doing learning to hack on a C64. Adding coder tools typed in from MLX code from Compute! or Compute!'s Gazette and learning assembly by taking apart programs in Monitor was a deep learning experience for many beginning 6502 coders. It's sad that they didn't last, but the chip was simply an evolutionary dead end. I've recently thought about finding one to give my grandson in an attempt to pry away from a tablet his parents gave him. The crop of tablet/cell-addicted kids is a sad state of affairs.
    - cmrdporcupine 10y ago
      I think you're missing the point of my argument. I was complaining that the ST and Amiga did not ship as easily programmable, nuts-and-bolts-visible, platforms like the C64 and Apple II etc. did. I too learned a lot on my VIC-20. But when I got my Atari ST I had to wait a while til I got a compiler (Personal Pascal and then later GFA BASIC and Modula-2) so I could do anything interesting with it. As for 6502 being an evolutionary dead end -- it's still being made, and still being used in actual products. You'd be surprised what still has a 6502 or 6502-based core in it. It is a wonderful chip - simple easy to learn ISA, absolutely stellar interrupt responsiveness, and easy to interface.
    - qwertyuiop924 10y ago
      >The crop of tablet/cell-addicted kids is a sad state of affairs. Speaking as one of them, yes. They're hard to program, no fun, and even the BASIC offering won't let you do much.
  - vidarh 10y ago
    On the Amiga, it basically meant that for many years assembler ruled larg segments of the hobbyist space, at a time when a lot of people would've benefited from transitioning to C. But it also resulted in things like AmigaE gaining a lot of popularity [1], and I suspect a lot of the reason ARexx got so widespread was because it took over some the space a decent Basic would have. Incidentaly, E had a big leg up on C on the Amiga because the C compilers were notoriously slow, while the AmigaE compiler was blazing fast (it was written entirely in assembler, and did things like reading the entire source into RAM and treating it as an array) [1] http://strlen.com/amiga-e http://strlen.com/amiga-e

by bhaak 10y ago

vidarh 10y ago

And AmigaBASIC was part of cementing Microsoft-hate amongst Amiga users early on. It's famously buggy and slow. One of the bugs involved hacky use of the top 8-bit of pointers to store flags - this worked on 68000 CPU's as thy only had a 24 bit address bus and ignored the rest, but would cause crashes on machines with faster CPUs/accelerators (there are patches, but AmigaBASIC was dropped for later AmigaOS releases anyway) > My first programming session involved changing the parameters of a ellipse drawing routine. Which was AFAIR from a programming tutorial that was part of the manual. This is one of the Commodore traditions I really miss: The manuals tended to start with programming, and also included fairly detailed schematics. I know many other home computers did this, but I think Commodore with the Amiga is one of the ones that persisted with it the longest, and were most systematic about it. The VIC-20 manual I think is one of the best early examples of this approach [1], where the manual starters with programming right after showing how to unpack and connect the computer. [1] https://archive.org/details/Personal_Computing_On_The_VIC-20_1982_Commodore https://archive.org/details/Personal_Computing_On_The_VIC-20...

cmrdporcupine 10y ago
Yeah it's interesting that both Atari and Commodore dropped the ball on the BASIC options for their 16-bit machines introduced at the same time. ST BASIC was also crappy, and it wasn't until later that better third party options became available (GFA BASIC was really good, as was HiSoft BASIC, but both came later). In large part the C64 and Apple II had their success because it was easy for owners to make the transition from hobbyist to programmer by starting with the built-in BASIC and just working up. BASIC was right there in front of you, always accessible, and you basically _had_ to learn it to use the machine. So it created a ready supply of new content producers. To program the ST effectively you really needed a C compiler, but of course back in those days that was not cheap at all or even readily available (or very fun on a machine with no hard drive, which was the common situation.) I can't help but wonder if those platforms could have had better success if they'd just shipped out of the box with a decent BASIC, an assembler, and maybe even a C compiler. Or at least had that available on launch day for a price accessible to hobbyists.
- tired_man 10y ago
  But that was half the fun of doing learning to hack on a C64. Adding coder tools typed in from MLX code from Compute! or Compute!'s Gazette and learning assembly by taking apart programs in Monitor was a deep learning experience for many beginning 6502 coders. It's sad that they didn't last, but the chip was simply an evolutionary dead end. I've recently thought about finding one to give my grandson in an attempt to pry away from a tablet his parents gave him. The crop of tablet/cell-addicted kids is a sad state of affairs.
  - cmrdporcupine 10y ago
    I think you're missing the point of my argument. I was complaining that the ST and Amiga did not ship as easily programmable, nuts-and-bolts-visible, platforms like the C64 and Apple II etc. did. I too learned a lot on my VIC-20. But when I got my Atari ST I had to wait a while til I got a compiler (Personal Pascal and then later GFA BASIC and Modula-2) so I could do anything interesting with it. As for 6502 being an evolutionary dead end -- it's still being made, and still being used in actual products. You'd be surprised what still has a 6502 or 6502-based core in it. It is a wonderful chip - simple easy to learn ISA, absolutely stellar interrupt responsiveness, and easy to interface.
  - qwertyuiop924 10y ago
    >The crop of tablet/cell-addicted kids is a sad state of affairs. Speaking as one of them, yes. They're hard to program, no fun, and even the BASIC offering won't let you do much.
- vidarh 10y ago
  On the Amiga, it basically meant that for many years assembler ruled larg segments of the hobbyist space, at a time when a lot of people would've benefited from transitioning to C. But it also resulted in things like AmigaE gaining a lot of popularity [1], and I suspect a lot of the reason ARexx got so widespread was because it took over some the space a decent Basic would have. Incidentaly, E had a big leg up on C on the Amiga because the C compilers were notoriously slow, while the AmigaE compiler was blazing fast (it was written entirely in assembler, and did things like reading the entire source into RAM and treating it as an array) [1] http://strlen.com/amiga-e http://strlen.com/amiga-e