Hmmm... it works for me. Can you send me the cube (fred.sdf) that you are using?

David


On 13 October 2010 14:53, Jane Buckle <[log in to unmask]> wrote:
> Hi David,
>
> This doesn't work for me:
>
> appcd:buckle> setaxis fred dim=3 mode=wcs
> appcd:buckle> maths exp='ia*cc+(ca+cb)*0' ia=fred out=fred2
> Segmentation fault
>
> I'm trying on a small cube, its only 11x11x25, with RA,DEC and velocity as
> axis1, axis2 and axis3 respectively.
>
> maths exp='ia*cc' ia=fred out=fred2
>
> does finish without a segmentation fault, but I don't think it gives me what
> I'm after.
>
> Regards,
>
> Jane
>
> David Berry wrote:
>>
>> Hi Jane,
>>              You should be able to use KAPPA:MATHS for this. Maths
>> allows you to create an NDF using an arbitrary algebraic combination
>> of the data, variance or coordinate values from other NDFs.  However
>> it's a bit out of date in that it has not been updated to use the
>> modern WCS system - it still uses the old AXIS system. So the first
>> thing you need to do is to create an AXIS structure describing the
>> velocity information in the WCS component:
>>
>> % setaxis fred dim=3 mode=wcs
>>
>> (assuming your velocity axis is axis 3). Then you run maths:
>>
>> % maths exp='ia*cc+(ca+cb)*0' ia=fred out=fred2
>>
>> The EXP parameter gives the algebraic expression to be stored in the
>> output NDF. In this expression, the "ia" token represents the data
>> value in the NDF associated with parameter "ia" (e.g. fred.sdf in this
>> case), and the "c*" tokens represent the coordinate values. The
>> alphabetically earliest "c*" token in the expression represents the
>> coordinate on the first axis, etc. This is why you need to include
>> "ca" and "cb" in the expression, as otherwise "cc" would be the
>> alphabetically earliest "c*" token in the expression and so would give
>> the coordinate value on the first axis, rather than the required third
>> axis. "ca" and "cb" are multiplied by zero in the expression so that
>> they do not affect the values stored in the output NDF.
>>
>> A bit convoluted I know. It would have been simpler for the meaning of
>> each "c*" token to be fixed - ca for the first axis, cb for the second
>> etc. But it's history now...
>>
>> David
>>
>>
>>
>>
>>
>>
>> On 20 August 2010 14:08, Jane Buckle <[log in to unmask]> wrote:
>>>
>>> Hi All,
>>>
>>> The help for kappa collapse with estimator iwc states "Iwc" --
>>> Intensity-weighted co-ordinate, being the sum of each value times its
>>> co-ordinate, all divided by the integrated value (see the "Integ"
>>> option).
>>>
>>> How does it do this? I would like to create a 3D cube, the same size as
>>> my
>>> original cube, but whose data values are the intensity multiplied by the
>>> velocity coordinate for each velocity channel.
>>>
>>> So, basically I would like each value times its coordinate, without any
>>> summing. This may well be a simple task, but the method is currently
>>> escaping me.
>>>
>>> cheers,
>>>
>>> Jane
>>> --
>>> Dr Jane V. Buckle
>>> Cavendish Astrophysics Group and Kavli Institute for Cosmology,
>>> Institute of Astronomy, University of Cambridge, Madingley Road,
>>> Cambridge
>>> CB3 0HE
>>> Tel: +44 (0)1223 746436
>>>
>
> --
> Dr Jane V. Buckle
> Cavendish Astrophysics Group and Kavli Institute for Cosmology,
> Institute of Astronomy, University of Cambridge, Madingley Road, Cambridge
> CB3 0HE
> Tel: +44 (0)1223 746436
>