Re: Contour plots

[Henry F. Mollet]

Shai,
I did find an email dated 2/20/05 from you in my Inbox which had the
following 3 files attached:

contourl.cc, contourc.m, and contour.m.

Why cannot I just use your contour.m which calls contourc.m which calls
contourl.cc? Not sure if I can compile an oct file with my version of octave
and capability. If I'm prepared to wait a little longer do I really need an
compiled oct file?
Henry



on 3/7/05 12:05 AM, Shai Ayal at address@hidden wrote:

> Henry,
> 
> Another way of fixing this would be to use another contour algorithm. There
> is one available on the maintainers list. It consists of 3 files -- two
> m-files and one oct file which you'll have to compile yourself using
> mkoctfile. It does not depend on anything else so you can just install it
> in a directory of your choice in your loadpath. This is an altogether
> different algorithm (taken from plplot) and you can get the actual contour
> lines ( i.e. y(x) ) from it.
> 
> I posted it to the maintainers list a few days ago, you can look it up from
> the octave homepage. If you have any problems, contact me.
> 
> Shai
> 
> 
> Henry F. Mollet wrote:
>> Thanks for suggestions from you and Per. I did not say but Per probably knew
>> that I was saving directly as pdf within AquaTerm, albeit an old version. I
>> will have to think about upgrading which appears to be always easier said
>> than actually done and I was hoping to get by with what I have until I buy a
>> new computer running Tiger.
>> 
>> I also take note that my biggest problem may have to do with contour.m
>> itself and I may have to do it the hard way and calculate the contours as y
>> = y(x) and plot those directly. This will also allow me to plot additional
>> "contours" needed.
>> Henry
>> 
>> 
>> 
>> on 3/6/05 12:17 PM, Dmitri A. Sergatskov at address@hidden wrote:
>> 
>> 
>>> Henry F. Mollet wrote:
>>> ...
>>> 
>>>> See attached for the best I was able to come up with. Of *most concern* is
>>>> the jaggedness of the contours as the parameter w/al increases. It does not
>>>> appear to be related to the meshgrid.
>>> 
>>> ...
>>> 
>>> It looks to me that you making a hardcopy of your screen picture with
>>> your screen resolution. Try to make a postscript or pdf figure
>>> directly by setting terminal and an output file.
>>> Unfortunately one cannot use print() command in multiplot mode.
>>> 
>>> I modified your script:
>>> 
>>> gset term post enh color solid "Times" 10
>>> gset out "henry11.ps"
>>> subplot (2,2,4)                                        %  subplot (2,2,4)
>>> gset origin 0.30,-0.07
>>> axis ([0.2,1.2,0,35], "square"); % problems with "normal"
>>> gset label 4 " D)  w/al = 12.0 " at 0.25, 27.5
>>> grid "on"
>>> x=linspace(0,1.2,350);
>>> y=linspace(1,35,350);
>>> [xx,yy]=meshgrid(x,y);
>>> z=xx./(yy.*(1-xx)) -\
>>> (12*yy-yy+1).*xx.^(12*yy-yy+1)./yy./(1-xx.^(12*yy-yy+1)) + eps;
>>> xlabel ("Sa/lambda1")
>>> ylabel ("alpha")
>>> contour (x,y,z)
>>> pause(1)
>>> 
>>> The attached file is epstopdf conversion of the resulting ps file.
>>> The thing *I* do not understand is why ylabel not rotated.
>>> Otherwise it looks good to me.
>>> 
>>> Regards,
>>> 
>>> Dmitri.
>> 
>> 
>> 
>> 
>> -------------------------------------------------------------
>> Octave is freely available under the terms of the GNU GPL.
>> 
>> Octave's home on the web:  http://www.octave.org
>> How to fund new projects:  http://www.octave.org/funding.html
>> Subscription information:  http://www.octave.org/archive.html
>> -------------------------------------------------------------
>> 
>> 



-------------------------------------------------------------
Octave is freely available under the terms of the GNU GPL.

Octave's home on the web:  http://www.octave.org
How to fund new projects:  http://www.octave.org/funding.html
Subscription information:  http://www.octave.org/archive.html
-------------------------------------------------------------