| Re: TRIGRID co-linearity problem [message #24984] |
Wed, 09 May 2001 14:54 |
Christopher Davey
Messages: 2
Registered: May 2001
|
Junior Member |
|
|
Roman,
I went ahead and tried doing what you recommended, using the randomu or
randomn functions to add random parts to my latitude and longitude
arrays. But I am still getting a co-linear error from trigrid. Are you
somehow determining specifically which x (longitude) and y (latitude)
points are co-linear and then adding the random perturbations to these
points only?
Roman Schreiber wrote:
> I had a similar problem few weeks ago. Without going into details
> I tried a "brute force" approach adding to x and y geographical
> coordinates small random parts (about 0.0001 * coordinate value).
> Error disappeared and the resulting contour plot was OK.
> But it does not of course explain the reason for the peculiar
> behavior of trigrid procedure...
>
> Christopher Davey wrote:
>>
>> The triangulation( ) procedure works just fine, but In my call to
>> trigrid( ), I get an error that I cannot figure out: "Points are
>> co-linear, no solution."
>
> Regards
>
> Roman Schreiber
>
> schreibe@ncac.torun.pl
|
|
|
|
| Re: TRIGRID co-linearity problem [message #24999 is a reply to message #24984] |
Wed, 09 May 2001 01:25  |
Roman Schreiber
Messages: 5
Registered: January 2001
|
Junior Member |
|
|
I had a similar problem few weeks ago. Without going into details
I tried a "brute force" approach adding to x and y geographical
coordinates small random parts (about 0.0001 * coordinate value).
Error disappeared and the resulting contour plot was OK.
But it does not of course explain the reason for the peculiar
behavior of trigrid procedure...
Christopher Davey wrote:
>
> The triangulation( ) procedure works just fine, but In my call to
> trigrid( ), I get an error that I cannot figure out: "Points are
> co-linear, no solution."
Regards
Roman Schreiber
schreibe@ncac.torun.pl
|
|
|
|
comp.lang.idl-pvwave archive
Members
Search
Help
Login
Home
