In gprolog, a float is represented as a C double which is represented
on 64 bits independently from the underlying architecture (32/64 bits
machines).
There is no way to increase this precision.
Daniel
Dave Sworin a écrit :
|
I checked a 64-bit and 32-bit system for max_integer with
the following results:
AMD Athlon 64 max_integer is 1152921504606846975 :)
Intel Pentium 3 max_integer is 268435455
However when I check the floating point representation I
always get the same
16 meaningfull digits. The maximum exponent was 307.
AMD Athlon 64: X is (1/3). X = 0.33333333333333331 ;)
Intel Pentium 3: X is (1/3). X = 0.33333333333333331
Is there a build option when increases the floating point
precision or is this
representation part of gprolog 1.3.0.
Dave
|
--
Ce message a été vérifié par
MailScanner
pour des virus ou des polluriels et rien de
suspect n'a été trouvé.
_______________________________________________
Users-prolog mailing list
[email protected]
http://lists.gnu.org/mailman/listinfo/users-prolog
|
_______________________________________________
Users-prolog mailing list
[email protected]
http://lists.gnu.org/mailman/listinfo/users-prolog
