high resolution floating point

[CdeMills]

Hello,

I tried to implement a multi-precision class based upon the gmp lib; except
it does not work and fails mysteriously on very simple operations. This
class is similar to double, except that the resolution may be adapted at
will. To be fully interesting, we should also agree that "MP_arg OP
standard_arg" should be evaluated as "MP_arg OP (MP version of
standard_arg)".

I think that basically I have to add a new type to ov-base.h, duplicate
things like ov-flt.h and so on. Would it be possible to open a
"multi-precision" branch, with the objective of integrating this new type,
and possibly merge it back into official Octave once full functionnal
integration is reached ?

Regards

Pascal



--
View this message in context: 
http://octave.1599824.n4.nabble.com/high-resolution-floating-point-tp4661592.html
Sent from the Octave - Maintainers mailing list archive at Nabble.com.