pith. sign in

arxiv: 2606.17636 · v1 · pith:IR2FK7V3new · submitted 2026-06-16 · 🌌 astro-ph.IM · astro-ph.EP· astro-ph.GA· astro-ph.HE· astro-ph.SR

A fast spectral-multigrid Poisson solver in non-Cartesian geometries

Ankush Mandal , Oliver Gressel , Udo Ziegler , Andrea Mignone This is my paper

Pith reviewed 2026-06-26 22:57 UTC · model grok-4.3

classification 🌌 astro-ph.IM astro-ph.EPastro-ph.GAastro-ph.HEastro-ph.SR
keywords Poisson solvermultigrid methodFourier decompositionspherical coordinatescylindrical coordinatesself-gravityastrophysical simulations
0
0 comments X

The pith

Spectral decomposition plus multigrid solves the Poisson equation to second order in spherical and cylindrical geometries.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper develops a Poisson solver for three-dimensional non-Cartesian grids by applying azimuthal Fourier decomposition to convert the 3D equation into independent 2D Helmholtz equations, each solved by a multigrid algorithm constructed to preserve geometric consistency and second-order accuracy on both uniform and logarithmic radial meshes. This matters for astrophysical fluid dynamics because self-gravity calculations must handle large radial dynamic ranges, open boundaries, and non-axisymmetric mass distributions without sacrificing either accuracy or parallel scalability. Vacuum boundaries are treated with a screening-mass technique, and the method is implemented inside the PLUTO code. Validation on analytic solutions and dynamical tests confirms the expected convergence, while weak-scaling runs to 4096 cores show the gravity solve remains cheaper than the magnetohydrodynamic update.

Core claim

The central claim is that azimuthal Fourier decomposition reduces the 3D Poisson equation to a set of 2D Helmholtz problems that can be solved by a geometrically consistent multigrid algorithm preserving second-order accuracy on both uniform and non-uniform (logarithmic) radial grids, with vacuum boundaries handled via a screening-mass approach, yielding an efficient, scalable solver for astrophysical self-gravity.

What carries the argument

azimuthal Fourier decomposition combined with a geometrically consistent multigrid algorithm applied mode-by-mode to the resulting Helmholtz equations

If this is right

  • Second-order convergence holds for both spherical and cylindrical geometries on uniform and stretched grids.
  • The Poisson solve remains subdominant to the magnetohydrodynamic update even at 4096 cores.
  • Vacuum boundaries and inner cavities are treated accurately without artificial reflections.
  • The solver supports large-scale simulations of star formation, accretion disks, and gravitational instabilities.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The faster convergence of higher Fourier modes could be exploited to truncate or adaptively refine the mode set for additional speed.
  • Integration with adaptive mesh refinement would allow the same accuracy on even larger dynamic ranges.
  • The same decomposition-plus-multigrid structure might apply to other elliptic operators or coordinate systems beyond the two tested here.

Load-bearing premise

The geometrically consistent multigrid algorithm preserves second-order accuracy on both uniform and non-uniform logarithmic grids when applied to the Helmholtz equations that result from the azimuthal Fourier decomposition.

What would settle it

A convergence test on a logarithmic radial grid with an analytic vacuum-boundary solution in which the measured error fails to decrease quadratically with grid spacing would disprove the accuracy claim.

Figures

Figures reproduced from arXiv: 2606.17636 by Andrea Mignone, Ankush Mandal, Oliver Gressel, Udo Ziegler.

Figure 1
Figure 1. Figure 1: Schematic diagrams of the area-weighted grid transfer operators in spherical coordinates. (a) Restriction operator R: A coarse-grid point (r 2h , θ2h ) (red circle) receives contributions from four surrounding fine-grid points (r h i , θh j ) (black circles), weighted by the overlapping shaded areas S i j. The coarse-grid residual is computed as an area-weighted average of the fine-grid residuals. (b) Prol… view at source ↗
Figure 2
Figure 2. Figure 2: Equatorial slices (θ = π/2 or z = 0) of the potential and its relative error for the double-sphere test on spherical (left-row) and cylindrical (right-right) grids. The results are shown for both the radially logarithmic (top) and uniform (bottom) grid. to probe both grid-aligned and non-grid-aligned density dis￾tributions, thereby evaluating the solver performance under idealized as well as more general g… view at source ↗
Figure 3
Figure 3. Figure 3: Comparison of the potential and its relative error for the mesh-segment test on logarithmic spherical (left) and cylindrical (right) grids. The top panels show equatorial slices (θ = π/2 or z = 0, for spherical or cylindrical grid, respectively), while the bottom panels show meridional slices at ϕ = 0.125π. where di denotes the distance from a field point to the center of the i th sphere. In spherical coor… view at source ↗
Figure 4
Figure 4. Figure 4: Same as [PITH_FULL_IMAGE:figures/full_fig_p015_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: The L2 of the relative error, |ϵ|2, as a function of resolution N, for the double-sphere (daashed lines) and mesh-segment (solid lines) tests. Each panel corresponds to a specific coordinate system and grid type as indicated in the panel title. The solid black line denotes the expected second-order convergence scaling. accuracy for both spherical and cylindrical coordinate sys￾tems when geometric discretiz… view at source ↗
Figure 6
Figure 6. Figure 6: Evolution of the central density during the collapse of a non-rotating uniform sphere. The top panel shows the time required for the central density to reach a value ρ, comparing simulations per￾formed on a spherical grid (green hexagons) and a cylindrical grid (red stars) with the analytical prediction (solid blue line; Eq. (85)). The dotted line, referenced to the right-hand axis, shows the relative cont… view at source ↗
Figure 7
Figure 7. Figure 7: Slices of the density (left-half of each panel) and azimuthal velocity, vϕ (right-half) distribution for the rotating uniform sphere test at t ∼ 1.32 tff are shown in the equatorial plane (θ = π/2 in spherical coordinates and z = 0 in cylindrical coordinates; left panels) and in the meridional plane (ϕ = π/2; right panels). The top panels correspond to results obtained on the spherical grid, while the bott… view at source ↗
Figure 8
Figure 8. Figure 8: The surface density distribution of the rotating cloud with m = 2 mode perturbation at t = 1.23 tff (top panels) and 1.26 tff (bottom panels), computed on the spherical (left) and cylindrical (right) grids. Note that, the top and bottom panels uses different color scale in order to show the structures at the respective time more clearly. (left) and cylindrical (right) grids. The initially small m = 2 pertu… view at source ↗
Figure 9
Figure 9. Figure 9: Top row: Surface density (Σ = R ρdz) distribution for the ring fragmentation test at t(Gρc) 1/2 = 18.0. The left panel shows the result from the spherical grid simulation, while the right panel corresponds to cylindrical grid. Bottom row: Time evolution of the Fourier amplitude Lm of the integrated ring density for m = 7, 8, 9 and 10 in spherical (left) and cylindrical (right) grid. The amplitudes are norm… view at source ↗
Figure 10
Figure 10. Figure 10: Result of weak scaling test. The average per-cell update time, tupdate, is plotted as a function of core count, Ncore, for spherical (left) and cylindrical (right) grids. Solid lines denote radially logarithmic grids, while dashed lines denote uniform radial grids. Blue hexagons show the total time taken by the Poisson solver, from density input to the computed potential. Magenta circles indicate the MHD … view at source ↗
Figure 11
Figure 11. Figure 11: Number of SOR iterations required for convergence as a function of the relaxation parameter ω for the radial and polar problems. The left and middle panels correspond to logarithmic and uniform radial grids, respectively, while the right panel shows results for a uniform polar grid. In each case, two resolutions are considered: N = 128 (green) and N = 256 (blue). The theoretical predictions (vertical line… view at source ↗
Figure 12
Figure 12. Figure 12: Left: Residual reduction as a function of multigrid V-cycles for the mesh-segment test on a logarithmic spherical grid. Each curve corresponds to a Fourier mode m, color-coded from m = 0 (dark blue) to m = Nϕ/2 (yellow). The residuals are normalized to their initial values. Lower-m modes converge more slowly, while higher-m modes converge within 1–2 V-cycles, consistent with the increasing diagonal domina… view at source ↗
Figure 13
Figure 13. Figure 13: Weak scaling breakdown of the Poisson solver components: FFTs (blue), first multigrid solve (green), DGF convolution (purple), second multigrid solve (red), and total (black). Left: absolute time per cell per timestep. Right: relative fraction of total time. approximately N 0.27, indicating a very weak dependence on resolution. This is consistent with theoretical expectations for variable-coefficient anis… view at source ↗
read the original abstract

Accurate and efficient computation of self-gravity is essential in astrophysical fluid dynamics, particularly in spherical and cylindrical geometries where large radial dynamic ranges and non-axisymmetric structures arise. Poisson solvers in such settings must simultaneously achieve high accuracy, scalability, and flexibility across a wide range of grid configurations and physical regimes. We present a robust and scalable Poisson solver for three-dimensional non-Cartesian geometries, supporting both spherical and cylindrical coordinates with either uniform or logarithmic radial discretizations. The method employs azimuthal Fourier decomposition to transform the 3D Poisson equation into a set of independent 2D Helmholtz equations. These are solved using a geometrically consistent multigrid algorithm that preserves second-order accuracy on both uniform and non-uniform grids. Vacuum boundary conditions are implemented through a screening-mass approach, enabling accurate solutions in domains with open boundaries, inner cavities, and strongly non-axisymmetric mass distributions. Owing to the differing convergence rates of Fourier modes -- where higher-order modes converge more rapidly -- the solver allows efficient mode-by-mode treatment. The combination of spectral decomposition and multigrid acceleration provides an efficient and flexible computational framework. The solver is implemented in the PLUTO code and validated against both analytical solutions and dynamical test problems in spherical and cylindrical geometries. Results demonstrate second-order convergence and excellent agreement with reference solutions. Weak-scaling tests up to 4096 cores show strong parallel performance, with the Poisson solve remaining subdominant to magnetohydrodynamic update cost. This makes the method well suited for large-scale simulations of star formation, accretion disks, and gravitational instabilities.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 0 minor

Summary. The manuscript presents a Poisson solver for 3D spherical and cylindrical geometries that decomposes the problem azimuthally via Fourier transforms into independent 2D Helmholtz equations, solved by a geometrically consistent multigrid algorithm on uniform or logarithmic radial grids. Vacuum boundaries are handled with a screening-mass method. The solver is implemented in PLUTO, with claims of second-order convergence, agreement with analytical and dynamical reference solutions, and good weak scaling to 4096 cores where the Poisson step remains subdominant to MHD updates.

Significance. If the second-order accuracy holds on logarithmic grids, the method would offer a scalable, flexible tool for self-gravity in astrophysical codes handling large radial dynamic ranges and non-axisymmetric structures, directly addressing needs in star formation and accretion-disk simulations.

major comments (2)
  1. [Abstract / validation sections] Abstract and validation sections: the central claim of second-order convergence on both uniform and logarithmic grids is stated but unsupported by any quantitative error tables, convergence plots, or discretization analysis in the provided text. This absence prevents verification that the multigrid transfers preserve the underlying finite-volume order for the variable-coefficient Helmholtz operator after Fourier decomposition.
  2. [Multigrid algorithm / weakest assumption] Geometrically consistent multigrid description: the assertion that the algorithm remains second-order accurate on logarithmic radial meshes (where cell volumes scale as r^2) requires an isolated test on a single azimuthal Fourier mode with an exact Helmholtz solution; without it, any mismatch between prolongation/restriction and the discrete divergence form could drop global accuracy below two, directly undermining the headline result.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the careful reading and constructive comments. We address each major comment below.

read point-by-point responses

  1. Referee: [Abstract / validation sections] Abstract and validation sections: the central claim of second-order convergence on both uniform and logarithmic grids is stated but unsupported by any quantitative error tables, convergence plots, or discretization analysis in the provided text. This absence prevents verification that the multigrid transfers preserve the underlying finite-volume order for the variable-coefficient Helmholtz operator after Fourier decomposition.

    Authors: We agree that the manuscript text as provided to the referee does not contain the requested quantitative tables, plots or discretization analysis. We will add explicit L2 and L-infinity error tables, convergence plots for both uniform and logarithmic radial grids, and a short section confirming that the multigrid restriction/prolongation operators preserve the second-order accuracy of the underlying finite-volume discretization of the Fourier-decomposed Helmholtz operator. revision: yes

  2. Referee: [Multigrid algorithm / weakest assumption] Geometrically consistent multigrid description: the assertion that the algorithm remains second-order accurate on logarithmic radial meshes (where cell volumes scale as r^2) requires an isolated test on a single azimuthal Fourier mode with an exact Helmholtz solution; without it, any mismatch between prolongation/restriction and the discrete divergence form could drop global accuracy below two, directly undermining the headline result.

    Authors: We agree that an isolated single-mode test would provide the strongest verification. We will insert a dedicated subsection containing an exact Helmholtz solution on a logarithmic radial grid for a single azimuthal Fourier mode, together with measured convergence rates that confirm the multigrid transfers remain consistent with the discrete divergence form and preserve global second-order accuracy. revision: yes

Circularity Check

0 steps flagged

No circularity; standard Fourier + multigrid construction with external analytical validation

full rationale

The derivation chain consists of azimuthal Fourier decomposition reducing the 3D Poisson equation to independent 2D Helmholtz problems, followed by a geometrically consistent multigrid solver on the resulting variable-coefficient operators. These steps are assembled from well-known components (Fourier transform, finite-volume discretization, multigrid transfers) without any self-referential definitions, fitted parameters renamed as predictions, or load-bearing self-citations. Convergence order is established by direct comparison to analytical solutions and reference codes rather than by construction from the inputs themselves. The paper therefore remains self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

The approach rests on standard Fourier decomposition and multigrid convergence properties plus the domain-specific claim that second-order accuracy is retained on non-uniform grids; no free parameters or new entities are introduced.

axioms (2)
  • standard math Azimuthal Fourier decomposition reduces the 3D Poisson equation to a set of independent 2D Helmholtz equations
    Invoked in the method description as the first transformation step.
  • domain assumption Geometrically consistent multigrid preserves second-order accuracy on uniform and logarithmic radial discretizations
    Central to the claim of accuracy on non-uniform grids; stated without derivation in the abstract.

pith-pipeline@v0.9.1-grok · 5829 in / 1379 out tokens · 34752 ms · 2026-06-26T22:57:27.434974+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

104 extracted references · 91 canonical work pages · 28 internal anchors

  1. [1]

    , keywords =

    A Self-gravity Module for the PLUTO Code. , keywords =. doi:10.3847/1538-4365/aced0a , archivePrefix =. 2306.05332 , primaryClass =

    work page doi:10.3847/1538-4365/aced0a

  2. [2]

    and Teukolsky, Saul A

    Press, William H. and Teukolsky, Saul A. and Vetterling, William T. and Flannery, Brian P. , title =. 1992 , isbn =

    1992

  3. [3]

    2000 , isbn =

    Trottenberg, Ulrich and Oosterlee, Cornelius and Schuller, Anton , title =. 2000 , isbn =

    2000

  4. [4]

    and Henson, Van Emden and McCormick, Steve F

    Briggs, William L. and Henson, Van Emden and McCormick, Steve F. , year =. A Multigrid Tutorial, Second Edition , ISBN =. doi:10.1137/1.9780898719505 , publisher =

    work page doi:10.1137/1.9780898719505

  5. [5]

    A simple multigrid scheme for solving the Poisson equation with arbitrary domain boundaries

    A simple multigrid scheme for solving the Poisson equation with arbitrary domain boundaries. Journal of Computational Physics , keywords =. doi:10.1016/j.jcp.2011.02.044 , archivePrefix =. 1104.1703 , primaryClass =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.1016/j.jcp.2011.02.044 2011

  6. [6]

    A Direct Multigrid Poisson Solver for Oct-Tree Adaptive Meshes

    A Direct Multigrid Poisson Solver for Oct-Tree Adaptive Meshes. ApJS , keywords =. doi:10.1086/526425 , archivePrefix =. 0710.4397 , primaryClass =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.1086/526425

  7. [7]

    ApJS , keywords =

    Self-gravitational Force Calculation of Second-order Accuracy Using the Multigrid Method on Nested Grids. ApJS , keywords =. doi:10.3847/1538-4365/ab66ba , archivePrefix =. 2001.02327 , primaryClass =

    work page doi:10.3847/1538-4365/ab66ba 2001

  8. [8]

    A&A , keywords =

    Self-gravitational adaptive mesh magnetohydrodynamics with the NIRVANA code. A&A , keywords =. doi:10.1051/0004-6361:20042451 , adsurl =

    work page doi:10.1051/0004-6361:20042451

  9. [9]

    PLUTO: a Numerical Code for Computational Astrophysics

    PLUTO: A Numerical Code for Computational Astrophysics. ApJS , keywords =. doi:10.1086/513316 , archivePrefix =. astro-ph/0701854 , primaryClass =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.1086/513316

  10. [10]

    The PLUTO Code for Adaptive Mesh Computations in Astrophysical Fluid Dynamics

    The PLUTO Code for Adaptive Mesh Computations in Astrophysical Fluid Dynamics. ApJS , keywords =. doi:10.1088/0067-0049/198/1/7 , archivePrefix =. 1110.0740 , primaryClass =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.1088/0067-0049/198/1/7

  11. [11]

    Y., V¨ ais¨ anen, P., Whitelock, P

    E pur si muove: Galilean-invariant cosmological hydrodynamical simulations on a moving mesh. MNRAS , keywords =. doi:10.1111/j.1365-2966.2009.15715.x , archivePrefix =. 0901.4107 , primaryClass =

    work page doi:10.1111/j.1365-2966.2009.15715.x 2009

  12. [12]

    , keywords =

    The cosmological simulation code GADGET-2. MNRAS , keywords =. doi:10.1111/j.1365-2966.2005.09655.x , archivePrefix =. astro-ph/0505010 , primaryClass =

    work page doi:10.1111/j.1365-2966.2005.09655.x 2005

  13. [13]

    MNRAS , keywords =

    Beyond N-body: 3D cosmological gas dynamics. MNRAS , keywords =. doi:10.1093/mnras/235.3.911 , adsurl =

    work page doi:10.1093/mnras/235.3.911

  14. [14]

    ApJ , keywords =

    Fragmentation in a rotating protostar: a comparison of two three-dimensional computer codes. ApJ , keywords =. doi:10.1086/157497 , adsurl =

    work page doi:10.1086/157497

  15. [15]

    MNRAS , keywords =

    Multiple Fragmentation in Collapsing Protostars. MNRAS , keywords =. doi:10.1093/mnras/264.4.798 , adsurl =

    work page doi:10.1093/mnras/264.4.798

  16. [16]

    ApJL , keywords =

    The Jeans Condition: A New Constraint on Spatial Resolution in Simulations of Isothermal Self-gravitational Hydrodynamics. ApJL , keywords =. doi:10.1086/310975 , adsurl =

    work page doi:10.1086/310975

  17. [17]

    ApJ , keywords =

    Self-gravitational Hydrodynamics with Three-dimensional Adaptive Mesh Refinement: Methodology and Applications to Molecular Cloud Collapse and Fragmentation. ApJ , keywords =. doi:10.1086/305329 , adsurl =

    work page doi:10.1086/305329

  18. [18]

    , keywords =

    A second-order accurate Super TimeStepping formulation for anisotropic thermal conduction. MNRAS , keywords =. doi:10.1111/j.1365-2966.2012.20744.x , adsurl =

    work page doi:10.1111/j.1365-2966.2012.20744.x 2012

  19. [19]

    Scalable explicit implementation of anisotropic diffusion with Runge-Kutta-Legendre super-time-stepping

    Scalable explicit implementation of anisotropic diffusion with Runge-Kutta-Legendre super-time stepping. MNRAS , keywords =. doi:10.1093/mnras/stx2176 , archivePrefix =. 1702.05487 , primaryClass =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.1093/mnras/stx2176

  20. [20]

    Journal of Computational Physics , year = 2014, month = jan, volume =

    A stabilized Runge-Kutta-Legendre method for explicit super-time-stepping of parabolic and mixed equations. Journal of Computational Physics , year = 2014, month = jan, volume =. doi:10.1016/j.jcp.2013.08.021 , adsurl =

    work page doi:10.1016/j.jcp.2013.08.021 2014

  21. [21]

    CASTRO: A New Compressible Astrophysical Solver. I. Hydrodynamics and Self-gravity. ApJ , keywords =. doi:10.1088/0004-637X/715/2/1221 , archivePrefix =. 1005.0114 , primaryClass =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.1088/0004-637x/715/2/1221

  22. [22]

    Asymptotic Methods for Integrals , ISBN =

    Temme, Nico M , year =. Asymptotic Methods for Integrals , ISBN =. doi:10.1142/9195 , journal =

    work page doi:10.1142/9195

  23. [23]

    Computing and Visualization in Science , publisher =

    Cell-centred multigrid revisited , volume =. Computing and Visualization in Science , publisher =. 2004 , month = oct, pages =. doi:10.1007/s00791-004-0137-0 , number =

    work page doi:10.1007/s00791-004-0137-0 2004

  24. [24]

    Galactic Dynamics: Second Edition , ISBN =

    Binney, James and Tremaine, Scott , year =. Galactic Dynamics: Second Edition , ISBN =. doi:10.2307/j.ctvc778ff , publisher =

    work page doi:10.2307/j.ctvc778ff

  25. [25]

    arXiv e-prints , keywords =

    Mathematical Vindications of the ``Jeans Swindle''. arXiv e-prints , keywords =

  26. [26]

    Why does the Jeans Swindle work?

    Why does the jeans swindle work?. MNRAS , keywords =. doi:10.1093/mnrasl/sls051 , archivePrefix =. 1210.3363 , primaryClass =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.1093/mnrasl/sls051

  27. [27]

    Conservation of Gravitational Energy , volume =

    Hanawa, Tomoyuki , year =. Conservation of Gravitational Energy , volume =. Journal of Physics: Conference Series , publisher =. doi:10.1088/1742-6596/1225/1/012015 , number =

    work page doi:10.1088/1742-6596/1225/1/012015

  28. [28]

    White Dwarf Mergers on Adaptive Meshes. I. Methodology and Code Verification. ApJ , keywords =. doi:10.3847/0004-637X/819/2/94 , archivePrefix =. 1512.06099 , primaryClass =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.3847/0004-637x/819/2/94

  29. [29]

    ApJS , keywords =

    An Extension of the Athena++ Framework for Fully Conservative Self-gravitating Hydrodynamics. ApJS , keywords =. doi:10.3847/1538-4365/abcfbd , archivePrefix =. 2012.01340 , primaryClass =

    work page doi:10.3847/1538-4365/abcfbd 2012

  30. [30]

    A New Way to Conserve Total Energy for Eulerian Hydrodynamic Simulations with Self-Gravity

    A new way to conserve total energy for Eulerian hydrodynamic simulations with self-gravity. NewA , keywords =. doi:10.1016/j.newast.2012.08.002 , archivePrefix =. 1208.1795 , primaryClass =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.1016/j.newast.2012.08.002 2012

  31. [31]

    , keywords =

    FLASH: An Adaptive Mesh Hydrodynamics Code for Modeling Astrophysical Thermonuclear Flashes. ApJS , keywords =. doi:10.1086/317361 , adsurl =

    work page doi:10.1086/317361

  32. [32]

    Computer Physics Communications , year = 1995, month = aug, volume =

    A piecewise parabolic method for cosmological hydrodynamics. Computer Physics Communications , year = 1995, month = aug, volume =. doi:10.1016/0010-4655(94)00191-4 , adsurl =

    work page doi:10.1016/0010-4655(94)00191-4 1995

  33. [33]

    A&A , keywords =

    On the capabilities and limits of smoothed particle hydrodynamics. A&A , keywords =

  34. [34]

    Resolution requirements for simulating gravitational fragmentation using SPH

    Resolution requirements for simulating gravitational fragmentation using SPH. A&A , keywords =. doi:10.1051/0004-6361:20054100 , archivePrefix =. astro-ph/0512247 , primaryClass =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.1051/0004-6361:20054100

  35. [35]

    GANDALF - Graphical Astrophysics code for N-body Dynamics And Lagrangian Fluids

    GANDALF - Graphical Astrophysics code for N-body Dynamics And Lagrangian Fluids. MNRAS , keywords =. doi:10.1093/mnras/stx2405 , archivePrefix =. 1709.04488 , primaryClass =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.1093/mnras/stx2405

  36. [36]

    ApJS , keywords =

    TREESPH: A Unification of SPH with the Hierarchical Tree Method. ApJS , keywords =. doi:10.1086/191344 , adsurl =

    work page doi:10.1086/191344

  37. [37]

    GADGET: A code for collisionless and gasdynamical cosmological simulations

    GADGET: a code for collisionless and gasdynamical cosmological simulations. NewA , keywords =. doi:10.1016/S1384-1076(01)00042-2 , archivePrefix =. astro-ph/0003162 , primaryClass =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.1016/s1384-1076(01)00042-2

  38. [38]

    Gasoline: An adaptable implementation of TreeSPH

    Gasoline: a flexible, parallel implementation of TreeSPH. NewA , keywords =. doi:10.1016/j.newast.2003.08.004 , archivePrefix =. astro-ph/0303521 , primaryClass =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.1016/j.newast.2003.08.004 2003

  39. [39]

    GIZMO: A New Class of Accurate, Mesh-Free Hydrodynamic Simulation Methods

    A new class of accurate, mesh-free hydrodynamic simulation methods. MNRAS , keywords =. doi:10.1093/mnras/stv195 , archivePrefix =. 1409.7395 , primaryClass =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.1093/mnras/stv195

  40. [40]

    ApJ , keywords =

    A new calculation on rotating protostar collapse. ApJ , keywords =. doi:10.1086/158185 , adsurl =

    work page doi:10.1086/158185

  41. [41]

    Protostellar Hydrodynamics: Constructing and Testing a Spatially and Temporally Second-Order--accurate Method. I. Spherical Coordinates. ApJS , keywords =. doi:10.1086/191739 , adsurl =

    work page doi:10.1086/191739

  42. [42]

    MNRAS , keywords =

    Resolution requirements for smoothed particle hydrodynamics calculations with self-gravity. MNRAS , keywords =. doi:10.1093/mnras/288.4.1060 , adsurl =

    work page doi:10.1093/mnras/288.4.1060

  43. [43]

    Journal of Computational and Applied Mathematics , keywords =

    Star formation with 3-D adaptive mesh refinement: the collapse and fragmentation of molecular clouds. Journal of Computational and Applied Mathematics , keywords =

  44. [44]

    MNRAS , keywords =

    Accelerating self-gravitating hydrodynamics simulations with adaptive force updates. MNRAS , keywords =. doi:10.1093/mnras/stab2208 , archivePrefix =. 2010.13792 , primaryClass =

    work page doi:10.1093/mnras/stab2208 2010

  45. [45]

    ApJ , keywords =

    A Production Mechanism for Clusters of Dense Cores. ApJ , keywords =. doi:10.1086/303982 , adsurl =

    work page doi:10.1086/303982

  46. [46]

    MNRAS , keywords =

    Impact of relativistic jets on the star formation rate: a turbulence-regulated framework. MNRAS , keywords =. doi:10.1093/mnras/stab2822 , archivePrefix =. 2109.13654 , primaryClass =

    work page doi:10.1093/mnras/stab2822

  47. [47]

    Density, Velocity, and Magnetic Field Structure in Turbulent Molecular Cloud Models

    Density, Velocity, and Magnetic Field Structure in Turbulent Molecular Cloud Models. ApJ , keywords =. doi:10.1086/318290 , archivePrefix =. astro-ph/0008454 , primaryClass =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.1086/318290

  48. [48]

    Malicious User Experience Design Research for Cybersecurity

    Theory of Star Formation. ARA&A , keywords =. doi:10.1146/annurev.astro.45.051806.110602 , archivePrefix =. 0707.3514 , primaryClass =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.1146/annurev.astro.45.051806.110602

  49. [49]

    Dimension as a Key to the Neutrino Mechanism of Core-Collapse Supernova Explosions

    Dimension as a Key to the Neutrino Mechanism of Core-collapse Supernova Explosions. ApJ , keywords =. doi:10.1088/0004-637X/720/1/694 , archivePrefix =. 1006.3792 , primaryClass =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.1088/0004-637x/720/1/694

  50. [50]

    Science , keywords =

    Giant planet formation by gravitational instability. Science , keywords =. doi:10.1126/science.276.5320.1836 , adsurl =

    work page doi:10.1126/science.276.5320.1836

  51. [51]

    ARA&A , year = 1998, month = jan, volume =

    Simulations of Structure Formation in the Universe. ARA&A , year = 1998, month = jan, volume =. doi:10.1146/annurev.astro.36.1.599 , adsurl =

    work page doi:10.1146/annurev.astro.36.1.599 1998

  52. [52]

    Galaxy Formation and Evolution , ISBN =

    Mo, Houjun and van den Bosch, Frank and White, Simon , year =. Galaxy Formation and Evolution , ISBN =. doi:10.1017/cbo9780511807244 , publisher =

    work page doi:10.1017/cbo9780511807244

  53. [53]

    2005, MNRAS, 364, L18, doi: 10.1111/j.1745-3933.2005.00096.x

    Investigating fragmentation conditions in self-gravitating accretion discs. MNRAS , keywords =. doi:10.1111/j.1745-3933.2005.00105.x , archivePrefix =. astro-ph/0509413 , primaryClass =

    work page doi:10.1111/j.1745-3933.2005.00105.x 2005

  54. [54]

    An Improved Multipole Approximation for Self-Gravity and Its Importance for Core-Collapse Supernova Simulations

    An Improved Multipole Approximation for Self-gravity and Its Importance for Core-collapse Supernova Simulations. ApJ , keywords =. doi:10.1088/0004-637X/778/2/181 , archivePrefix =. 1307.3135 , primaryClass =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.1088/0004-637x/778/2/181

  55. [55]

    Multi-level adaptive solutions to boundary-value problems , volume =

    Brandt, Achi , year =. Multi-level adaptive solutions to boundary-value problems , volume =. Mathematics of Computation , publisher =. doi:10.1090/s0025-5718-1977-0431719-x , number =

    work page doi:10.1090/s0025-5718-1977-0431719-x 1977

  56. [56]

    R., Millman, K

    Charles R. Harris and K. Jarrod Millman and St. Array programming with. 2020 , month = sep, journal =. doi:10.1038/s41586-020-2649-2 , publisher =

    work page doi:10.1038/s41586-020-2649-2 2020

  57. [57]

    2007, Computing in Science & Engineering, 9, 90, doi: 10.1109/MCSE.2007.55

    Hunter, John D. , year =. Matplotlib: A 2D Graphics Environment , volume =. Computing in Science & Engineering , publisher =. doi:10.1109/mcse.2007.55 , number =

    work page doi:10.1109/mcse.2007.55 2007

  58. [58]

    The Journal of Open Source Software , keywords =

    CMasher: Scientific colormaps for making accessible, informative and 'cmashing' plots. The Journal of Open Source Software , keywords =. doi:10.21105/joss.02004 , archivePrefix =. 2003.01069 , primaryClass =

    work page doi:10.21105/joss.02004 2003

  59. [59]

    Equilibrium Sequences and Gravitational Instability of Rotating Isothermal Rings

    Equilibrium Sequences and Gravitational Instability of Rotating Isothermal Rings. , keywords =. doi:10.3847/0004-637X/829/1/45 , archivePrefix =. 1607.03570 , primaryClass =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.3847/0004-637x/829/1/45

  60. [60]

    , keywords =

    A Fast Poisson Solver of Second-order Accuracy for Isolated Systems in Three-dimensional Cartesian and Cylindrical Coordinates. , keywords =. doi:10.3847/1538-4365/ab09e9 , archivePrefix =. 1902.08369 , primaryClass =

    work page doi:10.3847/1538-4365/ab09e9 1902

  61. [61]

    , keywords =

    A Fast, Second-order Accurate Poisson Solver in Spherical Polar Coordinates. , keywords =. doi:10.3847/1538-4365/adec68 , archivePrefix =. 2507.06784 , primaryClass =

    work page doi:10.3847/1538-4365/adec68

  62. [62]

    , keywords =

    A Versatile Method for Obtaining Structures of Rapidly Rotating Stars. , keywords =. doi:10.1086/191121 , adsurl =

    work page doi:10.1086/191121

  63. [63]

    Astronomische Nachrichten , keywords =

    Toward an efficient second-order method for computing the surface gravitational potential on spherical-polar meshes. Astronomische Nachrichten , keywords =. doi:10.1002/asna.20240056 , archivePrefix =. 2407.11648 , primaryClass =

    work page doi:10.1002/asna.20240056

  64. [64]

    arXiv e-prints , keywords =

    An efficient spectral Poisson solver for the nirvana-III code: the shearing-box case with vertical vacuum boundary conditions. arXiv e-prints , keywords =. doi:10.48550/arXiv.2510.10070 , archivePrefix =. 2510.10070 , primaryClass =

    work page doi:10.48550/arxiv.2510.10070

  65. [65]

    Self-gravity in curved mesh elements

    Self-gravity in curved mesh elements. Celestial Mechanics and Dynamical Astronomy , keywords =. doi:10.1007/s10569-014-9535-x , archivePrefix =. 1401.2758 , primaryClass =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.1007/s10569-014-9535-x

  66. [66]

    The solution of poisson’s equation for isolated source distributions , volume =

    James, R.A , year =. The solution of poisson’s equation for isolated source distributions , volume =. Journal of Computational Physics , publisher =. doi:10.1016/0021-9991(77)90013-4 , number =

    work page doi:10.1016/0021-9991(77)90013-4

  67. [67]

    , keywords =

    A Compact Cylindrical Green's Function Expansion for the Solution of Potential Problems. , keywords =. doi:10.1086/308062 , adsurl =

    work page doi:10.1086/308062

  68. [68]

    , keywords =

    GROMMET: an N-body code for high-resolution simulations of individual galaxies. , keywords =. doi:10.1111/j.1365-2966.2007.12344.x , archivePrefix =. astro-ph/0703406 , primaryClass =

    work page doi:10.1111/j.1365-2966.2007.12344.x 2007

  69. [69]

    and Johnson, Charles R

    Horn, Roger A. and Johnson, Charles R. , title=. 1985 , doi=

    1985

  70. [70]

    and Orszag, Steven A

    Bender, Carl M. and Orszag, Steven A. , title =. 1999 , doi=

    1999

  71. [71]

    , title =

    Young, David M. , title =. 1971 , doi=

    1971

  72. [72]

    , title =

    Varga, Richard S. , title =. 2009 , doi=

    2009

  73. [73]

    , year =

    Strikwerda, John C. , year =. Finite Difference Schemes and Partial Differential Equations, Second Edition , ISBN =. doi:10.1137/1.9780898717938 , publisher =

    work page doi:10.1137/1.9780898717938

  74. [74]

    Iterative methods for solving partial difference equations of elliptic type , volume =

    Young, David , year =. Iterative methods for solving partial difference equations of elliptic type , volume =. Transactions of the American Mathematical Society , publisher =. doi:10.1090/s0002-9947-1954-0059635-7 , number =

    work page doi:10.1090/s0002-9947-1954-0059635-7 1954

  75. [75]

    II ---Full Nonlinear Numerical Simulations---

    Fragmentation of Isothermal Sheet-Like Clouds. II ---Full Nonlinear Numerical Simulations---. Progress of Theoretical Physics , year = 1987, month = dec, volume =. doi:10.1143/PTP.78.1273 , adsurl =

    work page doi:10.1143/ptp.78.1273 1987

  76. [76]

    Dynamics of Protoplanetary Disks

    Dynamics of Protoplanetary Disks. , keywords =. doi:10.1146/annurev-astro-081710-102521 , archivePrefix =. 1011.1496 , primaryClass =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.1146/annurev-astro-081710-102521

  77. [77]

    , keywords =

    Star formation in molecular clouds: observation and theory. , keywords =. doi:10.1146/annurev.aa.25.090187.000323 , adsurl =

    work page doi:10.1146/annurev.aa.25.090187.000323

  78. [78]

    MHD Simulations of Global Accretion Disks with Vertical Magnetic Fields

    Magnetohydrodynamic Simulations of Global Accretion Disks with Vertical Magnetic Fields. , keywords =. doi:10.1088/0004-637X/784/2/121 , archivePrefix =. 1309.6916 , primaryClass =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.1088/0004-637x/784/2/121

  79. [79]

    Global Magnetohydrodynamical Simulations of Accretion Tori

    Global Magnetohydrodynamical Simulations of Accretion Tori. , keywords =. doi:10.1086/308180 , archivePrefix =. astro-ph/9907385 , primaryClass =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.1086/308180

  80. [80]

    1988 , doi =

    Computer Simulation Using Particles , publisher =. 1988 , doi =

    1988

Showing first 80 references.