[SciPy-Dev] New function in scipy.signal.lti to generate Nyquist plot

[Niklas Kröger]

Hi Pierre,

I think you are right. I chose the name because of the nyquist function
from octave, which I used before I had this python version coded. However
since my function doesn't generate the plot but just the values, you are
right.

Regarding the grid size "n":
I was unsure of that value and I just chose 10^4 based on the plots I
generated with the function. For simple functions like the example this
number is very high. However I ran into some functions which could not be
plotted precisely because of too few generated values. I will append an
example of what I mean (only showing the plot for w->+inf to make the plot
less crowded).
[image: Inline-Bild 1]
[image: Inline-Bild 2]
As you can see n=10^4 is still not enough to make a really nice and round
plot. But in my opinion it's close enough to see the actual plot. Of course
you could argue, that anyone who generates plots of such complex transfer
functions, could just set the value of n up himself. For me it's not
important what default n is set to, 10^4 was just the first value that
seemed to fit my needs.

Anyway, I will change the name of the function as you said and adjust the
docstring.

Best regards,
-Niklas.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scipy-dev/attachments/20130208/128f2e6c/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image.png
Type: image/png
Size: 30138 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/scipy-dev/attachments/20130208/128f2e6c/attachment.png>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image.png
Type: image/png
Size: 32481 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/scipy-dev/attachments/20130208/128f2e6c/attachment-0001.png>