[ale] so im crazy

[Irv Mullins]

On Wednesday 19 December 2001 03:17 pm, Charles Shapiro wrote:

> Uh, hmm.. I wrote in RCA 1802 machine code (not assembler) on a
> Netronics elf II for a time, similar to (but even smaller and less
> featureful than) the one documented at
> http://www.qwkslva.com/~abond/Museum/Netronics_Elf/netronics_elf.html. I
> still have the machine, but last I plugged it in it din't work, alas.
> It's an antique now. 

I have an Elf II, which served for years as a programmable timer in a 
gov't lab. Amazing what you can do with 0x46 bytes. Still works, too.

Another, which I wire-wrapped, has a 80 char x 24 line video display 
(on a 4" CRT) plus a 4 hex - digit LED readout. Haven't fired that one up in 
years. There are still some 1802's running in Viking and Voyager spacecraft, 
I think, which were still working when they departed the solar system headed 
for another galaxy.

> Writing in machine code is close to what you'd call
> "writing a program in hex"; I had to calculate and track things like
> destinations of jumps and data locations by hand, instead of sticking a
> label into the text and letting an assembler take care of that chore. It
> was grueling. I filled a lot of notebooks with hand-written columns of
> numbers.

Exactly - hand-assembled code is an excellent way to learn. 
Note that it has been 20+ years, and we still can recognize 
1802 programs just from a hex dump. ; )
I did eventually write a cross-assembler to run on the early IBM pc.
That's another worthwhile learning experience.

> Machine code//assembler//low-level code rapidly becomes
> machine-dependent. This tends to make it a good training exercise, but
> the specific knowledge you gain is of limited utility for anything
> except the very machine on which you're playing. Of course, writing a
> significant ( over 100 lines ) assembly-language program will teach you
> a lot both about how much you want to write code and how the guts of
> software really works. If you're gonna write code for a living, you
> should definitely write such a thing. But probably only once. I've
> written in assembly code for MS-DOS, but not for anything more complex.

Actually, check out http://grc.com/smgassembly.htm
Steve Gibson (inventor of SpinRite) shows how to do fast, tiny Windows 
programs in 100% assembler. It's pretty easy to understand, not at all 
daunting. (Not really necessary, however, since there are plenty of free, 
higher level languages for doing Windows programming)

Regards,
Irv



---
This message has been sent through the ALE general discussion list.
See http://www.ale.org/mailing-lists.shtml for more info. Problems should be 
sent to listmaster at ale dot org.