[GRASS-user] Does v.kernel have to take 16+ hours?

[Aren Cambre]

On Mon, Dec 10, 2012 at 2:03 AM, Moritz Lennert <
mlennert at club.worldonline.be> wrote:

> On 07/12/12 17:16, Markus Metz wrote:
>
>> On Thu, Dec 6, 2012 at 10:19 PM, Aren Cambre<aren at arencambre.com>  wrote:
>>
>>> OK, I think I get it now. After you've hit 4 SDs (or even arguably 3
>>> SDs!)
>>> the contribution of more points to that square fades to almost nothing
>>> without extreme clusters of points.
>>>
>>
>> Right.
>>
>>>
>>> Seems like this ought to be clarified in the UI and documentation?
>>> Should I
>>> file an enhancement request?
>>>
>>
>> I would rather have the option 'stddeviation' renamed to 'radius',
>> kernel radius in map units, which it actually is currently for all
>> kernels but gaussian, and internally adjust standard deviation instead
>> of radius for the gaussian kernel. I guess that the kernel radius is
>> of more importance to users than the standard deviation. Makes sense?
>>
>
>
> +1
>
> Moritz
>

Leave a comment at http://trac.osgeo.org/grass/ticket/1828 to help the
case.

Aren
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/grass-user/attachments/20121210/b3abf5c9/attachment-0001.html>