My PIC can't do the math! Can yours?

[paul borgmeier]

Sayzer,

Your PIC's math is working great!

Change the last part of your code to this:

FOR XA=0 to 150
XB=XB-2
IF XF = XA THEN
XF = XB
GOTO HERE
ENDIF
NEXT XA

HERE:
LCDOUT $FE,1,”Temp:”,#XF

Your code is actually changing the value of XF twice in the region you thought was flawed (e.g., for 46464, XB = 149. Your If Then sets XF = 149. Your loop continues and finds XF = 149. 303 - (149+1)*2 = 3 ... the math is correct)

Paul Borgmeier
Salt Lake City, Utah
USA

[sayzer]

Thanks Paul. I fixed the problem considering your catch.
I make the loop value the end number so it exits the loop.

XB=303
for XA=0 to 150

   XB=XB-2
   if xf=XA then 
     xf=XB
     XA=150    'once we have the match, lets exit immediately as Paul mentioned above.
   endif
 
next XA

Also, Thanks to you Jumper. I see what you are saying about ADCON1 but I have no issue with that register. I am now getting the correct results;

I found the bug :-)


------------

[Ingvar]

If this is all your program is supposed to do, i'd use something like this instead. Much faster.

ADCON1.7 = 1                          'right justified
Loop:
    ADCIN 1,TempRead
    IF tempread>800 THEN tempread=800 'Min. temp.
    IF tempread<650 THEN tempread=650 'Max. temp.
    XF = 301 - ((TempRead - 650)<<1)  '301-((TempRead-650)*2)
    Lcdout $fe,1,”Temp:”, #XF
GoTo loop

/Ingvar

[sayzer]

Thanks for all replies.

When I use "right justified" ADCON1 register, as Jumper and Ingvar suggested, the scale does not behave linear. Thus, I can not match the temp reading with Celsius.

Currently, “left justified” ADCON1 register provides a scale that is perfectly linear and the formula works great (thanks to Paul's catch).

By the way, Ingvar's code is indeed nice but I can not use it.



----------