[Scilab-users] scilab 5.4.1 simple calculations need rounding

[Arvid Rosén]

Also, when comparing to scilab 4, you should keep in mind that the older versions (4 and earlier, I think) used the extend floating point feature of the FPU, giving 80 bits of precision in some cases. This might yield slightly different results in some calculations.

Cheers,
Arvid

15 sep 2013 kl. 20:55 skrev Samuel Gougeon <sgougeon at free.fr>:

> The big number of digits asked for leads to that. You may see what does the following:
> 
> clc
> %eps
> format(16)
> x = poly(0,'x'); f = (x-1)^2, factors(f)
> format(19)
> x = poly(0,'x'); f = (x-1)^2, factors(f)
> 
> Regards
> Samuel
> _______________________________________________
> users mailing list
> users at lists.scilab.org
> http://lists.scilab.org/mailman/listinfo/users

--
Arvid Rosén

Softube AB
S:t Larsgatan 10c
SE-582 24 Linköping
Sweden
Mobile: (+46) 73 25 123 47
Office: (+46) 13 21 16 23
Fax: (+46) 13 10 00 23
Mail: arvid at softube.com
URL: www.softube.com

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.scilab.org/pipermail/users/attachments/20130916/8a2fe940/attachment.htm>