Hi Eddy,
I don't quite understand what approximation error you mean. The sub-
processes you refer to are just calculating sums separately, and then
summing those sums. There are no approximation errors in this process
(except for those related to machine precision), as sums are entirely
associative (you can group them in any way you want, you will always
get the same answer!
Cheers,
Saad.
On 18 Mar 2009, at 10:13, Eddy D.L. wrote:
> Hi Saad,
>
> I have a question:
> Can I calculate approximation error of my final template? Cause
> divided job
> in different part!
> or how can I tell that my template is good although construction of
> the
> model from several sub-processes?
>
>
> cheers,
>
> Eddy
>
Saad Jbabdi
Oxford University FMRIB Centre
JR Hospital, Headington, OX3 9DU, UK
+44 (0) 1865 222545 (fax 717)
www.fmrib.ox.ac.uk/~saad
|