Re: Weird Iso-Contour behaviour

Hi Ugo,

> has anyone seen some weird behaviour with iso-lines as shown in the figure:
>
> http://www.geogr.uni-jena.de/~p6taug/weird_1.jpg
>
> The iso-lines go off to the bottom right corner. First I thought it was
> my data, which have missing values (mapped to -9999). But it is not
> really the -9999 or the missing data (NaN), because zoom out still shows
> the lines ending up where a rainbow ends: nowhere.
>
> See also
> http://www.geogr.uni-jena.de/~p6taug/weird_2.jpg
>
> which confirms that the isolines end/start somewhere *fixed* at minus
> infinity.
>
> Furthermore, other simple test data containing missing don't show this.
>
> Changing the parameters
> http://www.geogr.uni-jena.de/~p6taug/weird_3.jpg
>
> causes what, in my view, is abnormal behaviour: I set the base to 600
> and suddenly only a few lines are drawn. (Lines are clipped in length,
> and not in height.)
>
> Any views, ideas or at least some expression of pity ;-)

Certainly sympathy. Even more, an interest in fixing this bug.
Do you have a test program that you can send to the list so
I can recreate this bug? Thank you.

Cheers,
Bill