Skip to content

Add nan fail test#12

Open
xywei wants to merge 2 commits into
inducer:mainfrom
xywei:add-nan-fail-test
Open

Add nan fail test#12
xywei wants to merge 2 commits into
inducer:mainfrom
xywei:add-nan-fail-test

Conversation

@xywei

@xywei xywei commented Aug 4, 2020

Copy link
Copy Markdown
Contributor

If the expansion basis is singular at the origin, and a source point lies at the expansion center, p2e fails with NaNs.
The added test shows that this is indeed an issue with Y2DMultipoleExpansion.

cf.

@inducer

inducer commented Aug 4, 2020

Copy link
Copy Markdown
Owner

Thanks!

cc @isuruf

@isuruf

isuruf commented Aug 4, 2020

Copy link
Copy Markdown
Collaborator

HelmholtzConformingVolumeTaylorMultipoleExpansion works? H2DMultipoleExpansion will not be the default Helmholtz expansion in the near future.

@xywei

xywei commented Aug 4, 2020

Copy link
Copy Markdown
Contributor Author

HelmholtzConformingVolumeTaylorMultipoleExpansion works? H2DMultipoleExpansion will not be the default Helmholtz expansion in the near future.

I think that Taylor expansions are okay because they are smooth. The issue pertains only to basis functions that are singular (like Bessel functions of the second kind).

@inducer

inducer commented Aug 4, 2020

Copy link
Copy Markdown
Owner

I think that Taylor expansions are okay because they are smooth.

Taylor multipoles aren't smooth though. (The multipole basis consists of derivatives of the kernel.)

Base automatically changed from master to main March 8, 2021 05:25
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants