| Author |
 Topic  |
|
|
lishoujun
China
1 Posts |
 Posted - 01/06/2012 : 06:07:29 AM
|
I found a bug of Origin Pro 8.1.
draw polar graph, rho = sin(2*theta)*cos(2*theta);
method 1:
--use MATLAB:
program:
theta = 0:0.01:2*pi;
rho = sin(2*theta).*cos(2*theta);
polar(theta,rho)
result is:
obvious, its correct!
--------------
method 2:
--use Origin Pro 8.1
program is:
result is:
obvious, its wrong!
please correct it !!!!
|
|
|
Drbobshepherd
USA
Posts |
Posted - 01/06/2012 : 11:12:35 AM
|
| This is not a function bug. It is a bug in the Origin polar plotting routine. Origin cannot plot polar coordinates when the radius is negative. Users should be made aware of this limitation, and OriginLab should correct it. |
 |
|
|
Hideo Fujii
USA
1582 Posts |
Posted - 01/06/2012 : 3:39:38 PM
|
Hi,
If I remember correctly, some years ago there was a discussion whether Origin should plot the point with a negative radius at the opposite position symmetric to the center of the circle, or not.
If we do so, the graph would show a kind of phantom points - for example, there is no such actual peak at x=70, y=0.5 (approx.) in the above sample graph. It's misleading. (See how the Y axis show the tick labels.)
On the other hand, if we don't do so, user CAN easily plot the absolute values (e.g., abs(sin(2*x)*cos(2*x)) ) if he or she wants to show.
So, my opinion is that it is a matter of design how it works - not either is right, or wrong.
--Hideo Fujii
OriginLab |
Edited by - Hideo Fujii on 01/06/2012 3:42:42 PM |
|
|
|
Drbobshepherd
USA
Posts |
Posted - 01/06/2012 : 4:33:21 PM
|
| Ah, but how Origin currently handles polar data with negative radii IS wrong! Origin ignores it, leading users to think their data with negative radii are displayed. Some users have a legitimate need to plot points with negative angles and negative radii on polar plots. These users shouldn't have to test their polar data and make it conform to Origin limitations. Origin should be intellegent enough to display the data points. MATLAB has it right. The Origin polar plot routine IS flawed! Who knows how many people trusted Origin to display their polar data correctly? |
|
|
|
Sam Fang
292 Posts |
Posted - 01/20/2012 : 03:59:33 AM
|
Hi Drbobshepherd,
Thanks for your suggestion.
We have added it to our bug tracking database: ID ORG-1794's P2. Origin will support it in the future.
Thanks.
Sam
OriginLab Technical Services |
|
|
|
Drbobshepherd
USA
Posts |
Posted - 07/20/2012 : 6:07:41 PM
|
Sam,
I typed in the bug ID you gave into the OriginLab Tracker website and got no match. What is the deal? |
|
|
|
greg
USA
1378 Posts |
Posted - 07/23/2012 : 4:10:02 PM
|
Bugs may be tracked only when they have been cleared and this one has not been cleared because there is no bug.
There is a rather obscure option to allow display of negative values in the opposite quadrant - which is what is needed here.
Open the Script or Command Window and type
@YN=1
and press Enter.
Refresh the graph. |
|
|
|
Drbobshepherd
USA
Posts |
Posted - 07/24/2012 : 12:16:32 PM
|
I'm glad Origin has a way to draw polar plots with negative radii, but the point I am trying to make is I think this option should be the default.
In all other plot types, Origin automatically scales the initial plot to show all your data. But in the case of Polar plots, by default, the points with negative radii are treated as out of range. I had no idea that some of my data was going undisplayed.
I can only think of one case in which I would not want my negative radius values to be flipped to the opposite quadrant-- log-polar plots in which the radial scale goes from a negative number to a positive one. But I still want to see all my data shown in the initial plot. (BTW, can Origin do log-polar plots?)
Now, I know that Origin may not show all my data in the initial polar plot; but, I'll bet that most Origin users do not know this, and that some have turned in faulty plots because of this obscure fact. |
|
|
|
Sam Fang
292 Posts |
Posted - 07/25/2012 : 04:38:10 AM
|
Thanks for your suggestions, we created two items in our bug tracking database for them:
1. Provide an option in the dialog to show negative radius points in the opposite quadrant.
2. Support to change From value of Y axis in Polar Function Plot, not just fixed to 0.
As you said, sometimes people may want to show negative radius instead of postive radius in the opposite quadrant, e.g. antenna radiation pattern data (Rhos) in dB (decibels) vs angle (degrees). And in this case @YB=1 is not proper.
Origin's polar function plot doesn't support negative radius because its From value of Y is fixed to 0. However you can use data in the worksheet to plot them, select Plot: Specialized: Polar theta(X) r(Y) or Polar r(X) theta(Y), and choose Use Y-Axis Range in the polar graph. And set the scale for Y axis. Negative radius in Y axis can be shown.
Sam
OriginLab Technical Services |
Edited by - Sam Fang on 07/25/2012 04:46:59 AM |
|
|
|
Drbobshepherd
USA
Posts |
Posted - 07/25/2012 : 11:16:28 AM
|
Sam,
Thank you for responding, but I still feel you have not addressed the main point I am trying to make-- simply put, I believe @YN=1 should be the default condition for polar plots.
I believe I speak for many users like myself who rely on Origin to automatically show all points in the dataset in that initial plot. Only after Origin shows me all my data, should I select options to reduce the number of points plotted (such as scale changes or selecting @YN=0). It's wrong for Origin to do this for me, and without any indication, and as the default condition.
I like Origin. It is a powerful analytical tool, and I use it often. This is why I would like to see it improved.
I like the changes you suggested, but I would like it better if Item 1 was an option to turn off Quadrant-flipping, which in my opinion should be the default condition.
DrBobShepherd |
|
|
|
Drbobshepherd
USA
Posts |
Posted - 07/25/2012 : 12:19:44 PM
|
P.S.
An antenna pattern was one of the cases I had in mind when I wrote about log-polar plots being the only condition I could think of where @YN=0 should apply. Typical y-values vary from negative dB values to positive ones; the center point of the plot would be a negative value, the outer-most circle would be positive, and zero would correspond to an inner circle. Negative values should plot continuously across the zero-cricle, not flip to the opposite quadrant in this case.
But, Origin doesn't currently handle polar plots with a negative center. In fact, it doesn't allow for the center to be anything but zero, hence the need for change Item 2 (which BTW I am greatly in favor of). So, why is Q-flipping not the default? There is no posibility of ambiguity here. With the center fixed at 0, (30 degrees, -1.2 volts) can only be plotted at one position.
Perhaps the people at OriginLab assume negative radius values are rare. I can assure you, they occur quite frequently. For example, the spectrum from a FTIR spectrometer (every chem lab in the world has one) could vary in magnitude from -5V to +5V, and the phase from -90 degrees to +90 degrees. The FT tool in Origin often generates negative magnitudes. It is ironic that Origin can't plot them as a function of phase, unless the user remembers to make the obscure change @YN=1.
To sum it up: I can't always remember to set @YN=1, and I bet a most users don't even know about @YN. So why not set @YN=1 as the default? It would be more user friendly.
DrBobShepherd |
|
|
|
Sam Fang
292 Posts |
Posted - 07/26/2012 : 06:53:56 AM
|
Hi DrBobShepherd,
Thanks for your explanation.
What @YN=1 does is only to show negative radius points in the opposite quadrant when From value of Y scale is 0, and it doesn't take effect for other From values of Y scale.
In fact Origin supports a negative center in polar plot when you use data points to plot. It just doesn't support in Polar Function Plot.
For the suggestion setting @YN=1 by default, we will discuss it and decide whether to make this change.
Thanks.
Sam
OriginLab Technical Services |
|
|
|
Drbobshepherd
USA
Posts |
Posted - 07/26/2012 : 10:00:40 AM
|
Once again, thank you for responding. And thank you for considering my suggestion. And my apologies for inaccurately stating that Origin can't handle negative centers in its polar plots; I stand corrected.
DrBobShepherd |
|
|
|
attomaz
1 Posts |
Posted - 09/21/2012 : 12:37:47 PM
|
Dear People?!.
I need help in plot data OriginPro 8 in polar coor. Details is antennas radiation diagram in polar plot. I use of the CST MWS simulations, and need reproductions this data using Origin. Please, some people help? Rgs. |
|
|
|
Sam Fang
292 Posts |
Posted - 09/24/2012 : 02:14:06 AM
|
Origin8.0SR6 can plot antennas radiation diagram using default setting.
You need two columns, one column for angle (Unit: Degree), the other column for radiation strength (Unit: dB).
Then select Plot: Specialized: Polar theta(X) r(Y) or Plot: Specialized: Polar r(X) theta(Y) from Origin menu.
Sam
OriginLab Technical Services |
Edited by - Sam Fang on 09/24/2012 07:15:45 AM |
|
|
|
Drbobshepherd
USA
Posts |
Posted - 01/09/2013 : 11:10:02 AM
|
Cheers to OriginLab for fixing the polar-plot bug in the latest release, Origin 9.0 SR1! Now, Origin will plot any polar data point, not just the ones with positive radius values, as the default condition. This is a good fix for what I considered a major short-coming in Origin.
Thank you,
DrBobShepherd |
|
|
| |
Topic |
|
|
|
fuction Bug of Origin Pro 8