Different specification of parameters; octave and gnuplot

classic Classic list List threaded Threaded
6 messages Options
Reply | Threaded
Open this post in threaded view
|

Different specification of parameters; octave and gnuplot

pathematica
I have no problem, this is not urgent and nothing bad will happen if no one replies!

I am merely interested to know something. It is possible that my confusion is based on missing a key bit of explanation in the manual.

So ... using gnuplot for 3d plots from octave

I note that when specifying a particular view for a 3d plot (for example to align the visual axis with the line through the origin with the homogeneous coordinate on the projective plane P2 [k : k : k] to generate radial symmetry of order three), it is possible to send parameters for azimuth and elevation as a 1x2 vector using view().

As I am sure you will all know, gnuplot reports the elevation and azimuth on the plot. The parameters are reported in that particular order, whereas octave requires the parameters to be specified in the opposite order, ie azimuth then elevation. Also, the octave requires the elevation to be specified as a negative offset from 90 degrees compared with the elevation reported in gnuplot.

Have I understood this correctly and, if so, is there an interesting explanation for the different convention (or, perhaps more accurately, the lack of convention)? Also, is there a reason why the parameters are specified in degrees rather than radians?

As noted at the beginning - this is merely of interest and it is not urgent.

Thank you so much for octave. (Seriously, thank you; - compliments like this will always be a form of litotes and they can never do it justice!).
However good you think Octave is, it's much, much better.
Reply | Threaded
Open this post in threaded view
|

Re: Different specification of parameters; octave and gnuplot

c.-2

On 22 Aug 2010, at 14:03, pathematica wrote:

>
> I have no problem, this is not urgent and nothing bad will happen if  
> no one
> replies!
>
> I am merely interested to know something. It is possible that my  
> confusion
> is based on missing a key bit of explanation in the manual.
>
> So ... using gnuplot for 3d plots from octave
>
> I note that when specifying a particular view for a 3d plot (for  
> example to
> align the visual axis with the line through the origin with the  
> homogeneous
> coordinate on the projective plane P2 [k : k : k] to generate radial
> symmetry of order three), it is possible to send parameters for  
> azimuth and
> elevation as a 1x2 vector using view().
>
> As I am sure you will all know, gnuplot reports the elevation and  
> azimuth on
> the plot. The parameters are reported in that particular order,  
> whereas
> octave requires the parameters to be specified in the opposite  
> order, ie
> azimuth then elevation. Also, the octave requires the elevation to be
> specified as a negative offset from 90 degrees compared with the  
> elevation
> reported in gnuplot.
>
> Have I understood this correctly and, if so, is there an interesting
> explanation for the different convention (or, perhaps more  
> accurately, the
> lack of convention)? Also, is there a reason why the parameters are
> specified in degrees rather than radians?
>
> As noted at the beginning - this is merely of interest and it is not  
> urgent.
>
> Thank you so much for octave. (Seriously, thank you; - compliments  
> like this
> will always be a form of litotes and they can never do it justice!).


Octave tries to be, as far as possible, compatible with Matlab, so the  
order and meaning of the parameters of the 'view' command has most  
likely been chosen to follow those of the same command in Matlab.
HTH,
c.



_______________________________________________
Help-octave mailing list
[hidden email]
https://www-old.cae.wisc.edu/mailman/listinfo/help-octave
Reply | Threaded
Open this post in threaded view
|

Re: Different specification of parameters; octave and gnuplot

bpabbott
Administrator
In reply to this post by pathematica
On Aug 22, 2010, at 8:03 AM, pathematica wrote:

> I have no problem, this is not urgent and nothing bad will happen if no one
> replies!
>
> I am merely interested to know something. It is possible that my confusion
> is based on missing a key bit of explanation in the manual.
>
> So ... using gnuplot for 3d plots from octave
>
> I note that when specifying a particular view for a 3d plot (for example to
> align the visual axis with the line through the origin with the homogeneous
> coordinate on the projective plane P2 [k : k : k] to generate radial
> symmetry of order three), it is possible to send parameters for azimuth and
> elevation as a 1x2 vector using view().
>
> As I am sure you will all know, gnuplot reports the elevation and azimuth on
> the plot. The parameters are reported in that particular order, whereas
> octave requires the parameters to be specified in the opposite order, ie
> azimuth then elevation. Also, the octave requires the elevation to be
> specified as a negative offset from 90 degrees compared with the elevation
> reported in gnuplot.
>
> Have I understood this correctly and, if so, is there an interesting
> explanation for the different convention (or, perhaps more accurately, the
> lack of convention)? Also, is there a reason why the parameters are
> specified in degrees rather than radians?
>
> As noted at the beginning - this is merely of interest and it is not urgent.
>
> Thank you so much for octave. (Seriously, thank you; - compliments like this
> will always be a form of litotes and they can never do it justice!).

The developers of Octave and Gnuplot have some degree of cooperation. However, as Octave has chosen to favor compatibility with Matlab there will necessarily be differences in conventions between Octave and Gnuplot.

Thus, regarding the convention for specifying the axes view property, Octave has adopted the approach used by Matlab, which is different from that of Gnuplot.

Ben


_______________________________________________
Help-octave mailing list
[hidden email]
https://www-old.cae.wisc.edu/mailman/listinfo/help-octave
Reply | Threaded
Open this post in threaded view
|

Re: Different specification of parameters; octave and gnuplot

pathematica
In reply to this post by c.-2
That would make sense. I have practically no experience of using matlab (I grew up with Mathcad with the Open University).

I have just noticed that the mailing list etiquette discourages small talk so this thread might not be a suitable place for questions and discussions of this type. This has been noted and I send my apologies.
However good you think Octave is, it's much, much better.
Reply | Threaded
Open this post in threaded view
|

Re: Different specification of parameters; octave and gnuplot

bpabbott
Administrator
On Aug 22, 2010, at 11:25 AM, pathematica wrote:

> That would make sense. I have practically no experience of using matlab (I
> grew up with Mathcad with the Open University).
>
> I have just noticed that the mailing list etiquette discourages small talk
> so this thread might not be a suitable place for questions and discussions
> of this type. This has been noted and I send my apologies.

The differences in convention between Octave, Matlab, and Gnuplot is a common source of questions.  I expect others are curious about these things as well. No need for apologies.

Ben
_______________________________________________
Help-octave mailing list
[hidden email]
https://www-old.cae.wisc.edu/mailman/listinfo/help-octave
Reply | Threaded
Open this post in threaded view
|

Re: Different specification of parameters; octave and gnuplot

pathematica
In that case, despite the mailing list etiquette, I'll risk saying "thank you"!
However good you think Octave is, it's much, much better.