| PLOYY plot [message #60976] |
Wed, 02 July 2008 02:03  |
d.poreh
Messages: 406
Registered: October 2007
|
Senior Member |
|
|
Folks
I want to plot, to set of data in a single figure in IDL (something
like polyy in MATLAB) one set x,y for example with normal axis and
another on (say xx,yy)whit logarithmic axis.
Any help?
Cheers
Dave
|
|
|
|
| Re: PLOYY plot [message #61112 is a reply to message #60976] |
Wed, 02 July 2008 09:47  |
Paul Van Delst[1]
Messages: 1157
Registered: April 2002
|
Senior Member |
|
|
d.poreh@gmail.com wrote:
> On Jul 2, 5:51 pm, Paul van Delst <Paul.vanDe...@noaa.gov> wrote:
>> d.po...@gmail.com wrote:
>>> On Jul 2, 5:14 pm, Brian Larsen <balar...@gmail.com> wrote:
>>>> How is Chris' post not what you want? We all obviously need a little
>>>> more explanation here. What I see in Chris' plot is a "wiggle"
>>>> plotted on the bottom x-axis and a "parabola" plotted on the top x-
>>>> axis.
>>>> Brian
>>>> ------------------------------------------------------------ --------------
>>>> Brian Larsen
>>>> Boston University
>>>> Center for Space Physicshttp://people.bu.edu/balarsen/Home/IDL
>>> Brian
>>> i want upper x also loarithmic. if i change and put *xlog=1* it is not
>>> work. i want something like this:
>>> https://www.rsg.tu-freiberg.de/twiki/pub/Main/DavodPoreh/as. pdf
>>> as you can see y1=f(x) (left and down) is normal and y2=f(x2)(rigth
>>> and up) is logarithmic.
>> Why? The x-axes aren't related -- at least they shouldn't be since the lower one starts at 0.
>>
>> Why plot two disparate datasets on the same figure? Even if it is valid numbers-wise, it
>> will still be confusing. I don't know anything about the data, but it smacks of advanced
>> plotology to me, i.e. displaying data to make it look a certain way. The plot equivalent
>> of the drunk man under the street light statistics story.
>>
>> cheers,
>>
>> paulv
>
> Paul
> if you search in net Plotyy you can see alot about this for example:
> http://www.sgr.nada.kth.se/unix/software/matlab/senaste/tech doc/ref/plotyy.html
> it is very useful in scince.
Oh, I have no argument about the technique in general. I've done similar to that shown in
the matlab example above. My issue (such as it is) was with the example plot you showed.
cheers,
paulv
|
|
|
|
| Re: PLOYY plot [message #61114 is a reply to message #60976] |
Wed, 02 July 2008 09:12 |
d.poreh
Messages: 406
Registered: October 2007
|
Senior Member |
|
|
On Jul 2, 5:58 pm, d.po...@gmail.com wrote:
> On Jul 2, 5:51 pm, Paul van Delst <Paul.vanDe...@noaa.gov> wrote:
>
>
>
>> d.po...@gmail.com wrote:
>>> On Jul 2, 5:14 pm, Brian Larsen <balar...@gmail.com> wrote:
>>>> How is Chris' post not what you want? We all obviously need a little
>>>> more explanation here. What I see in Chris' plot is a "wiggle"
>>>> plotted on the bottom x-axis and a "parabola" plotted on the top x-
>>>> axis.
>
>>>> Brian
>
>>>> ------------------------------------------------------------ --------------
>>>> Brian Larsen
>>>> Boston University
>>>> Center for Space Physicshttp://people.bu.edu/balarsen/Home/IDL
>
>>> Brian
>>> i want upper x also loarithmic. if i change and put *xlog=1* it is not
>>> work. i want something like this:
>>> https://www.rsg.tu-freiberg.de/twiki/pub/Main/DavodPoreh/as. pdf
>>> as you can see y1=f(x) (left and down) is normal and y2=f(x2)(rigth
>>> and up) is logarithmic.
>
>> Why? The x-axes aren't related -- at least they shouldn't be since the lower one starts at 0.
>
>> Why plot two disparate datasets on the same figure? Even if it is valid numbers-wise, it
>> will still be confusing. I don't know anything about the data, but it smacks of advanced
>> plotology to me, i.e. displaying data to make it look a certain way. The plot equivalent
>> of the drunk man under the street light statistics story.
>
>> cheers,
>
>> paulv
>
> Paul
> if you search in net Plotyy you can see alot about this for example:http://www.sgr.nada.kth.se/unix/software/matlab/sena ste/techdoc/ref/p...
> it is very useful in scince.
> Cheers
Brian it works now. thanks for all
for helping. now i want to do that in ITools.
Cheers
|
|
|
|
| Re: PLOYY plot [message #61115 is a reply to message #60976] |
Wed, 02 July 2008 08:58 |
d.poreh
Messages: 406
Registered: October 2007
|
Senior Member |
|
|
On Jul 2, 5:51 pm, Paul van Delst <Paul.vanDe...@noaa.gov> wrote:
> d.po...@gmail.com wrote:
>> On Jul 2, 5:14 pm, Brian Larsen <balar...@gmail.com> wrote:
>>> How is Chris' post not what you want? We all obviously need a little
>>> more explanation here. What I see in Chris' plot is a "wiggle"
>>> plotted on the bottom x-axis and a "parabola" plotted on the top x-
>>> axis.
>
>>> Brian
>
>>> ------------------------------------------------------------ --------------
>>> Brian Larsen
>>> Boston University
>>> Center for Space Physicshttp://people.bu.edu/balarsen/Home/IDL
>
>> Brian
>> i want upper x also loarithmic. if i change and put *xlog=1* it is not
>> work. i want something like this:
>> https://www.rsg.tu-freiberg.de/twiki/pub/Main/DavodPoreh/as. pdf
>> as you can see y1=f(x) (left and down) is normal and y2=f(x2)(rigth
>> and up) is logarithmic.
>
> Why? The x-axes aren't related -- at least they shouldn't be since the lower one starts at 0.
>
> Why plot two disparate datasets on the same figure? Even if it is valid numbers-wise, it
> will still be confusing. I don't know anything about the data, but it smacks of advanced
> plotology to me, i.e. displaying data to make it look a certain way. The plot equivalent
> of the drunk man under the street light statistics story.
>
> cheers,
>
> paulv
Paul
if you search in net Plotyy you can see alot about this for example:
http://www.sgr.nada.kth.se/unix/software/matlab/senaste/tech doc/ref/plotyy.html
it is very useful in scince.
Cheers
|
|
|
|
| Re: PLOYY plot [message #61116 is a reply to message #60976] |
Wed, 02 July 2008 08:51 |
Paul Van Delst[1]
Messages: 1157
Registered: April 2002
|
Senior Member |
|
|
d.poreh@gmail.com wrote:
> On Jul 2, 5:14 pm, Brian Larsen <balar...@gmail.com> wrote:
>> How is Chris' post not what you want? We all obviously need a little
>> more explanation here. What I see in Chris' plot is a "wiggle"
>> plotted on the bottom x-axis and a "parabola" plotted on the top x-
>> axis.
>>
>> Brian
>>
>> ------------------------------------------------------------ --------------
>> Brian Larsen
>> Boston University
>> Center for Space Physicshttp://people.bu.edu/balarsen/Home/IDL
>
> Brian
> i want upper x also loarithmic. if i change and put *xlog=1* it is not
> work. i want something like this:
> https://www.rsg.tu-freiberg.de/twiki/pub/Main/DavodPoreh/as. pdf
> as you can see y1=f(x) (left and down) is normal and y2=f(x2)(rigth
> and up) is logarithmic.
Why? The x-axes aren't related -- at least they shouldn't be since the lower one starts at 0.
Why plot two disparate datasets on the same figure? Even if it is valid numbers-wise, it
will still be confusing. I don't know anything about the data, but it smacks of advanced
plotology to me, i.e. displaying data to make it look a certain way. The plot equivalent
of the drunk man under the street light statistics story.
cheers,
paulv
|
|
|
|
| Re: PLOYY plot [message #61117 is a reply to message #60976] |
Wed, 02 July 2008 08:36 |
d.poreh
Messages: 406
Registered: October 2007
|
Senior Member |
|
|
On Jul 2, 5:14 pm, Brian Larsen <balar...@gmail.com> wrote:
> How is Chris' post not what you want? We all obviously need a little
> more explanation here. What I see in Chris' plot is a "wiggle"
> plotted on the bottom x-axis and a "parabola" plotted on the top x-
> axis.
>
> Brian
>
> ------------------------------------------------------------ --------------
> Brian Larsen
> Boston University
> Center for Space Physicshttp://people.bu.edu/balarsen/Home/IDL
Brian
i want upper x also loarithmic. if i change and put *xlog=1* it is not
work. i want something like this:
https://www.rsg.tu-freiberg.de/twiki/pub/Main/DavodPoreh/as. pdf
as you can see y1=f(x) (left and down) is normal and y2=f(x2)(rigth
and up) is logarithmic.
Cheers
|
|
|
|
| Re: PLOYY plot [message #61118 is a reply to message #60976] |
Wed, 02 July 2008 08:14 |
Brian Larsen
Messages: 270
Registered: June 2006
|
Senior Member |
|
|
How is Chris' post not what you want? We all obviously need a little
more explanation here. What I see in Chris' plot is a "wiggle"
plotted on the bottom x-axis and a "parabola" plotted on the top x-
axis.
Brian
------------------------------------------------------------ --------------
Brian Larsen
Boston University
Center for Space Physics
http://people.bu.edu/balarsen/Home/IDL
|
|
|
|
| Re: PLOYY plot [message #61119 is a reply to message #60976] |
Wed, 02 July 2008 08:08 |
d.poreh
Messages: 406
Registered: October 2007
|
Senior Member |
|
|
On Jul 2, 4:44 pm, Spon <christoph.b...@gmail.com> wrote:
> On Jul 2, 2:23 pm, d.po...@gmail.com wrote:
>
>
>
>> On Jul 2, 2:15 pm, Brian Larsen <balar...@gmail.com> wrote:
>
>>> Does this work?
>
>>> http://www.dfanning.com/tips/another_yaxis.html
>
>>> Try using that site for all you IDL needs, it is a great first (and
>>> last) stop.
>
>>> Cheers,
>
>>> Brian
>
>>> ------------------------------------------------------------ --------------
>>> Brian Larsen
>>> Boston University
>>> Center for Space Physicshttp://people.bu.edu/balarsen/Home/IDL
>
>> Hi Brian
>> I saw that before no it is not what I need. In that image x is same
>> for both y. I want to inter another x (in the top of box) that y=f(x)
>> is for right and top.
>> Cheers
>
> You have to call AXIS twice, once for each new axis you require.
> To extend David's example:
>
> x = FINDGEN(100)
> theta = x/5
> curve = SIN(x/5) / EXP(x/50)
> PLOT, curve, YSTYLE=8
> AXIS, YAXIS=1, YLOG=1, YRANGE=[0.1, 100], /SAVE
> AXIS, XAXIS=1, XRANGE =[0, 200], /SAVE
> OPLOT, [theta, REVERSE(theta)]
>
> If you have the same problem as I do with overlayed axes disappearing
> you could get around it by setting YTHICK = 2 & XTHICK = 2 in your
> respective AXIS calls; but this might be a platform-dependent problem
> that doesn't affect you.
>
> Regards,
> Chris
no
what you had done is just put y1=f(x)&y2=f(x) in a same figure in need
somthing like this:
https://www.rsg.tu-freiberg.de/twiki/pub/Main/DavodPoreh/as. pdf
that means plotting y1=f(x1) & y2=f(x2) in diffrent axis format as you
can see.
cheers
|
|
|
|
comp.lang.idl-pvwave archive
Members
Search
Help
Login
Home
