Since Marcel will need to save the 12(16)-bit number from
one iteration to the next, Andy's otherwise perfectly
working routine needs a slight modification.

Andrew Warren wrote:

>        MOVF    OLDLO,W
>        SUBWF   NEWLO,W
>        MOVF    OLDHI,W
>        SKPC
>        INCFSZ  OLDHI,W
>        SUBWF   NEWHI,W

         MOVF    NEWLO,W
         MOVWF   OLDLO
         MOVF    NEWHI,W
         MOVWF   OLDHI

>        BC      NEW_IS_LARGER_OR_EQUAL
>        GOTO    NEW_IS_SMALLER


It's hardly worth showing, but it's possible to shave a
cycle off of this 'compare and save' algorithm:

    movf   new+1,W   ;W = LSB of new
    subwf  old+1,W   ;W = old - new  (LSB)
    subwf  old+1,F   ;old = old - (old - new) = new

  ;at this point, the carry bit is set if the low byte of old
  ;is greater than or equal to the low byte of new. Also, new
  ;has been copied to old's location.

    movf   new,W
    skpc
     incfsz new,W    ;
      subwf old,W

    movf   new,W     ;Now copy the high byte. Note that
    movwf  old       ;the subwf trick won't work here.
                     ;But then again, Dmitry hasn't put his
                     ;two rubles in yet.

    skpnc
     goto  old_is_larger_or_equal
    goto   old_is_smaller



BTW Andy, you forgot your comments!

Scott