In message <a467mvc43cqjdek5m93u62toemvt6asqq1@[EMAIL PROTECTED]
>, Roedy Green
<roedy@[EMAIL PROTECTED]
> writes
>The PDP-11 started out simply enough when 64K seemed like infinite
>RAM, but it got a little hairier once you tried to crack that barrier.
64K, when I was a lad, I had 5K, well thats what Commodore advertised on
the box. Nasty marketing folks, I had 3.5K! 0.5K had gone to the screen
and the rest to the OS, but if you knew where the cassette read buffer
was you could poke a nifty assembler routine in there too. Vic=20 time!
Anyone else?
>When you are coding Intel assembler, even after a long familiarity,
>you are always looking things up in the book.
Oh yeah. Wrote my first (and only) game on the 80x286 in 1998 in my
lunch hour. I now use the Sybex 80386 book I bought in the same year, 15
years later. And you are right, I do look up the instructions on a
regular basis. Its a minefield, variable length instructions indeed.
>not quite as dense. As I recall, the only mildly tricky thing about
>68K assembler is knowing when you will get sign extension and when
>unsigned.
Yup, caught me out a treat ****ting the x86 game to the Atari ST.
>The thing that puzzles me is the complexity of the PowerPC and the
>Itanium. I thought the world was going RISC. These machines have more
Itanium, bluergh! I ****ted a huge Win32 CAD application to Win64 on
Itanium. Where is the stack? Er, its hidden. Where? Can't tell you. Why
not? Well, er, the register stack engine knows where it is. The what? We
don't want you munging about in the stack with your debuggery things, so
we've hidden the stack so the processors register renaming can really
much about with things. The stack, well its there, but we aren't telling
you where it is.
Horrible, Horrible, Horrible, from a debugging point of view.
Didn't help that my desktop Pentium could outpace the Itanium without
breaking into a sweat. Athlon 64 is the way to go. No competition there.
>instructions than you could shake a stick at. Many of the Itanium
>instructions I can't even think what on earth you would use them for.
RISC is a farce. The most RISC processor I've found is the 6502. 51
instructions (from memory). Lots of flexibility. 3 registers (A, X, Y).
Problem is, 6502 is a CISC, not a RISC.
>The best toy I ever had was a DEC MINC (a PDP-11 with fancy I/O
>interfaces) and the BC Hydro solar energy research lab. However, the
>DEC software made it TOO easy.
You make me feel young writing all that stuff. Wow!
Stephen
--
Stephen Kellett
Object Media Limited http://www.objmedia.demon.co.uk
RSI Information: http://www.objmedia.demon.co.uk/rsi.html
|
