Re: problem with example mjoclivar_14

From: Dennis Shea <shea_at_nyahnyahspammersnyahnyah>
Date: Thu Jan 07 2010 - 07:08:57 MST

shinn wong wrote:
> Dear ncl users,
> Attached is the code for computing multivariate eof. I have two
> questions concerning the multi-variate eof.
> 1)Is the sign of the combined eof arbitrary? Is it ok for me to change
> the sign of the eof?

As noted by Maurice, you have already asked this question.

(a) The sign of the EOF is arbitrary.

(b) It is "ok" to change the sign.

> 2)There are errors when I'm trying to change the sign of the eof by
> multiplying eof_cdata by -1 (line 151 to 153).

It is ok to multiply by -1 [integer or float]

  The error is:
> fatal:NclGetArgValue: In correct argument type passed to the function 5
> [main] ncl 2584_cygtls::handle_exceptions:Error while dumping state
> <probably corrupted stack> Segmentation fault <core dumped>. So i want
> to know what i actually did wrong.

This is an handle_exception "probably corrupted stack" ... I
think this is a CYGWIN issue *not* NCL.

===
I ran the script on locally available data. 2006 and 2007
were not available. The data spanned 1985-2005. You can download
the output at:

    http://www.cgd.ucar.edu/~shea/mjoclivar.1985-2005.ps.gz

%> gzip -d mjoclivar.1985-2005.ps.gz

then look at the postscript

> Thank you.
>
>
> ------------------------------------------------------------------------
>
> _______________________________________________
> ncl-talk mailing list
> List instructions, subscriber options, unsubscribe:
> http://mailman.ucar.edu/mailman/listinfo/ncl-talk
_______________________________________________
ncl-talk mailing list
List instructions, subscriber options, unsubscribe:
http://mailman.ucar.edu/mailman/listinfo/ncl-talk
Received on Thu Jan 7 07:10:28 2010

This archive was generated by hypermail 2.1.8 : Tue Jan 12 2010 - 15:38:20 MST