Back to the topic of the binary to BCD conversion, I found in the
PICLIST Source Code Library a shift-and-add-3 routine for
32-bit-to-10-digits by Ron Kreymborg and Mike Keitz
(<http://www.piclist.com/techref/microchip/math/radix/b2bp-32b10d.htm>)
that is more optimized than the one I posted here.

It takes 33 instruction words an runs in 2250 instruction cycles. I made
an unrolled loop version that takes 54 instruction words and runs in
1514 instruction cycles.

I wish to update my MPLAB-C18 'printf' and I think this routine should
be much more efficient than the Microchip's original approach of
divide/module,  but perhaps Scott's algorithm can be expanded to this
size (32-bit to 10 digits) with much better efficiency. What do you
think? Scott?.


Best regards,

Isaac

__________________________________________________
Fa=E7a liga=E7=F5es para outros computadores com o novo Yahoo! Messenger =

http://br.beta.messenger.yahoo.com/ =


-- =

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