Re: [sage-devel] Re: spherical harmonics still broken in 9.5.beta8
Hi Eric, On Mon, Mar 14, 2022 at 12:58:26PM -0700, Eric Gourgoulhon wrote: > The branch of the ticket https://trac.sagemath.org/ticket/33117 has been > merged in Sage 9.6.beta5, so in Sage 9.6 spherical harmonics will agree > with those of SymPy, SciPy, Mathematica and Wikipedia, and will have > correct derivatives. There remains the issue of simplifying some > sqrt(sin(theta)^2) terms which appear for odd orders m. This is now > https://trac.sagemath.org/ticket/33501. This is a big improvement already -- thanks a lot for all your efforts on this (and SageManifolds in general)! All the best, keep safe and COVID-free, -- Jonathan -- You received this message because you are subscribed to the Google Groups "sage-devel" group. To unsubscribe from this group and stop receiving emails from it, send an email to sage-devel+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/sage-devel/YjA700OWMUb6WEAr%40gold.bkis-orchard.net.
[sage-devel] Re: spherical harmonics still broken in 9.5.beta8
Hi, The branch of the ticket https://trac.sagemath.org/ticket/33117 has been merged in Sage 9.6.beta5, so in Sage 9.6 spherical harmonics will agree with those of SymPy, SciPy, Mathematica and Wikipedia, and will have correct derivatives. There remains the issue of simplifying some sqrt(sin(theta)^2) terms which appear for odd orders m. This is now https://trac.sagemath.org/ticket/33501. Eric. Le mercredi 5 janvier 2022 à 09:14:23 UTC+1, Eric Gourgoulhon a écrit : > Le mercredi 5 janvier 2022 à 08:27:56 UTC+1, Eric Gourgoulhon a écrit : > >> >> Actually, the difference between the two results is essentially due to a >> different convention in the Condon-Shortley phase >> (cf. >> https://en.wikipedia.org/wiki/Spherical_harmonics#Condon%E2%80%93Shortley_phase >> ), >> which makes Sage's spherical harmonics Y_l^m differ from Wikipedia and >> Mathematica ones by a factor (-1)^m. >> The other difference in the above example is a lack of simplification of >> sqrt(sin(theta)^2). >> >> I would vote for including the Condon-Shortley phase in Sage's spherical >> harmonics, since this is standard in quantum mechanics and this would make >> Sage agree with Wikipedia and Mathematica. >> > > I've opened > https://trac.sagemath.org/ticket/33117 > for this. > > In doing so, I've noticed that current Sage's spherical harmonics > disagree with SymPy as well. > I've also found a very serious bug in the computation of derivatives of > spherical harmonics (see the ticket for details). This has not been seen > earlier probably because before https://trac.sagemath.org/ticket/25034 > (merged in Sage 9.3), spherical harmonics were basically not usable in > Sage. > > Eric. > -- You received this message because you are subscribed to the Google Groups "sage-devel" group. To unsubscribe from this group and stop receiving emails from it, send an email to sage-devel+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/sage-devel/3034012d-c5a8-4626-99a1-9944fc264234n%40googlegroups.com.
Re: [sage-devel] Re: spherical harmonics still broken in 9.5.beta8
Le jeudi 6 janvier 2022 à 06:20:28 UTC+1, Jonathan Thornburg a écrit : > I think this problem is worse than "just" a lack of simplification: > if sin(theta) < 0 then sqrt(sin(theta)^2 != sin(theta), i.e., the > theta dependence is wrong, not "just" not-fully-simplified. > > Well, the standard spherical coordinate theta (the colatitude on S^2) lies in [0, pi], so that sin(theta) >= 0. This simplification issue should be dealt either in https://trac.sagemath.org/ticket/33117 (since it seems pretty easy to implement), or in a follow up ticket. Eric. -- You received this message because you are subscribed to the Google Groups "sage-devel" group. To unsubscribe from this group and stop receiving emails from it, send an email to sage-devel+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/sage-devel/261963d4-32c8-4316-aff8-67b66f0c79een%40googlegroups.com.
Re: [sage-devel] Re: spherical harmonics still broken in 9.5.beta8
On Wed, Jan 05, 2022 at 12:14:23AM -0800, Eric Gourgoulhon wrote: > The other difference in the above example is a lack of simplification of > sqrt(sin(theta)^2). I think this problem is worse than "just" a lack of simplification: if sin(theta) < 0 then sqrt(sin(theta)^2 != sin(theta), i.e., the theta dependence is wrong, not "just" not-fully-simplified. Should we have another ticket this? Eric also wrote: > Actually, the difference between the two results is essentially due to a > different convention in the Condon-Shortley phase > (cf. > https://en.wikipedia.org/wiki/Spherical_harmonics#Condon%E2%80%93Shortley_phase > ), > which makes Sage's spherical harmonics Y_l^m differ from Wikipedia and > Mathematica ones by a factor (-1)^m. [[...]] > I would vote for including the Condon-Shortley phase in Sage's spherical > harmonics, since this is standard in quantum mechanics and this would make > Sage agree with Wikipedia and Mathematica. +1 on this. > I've opened > https://trac.sagemath.org/ticket/33117 > for this. -- -- "Jonathan Thornburg [remove color- to reply]" on the west coast of Canada, eh? "There was of course no way of knowing whether you were being watched at any given moment. How often, or on what system, the Thought Police plugged in on any individual wire was guesswork. It was even conceivable that they watched everybody all the time." -- George Orwell, "1984" -- You received this message because you are subscribed to the Google Groups "sage-devel" group. To unsubscribe from this group and stop receiving emails from it, send an email to sage-devel+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/sage-devel/YdZ3W6iYQDLa7EaA%40gold.bkis-orchard.net.
[sage-devel] Re: spherical harmonics still broken in 9.5.beta8
Le mercredi 5 janvier 2022 à 08:27:56 UTC+1, Eric Gourgoulhon a écrit : > > Actually, the difference between the two results is essentially due to a > different convention in the Condon-Shortley phase > (cf. > https://en.wikipedia.org/wiki/Spherical_harmonics#Condon%E2%80%93Shortley_phase > ), > which makes Sage's spherical harmonics Y_l^m differ from Wikipedia and > Mathematica ones by a factor (-1)^m. > The other difference in the above example is a lack of simplification of > sqrt(sin(theta)^2). > > I would vote for including the Condon-Shortley phase in Sage's spherical > harmonics, since this is standard in quantum mechanics and this would make > Sage agree with Wikipedia and Mathematica. > I've opened https://trac.sagemath.org/ticket/33117 for this. In doing so, I've noticed that current Sage's spherical harmonics disagree with SymPy as well. I've also found a very serious bug in the computation of derivatives of spherical harmonics (see the ticket for details). This has not been seen earlier probably because before https://trac.sagemath.org/ticket/25034 (merged in Sage 9.3), spherical harmonics were basically not usable in Sage. Eric. -- You received this message because you are subscribed to the Google Groups "sage-devel" group. To unsubscribe from this group and stop receiving emails from it, send an email to sage-devel+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/sage-devel/3772782b-8f69-4904-9770-ceb427249414n%40googlegroups.com.
[sage-devel] Re: spherical harmonics still broken in 9.5.beta8
Hi,
Le mercredi 5 janvier 2022 à 00:36:11 UTC+1, Jonathan Thornburg a écrit :
> Sage has long had problems with spherical harmonics, e.g., this thread
> from June 2019:
> https://groups.google.com/g/sage-support/c/I_d_meMxRbM/m/Esxo5UO2BAAJ
>
> As of 9.5.beta8, spherical harmonics are (still) broken for some
> arguments,
> with the test case noted in that earlier thread still giving the same
> (wrong) result:
> sage: theta,phi = var('theta,phi')
> sage: spherical_harmonic(1,1,theta,phi)
> 1/4*sqrt(3)*sqrt(2)*sqrt(sin(theta)^2)*e^(I*phi)/sqrt(pi)
> The correct result would be
> -1/4*sqrt(6)*e^(I*phi)*sin(theta)/sqrt(pi)
> (see, e.g., https://en.wikipedia.org/wiki/Table_of_spherical_harmonics).
>
>
Actually, the difference between the two results is essentially due to a
different convention in the Condon-Shortley phase
(cf.
https://en.wikipedia.org/wiki/Spherical_harmonics#Condon%E2%80%93Shortley_phase),
which makes Sage's spherical harmonics Y_l^m differ from Wikipedia and
Mathematica ones by a factor (-1)^m.
The other difference in the above example is a lack of simplification of
sqrt(sin(theta)^2).
I would vote for including the Condon-Shortley phase in Sage's spherical
harmonics, since this is standard in quantum mechanics and this would make
Sage agree with Wikipedia and Mathematica.
Eric.
--
You received this message because you are subscribed to the Google Groups
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to sage-devel+unsubscr...@googlegroups.com.
To view this discussion on the web visit
https://groups.google.com/d/msgid/sage-devel/b7f40005-d3f0-4289-a51d-7b7c216c85fen%40googlegroups.com.
