> The start bit idea is clever! It allows a character to have
> up to 7 dits or dahs.

Heh.  The motorola "Black Widow" design contest (for their 8pin
cpus, essentially) has as one of the "steps" to finish a morse
code program.  The core that they supply stores the translation
table as STRINGS:

>>> char *convert(char c){    //Convert ASCII received 	switch(c){
>>> 		case 'a':
>>> 			return ".-";
>>> 		case 'b':
>>> 			return "-...";
>>> 		case 'c':
>>> 			return "-.-.";
>>>
I guess that says something about changes in attention paid to
code size by "youngins" these days.  And C vs ASM, perhaps.
Although since the high-end 8pin chip contains 8kbyte of flash
(more than any other 8pin uC?) I guess it doesn't matter.

> SendMorseCharA
>        RLF   MorseChar,f  ; shift msb into carry
>        movf  MorseChar,f  ; test to see if we're done
>        bz    SendMorseCharZ ; We're done, get out
>        skpc               ; skip if dit
>        call  DoDah        ; go send dah - PRESERVE CARRY!
>        skpnc
>        call  DoDit        ; go send dit
>        call  InterCharDelay
>        goto  SendMorseCharA ; go send next bit, if any
> SendMorseCharZ
>        return              ; we're outta here!

Speaking of C compilers, this is a good example of the sort of
code that I think it would be really difficult for a C compiler
to generate; It's what I meant by "explicit use of carry."  Does
anybody want to take a shot at C code/compiler implementation
that comes close to outputting this?

(And on the third hand, it is perhaps also an example of optimization
that doesn't need to be done.)

BillW
-- 
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist