[Scilab-users] Floating precision differences

[Adelson Oliveira]

Unfortunately not! It is 1.000D+16.

A friend of mine (unix guru) showed me that the command,

$ export LC_ALL=C

before calling scilab, is a workaround.

It seems that the problem comes from glibc ...

Thanks

2014-10-06 4:48 GMT-03:00 Frederic Jourdin <frederic.jourdin at shom.fr>:

>  I guess it should be: 1.000D-16 isn't it?
> You are testing limits of different computers, so slight differences are
> usual I think.
>
>
> Le 03/10/2014 16:34, Adelson Oliveira a écrit :
>
>     Hi,
>
> I've Downloaded scilab-5.5.1. I've installed it in Openmandriva 2014.1
> Linux and in Red Hat Enterprise 6.5.
>  Machines are different (intel i7 and intel Xeon)
>
>  At Openmandriva,
>
> --> 1.0000000000000001-1.
>   ans  =
>
>      0.
>
>  At Red Hat 6.5,
>
> --> 1.0000000000000001-1.
>   ans  =
>
>       1.000D+16
>
>  Why is it different?
>
>  Thanks
>
>
> _______________________________________________
> users mailing listusers at lists.scilab.orghttp://lists.scilab.org/mailman/listinfo/users
>
>
>
> _______________________________________________
> users mailing list
> users at lists.scilab.org
> http://lists.scilab.org/mailman/listinfo/users
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.scilab.org/pipermail/users/attachments/20141006/1c414e8d/attachment.htm>