PB DOD "PRINT" Oddity!

[Arthur Gomide]

Originally posted by James Horner:
You didn't read my post!

I apologize!
PB32 has this problem - to be safe you must use the trick of Mr. Bishop says above.
PB35 runs your code fine!

Best Regards,

Arthur.




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

[James Horner]

You didn't read my post!
N is double word
N = 98765432
PRINT N gives 8.876543E7

BUT, PRINT (N - 10000000 + 10000000)
gives the 8 digit integer 98765432

Nothing is different about the number in question
but PRINT renders then differently.

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

[James Horner]

As I said in the riginal post, N was declared as
a double word variable!

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

[Michael Mattias]

Help file under PRINT tells you why..

PRINT displays single-precision quantities with up to 6 significant
digits, and other quantities with up to 16 digits. Extended-precision,
binary coded decimal, and quad integer values with 17 or 18 digits must
be printed via the STR$ function if all digits are to be displayed.

.. which is, "because that's the way it works."

[Arthur Gomide]

Originally posted by James Horner:

Any guesses why??

The type of N means to be single-precision (default for variables without type declaration in PB32) in your example, and single-precision has a precision of 6 to 7 decimal digits.
Try defining N as a double word with:

DEFDWD N

'and then

N = 23456789
PRINT N ' now gives 23456789
PRINT (N - 10000000 + 10000000) ' now gives 23456789

Best Regards,

Arthur.

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

[Mel Bishop]

I can't answer your question but when dealing with numbers like
that, I usually use something like:

fmt$ = "##,###,###.##"   ' or use whatever format style that
print;using$(fmt$,n)     ' floats your boat.

This gives proper display 99 times out of a hundred.


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

[James Horner]

I stubled onto this oddity with PBDOS 3.2
using "PRINT" command.
I'm working with very large positive integers --in
declared as DOUBLWRD
With numbers N > 10,000,000 , PRINT N
showa on screen the a.bcd...E7 format
'and I lose the final digit.
However for such numbere, if I program it as
PRINT (N - 10000000 + 10000000) I get the full
eight digit numer on screen!
e.g.
N = 23456789
PRINT N gives 2.345678E7, But
PRINT (N - 10000000 + 10000000) gives 2345678 !
Any guesses why??

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