Dear all,
I am trying to create a simple parfile to help a PhD student. The goal is to set up puncture initial data with NRPyEllipticET and then use QuasiLocalMeasures/AHFinderDirect to compute properties of the punctures and compare with the expected results. No evolution thorn is expected/required and the simulation ends after the initial data has been set and the first round of analysis has finished.
The parfile attached partially works. The puncture initial data is set up and NRPyEllipticET's output is what I'd expect. However, the diagnostics are clearly wrong since: 1) AHFinderDirect fails to find the puncture (in this example I am setting the mass of one of the punctures to zero); and 2) the QuasiLocalMeasures diagnostics is never triggered, despite being scheduled.
Could someone provide some guidance on how to set up these thorns correctly, specifically SphericalSurfaces, QuasiLocalMeasures, and AHFinderDirect? Am I missing additional thorns?
Also, if someone has a working TwoPuncture example parfile instead, please share it with me as I'd be happy to adapt it!
Thanks in advance! Leo
Leonardo
I've briefly looked at the parameter file. The horizon finder needs to know an initial guess for both position and shape of the horizon. You are setting this location to 0, but are placing the puncture at x=5. The horizon finder then fails. QuasiLocalMeasures then sees that there is no horizon and does not run.
-erik
On Wed, Jun 28, 2023 at 3:50 PM Leonardo Werneck wernecklr@gmail.com wrote:
Dear all,
I am trying to create a simple parfile to help a PhD student. The goal is to set up puncture initial data with NRPyEllipticET and then use QuasiLocalMeasures/AHFinderDirect to compute properties of the punctures and compare with the expected results. No evolution thorn is expected/required and the simulation ends after the initial data has been set and the first round of analysis has finished.
The parfile attached partially works. The puncture initial data is set up and NRPyEllipticET's output is what I'd expect. However, the diagnostics are clearly wrong since: 1) AHFinderDirect fails to find the puncture (in this example I am setting the mass of one of the punctures to zero); and 2) the QuasiLocalMeasures diagnostics is never triggered, despite being scheduled.
Could someone provide some guidance on how to set up these thorns correctly, specifically SphericalSurfaces, QuasiLocalMeasures, and AHFinderDirect? Am I missing additional thorns?
Also, if someone has a working TwoPuncture example parfile instead, please share it with me as I'd be happy to adapt it!
Thanks in advance! Leo
--
Dr. Leonardo R. Werneck, Ph.D. _______________________________________________ Users mailing list Users@einsteintoolkit.org http://lists.einsteintoolkit.org/mailman/listinfo/users
Hi Leo,
I actually have the same test for the FUKA importer using a single BH. The par can be found with the KadathImporter thorn or directly here: https://bitbucket.org/fukaws/kadathimporter/src/master/test/bhtest.par
Cheers,
Samuel
-------------------------------- Postdoctoral Researcher Institue for Theoretical Physics Goethe University Frankfurt https://relastro.uni-frankfurt.de/
On 6/28/23 10:49 PM, Erik Schnetter wrote:
Leonardo
I've briefly looked at the parameter file. The horizon finder needs to know an initial guess for both position and shape of the horizon. You are setting this location to 0, but are placing the puncture at x=5. The horizon finder then fails. QuasiLocalMeasures then sees that there is no horizon and does not run.
-erik
On Wed, Jun 28, 2023 at 3:50 PM Leonardo Werneck wernecklr@gmail.com wrote:
Dear all,
I am trying to create a simple parfile to help a PhD student. The goal is to set up puncture initial data with NRPyEllipticET and then use QuasiLocalMeasures/AHFinderDirect to compute properties of the punctures and compare with the expected results. No evolution thorn is expected/required and the simulation ends after the initial data has been set and the first round of analysis has finished.
The parfile attached partially works. The puncture initial data is set up and NRPyEllipticET's output is what I'd expect. However, the diagnostics are clearly wrong since: 1) AHFinderDirect fails to find the puncture (in this example I am setting the mass of one of the punctures to zero); and 2) the QuasiLocalMeasures diagnostics is never triggered, despite being scheduled.
Could someone provide some guidance on how to set up these thorns correctly, specifically SphericalSurfaces, QuasiLocalMeasures, and AHFinderDirect? Am I missing additional thorns?
Also, if someone has a working TwoPuncture example parfile instead, please share it with me as I'd be happy to adapt it!
Thanks in advance! Leo
--
Dr. Leonardo R. Werneck, Ph.D. _______________________________________________ Users mailing list Users@einsteintoolkit.org http://lists.einsteintoolkit.org/mailman/listinfo/users
He Erik,
Thanks for the response!
The puncture is placed at x=5 because of the nature of the grid used by the elliptic solver. However, when the solution is interpolated to the carpet grid the position of the puncture is shifted to x=0 (this is controlled by the parameter NRPyEllipticET::position_shift[0]; please see the attached plot of the lapse function at t=0).
Cheers, Leo
[image: image.png]
On Wed, Jun 28, 2023 at 1:49 PM Erik Schnetter schnetter@gmail.com wrote:
Leonardo
I've briefly looked at the parameter file. The horizon finder needs to know an initial guess for both position and shape of the horizon. You are setting this location to 0, but are placing the puncture at x=5. The horizon finder then fails. QuasiLocalMeasures then sees that there is no horizon and does not run.
-erik
On Wed, Jun 28, 2023 at 3:50 PM Leonardo Werneck wernecklr@gmail.com wrote:
Dear all,
I am trying to create a simple parfile to help a PhD student. The goal
is to set up puncture initial data with NRPyEllipticET and then use QuasiLocalMeasures/AHFinderDirect to compute properties of the punctures and compare with the expected results. No evolution thorn is expected/required and the simulation ends after the initial data has been set and the first round of analysis has finished.
The parfile attached partially works. The puncture initial data is set
up and NRPyEllipticET's output is what I'd expect. However, the diagnostics are clearly wrong since: 1) AHFinderDirect fails to find the puncture (in this example I am setting the mass of one of the punctures to zero); and 2) the QuasiLocalMeasures diagnostics is never triggered, despite being scheduled.
Could someone provide some guidance on how to set up these thorns
correctly, specifically SphericalSurfaces, QuasiLocalMeasures, and AHFinderDirect? Am I missing additional thorns?
Also, if someone has a working TwoPuncture example parfile instead,
please share it with me as I'd be happy to adapt it!
Thanks in advance! Leo
--
Dr. Leonardo R. Werneck, Ph.D. _______________________________________________ Users mailing list Users@einsteintoolkit.org http://lists.einsteintoolkit.org/mailman/listinfo/users
-- Erik Schnetter schnetter@gmail.com http://www.perimeterinstitute.ca/personal/eschnetter/
Leonardo
In this case the initial guess for the horizon radius might be off. As a rule of thumb, the surface where \alpha = 0.3 is the horizon. Your guess of r = 1 might be off.
-erik
On Wed, Jun 28, 2023 at 4:58 PM Leonardo Werneck wernecklr@gmail.com wrote:
He Erik,
Thanks for the response!
The puncture is placed at x=5 because of the nature of the grid used by the elliptic solver. However, when the solution is interpolated to the carpet grid the position of the puncture is shifted to x=0 (this is controlled by the parameter NRPyEllipticET::position_shift[0]; please see the attached plot of the lapse function at t=0).
Cheers, Leo
[image: image.png]
On Wed, Jun 28, 2023 at 1:49 PM Erik Schnetter schnetter@gmail.com wrote:
Leonardo
I've briefly looked at the parameter file. The horizon finder needs to know an initial guess for both position and shape of the horizon. You are setting this location to 0, but are placing the puncture at x=5. The horizon finder then fails. QuasiLocalMeasures then sees that there is no horizon and does not run.
-erik
On Wed, Jun 28, 2023 at 3:50 PM Leonardo Werneck wernecklr@gmail.com wrote:
Dear all,
I am trying to create a simple parfile to help a PhD student. The goal
is to set up puncture initial data with NRPyEllipticET and then use QuasiLocalMeasures/AHFinderDirect to compute properties of the punctures and compare with the expected results. No evolution thorn is expected/required and the simulation ends after the initial data has been set and the first round of analysis has finished.
The parfile attached partially works. The puncture initial data is set
up and NRPyEllipticET's output is what I'd expect. However, the diagnostics are clearly wrong since: 1) AHFinderDirect fails to find the puncture (in this example I am setting the mass of one of the punctures to zero); and 2) the QuasiLocalMeasures diagnostics is never triggered, despite being scheduled.
Could someone provide some guidance on how to set up these thorns
correctly, specifically SphericalSurfaces, QuasiLocalMeasures, and AHFinderDirect? Am I missing additional thorns?
Also, if someone has a working TwoPuncture example parfile instead,
please share it with me as I'd be happy to adapt it!
Thanks in advance! Leo
--
Dr. Leonardo R. Werneck, Ph.D. _______________________________________________ Users mailing list Users@einsteintoolkit.org http://lists.einsteintoolkit.org/mailman/listinfo/users
-- Erik Schnetter schnetter@gmail.com http://www.perimeterinstitute.ca/personal/eschnetter/
users@lists.einsteintoolkit.org
-
Erik Schnetter -
Leonardo Werneck -
Samuel Tootle