python-meep team mailing list archive

[Nizamov Shawkat <nizamov.shawkat@xxxxxxxxx>]

It is OK. Actually, I think that your suggestion for making
python-meep consistent with Scheme interface was right. If I recoll
correctly, now only volume classes are strictly with C++ conventions.

wbr
Shawkat

2010/1/21, Emmanuel Lambert <Emmanuel.Lambert@xxxxxxxxxxxxxx>:
> Shavkat,
>
> I would like to point out a discrepancy between the C++
> "gaussian_src_time" and the Scheme "gaussian_src_time".
>
> Both take the same parameters (f, fwidth, s), but define the source with
> different cutoff time.
>
> In fact: let's take as parameters:
> fcen = 0.15
> fwidth = 0.10
> s = 5.0 (the default)
>
> * In Scheme, using the definition from "meep.scm.in", this is translated
> in a call to the C++ function gaussian_src_time(f,t,st,et) as follows:
> gaussian_src_time(0.15, 1/0.1, 0, 2*(1/0.1) * 5.0)
> =  gaussian_src_time(0.15, 10, 0, 100)
> which defines the attributes of the source as follows:
> freq = 0.15
> width = 10
> peak_time = 50
> cutoff = 50
>
> * In C++, the function gaussian_src_time(f,fwidth,s) called with values
> 0.15,0.10 and 5.0 defines the attributes of the source as follows:
> freq = 0.15 (identical)
> width = 1.0 / 0.10 = 10 (identical)
> peak_time = 50 (identical)
> cutoff = 10 * 5.0 * 2 = 100
>
> Therefore I will add to "meep-common.py" a redefinition of
> "gaussian_src_time" that will make the function call consistent with
> Scheme.
>
> Is this OK for you?
>
> PS: I'm currently also working on a Harminv helper function, which will
> be ready very soon.
>
> wbr,
> Emmanuel
>
>
>
> _______________________________________________
> Mailing list: https://launchpad.net/~python-meep
> Post to     : python-meep@xxxxxxxxxxxxxxxxxxx
> Unsubscribe : https://launchpad.net/~python-meep
> More help   : https://help.launchpad.net/ListHelp
>