Hi,
Following up on the last weeks thread on digital filtering
on pic's I managed to get a few hours of 'free' time and
implement an 24 bit version of Scott Dattalo's 'twist' routine.
Enclosed you will find an ( albeit a bit bloated :-) ) fully
working 256 step IIR filter for 24 bit values.
And by 256 steps I mean how many steps one has to weigh against
new readings versus old readings.

I'm by no means an good PIC programmer, so I guess that
the code could be improved quite a bit. :-)

Also the code is very easy to modify for 16 or even 8 bit variables.

Have fun,

/Tony

BTW How is it with 'enclosed' code ? acceptable ? it IS regarding
PIS's and I consider it to be somewhat useful, but others might
not agree with me, please give me a note if You find this 'unwanted'.
Unfortunately I dont have access to private webspace DS


Tony KŸbek, Flintab AB            
ΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣ
E-mail: tony.kubek@flintab.com
ΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣΣ


Content-Type: application/octet-stream; name="Twist24.asm"
Content-Disposition: attachment; filename="Twist24.asm"

Attachment converted: creation:Twist24.asm (????/----) (00015824)