Peter L. Peres wrote:
>
> > better
>
> decf    mid, F
> decf    mid, F
> decfsz  low, F   This test happens on LOW= 1 to 0, not 0 to FF.
>   incf  mid, F
> incfsz  mid, F
>   decf  high, F
>
> 6 T cycles, 6 words run time (constant), and no changes in W ;). I think
> it's optimal.
>
> Peter

  It should break there.  Also the mid test is it going back
up to zero, not ending on FF.  But giving me other ideas..

decf   low,f
incfsz low,f
decf   mid,f
decf   low,f
decf   mid,f
incfsz mid,f
decf   high,f
decf   mid,f

  Should work.  Move each down then up to test for zero and
modify next up before actual dec.  Still 8 cycles though..