Pith. sign in

REVIEW 3 major objections 5 minor 29 references

This paper presents a method to construct cosmological initial data by solving the Einstein constraint equations outward from the origin, with no boundary conditions, using regularity conditions at r=0 to fix a unique solution.

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

T0 review · deepseek-v4-flash

2026-08-01 07:50 UTC pith:QN5MRJKC

load-bearing objection A genuinely useful extension of the parabolic-hyperbolic constraint method to spherical cosmological data, but the advertised uniqueness result is overclaimed and the missing data artifacts should be fixed before publication. the 3 major comments →

arxiv 2607.21308 v1 pith:QN5MRJKC submitted 2026-07-23 gr-qc

Cosmological initial data without periodic boundary conditions

K\'aroly Csuk\'as This is my paper
classification gr-qc MSC 83C0583F05
keywords Einstein constraint equationsparabolic-hyperbolic formulationcosmological initial dataregularity at originperfect fluid perturbationsno boundary conditionsuniquenessnumerical relativity
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

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

The paper aims to establish that cosmological initial data with localized matter perturbations can be generated without specifying any boundary conditions. Standard methods solve the Einstein constraint equations as elliptic problems and require boundary conditions (often periodic boxes); here the constraints are recast as a parabolic-hyperbolic system and integrated radially outward. Starting from regular data at the origin, fixed by smoothness assumptions, the evolution yields a unique solution on a spherical domain. If correct, this provides a way to isolate the systematic bias that periodic or other boundary conditions introduce, and avoids the non-uniqueness that plagues elliptic solvers.

Core claim

On the paper's own terms, the central claim is that the parabolic-hyperbolic form of the Einstein constraints, with free data describing a flat FLRW background plus localized anisotropic perfect-fluid perturbations, can be integrated outward from r=0 without any boundary conditions. The regularity conditions at the origin—derived by evaluating the r→0 limit of the equations and using l'Hopital's rule and a maximum-principle argument—uniquely fix the initial values (bN|0=1, k|0=0, K|0=2κ). The numerical implementation produces smooth data to the Hubble radius, with 4th-order convergence in radius and spectral convergence in angle, and the paper reports that higher-ℓ modes of the lapse fall of

What carries the argument

The central object is the parabolic-hyperbolic decomposition of the Einstein constraints on a spherical foliation: the Hamiltonian constraint becomes a parabolic equation for the secondary lapse bN and the momentum constraints become a symmetric hyperbolic system for the variables k and K. The sign of the quantity ⋆K = 2/r for the flat FLRW slicing sets the integration direction outward. Regularity at the origin is fixed by expanding all variables in a Taylor series and solving order by order; the leading conditions come from l'Hopital's rule and the maximum principle applied to the angular Laplacian equation for bN|0.

Load-bearing premise

The construction assumes the apparent singularity at r=0 is entirely an artifact of the foliation—all fundamental fields are smooth there—so that l'Hopital's rule and the maximum principle uniquely fix the Cauchy data; if a physical singularity, non-smooth field, or an alternative branch such as bN|0=0 existed, the unique specification would fail.

What would settle it

Solve the same free data with a standard elliptic method on a large spherical domain and compare with the outward-integrated solution: agreement would corroborate the regularity conditions, while any disagreement would indicate that either the boundary conditions or the origin regularity assumptions are not equivalent to the physical problem. A more direct test: attempt an inward integration from a finite radius to r=0 and look for a singularity or a second smooth branch (e.g., bN|0 = -1) that still satisfies the regularity equations; finding such a branch would break the claimed uniqueness.

Watch this falsifier — get emailed when new claim-graph text bears on it.

If this is right

  • If correct, initial data for cosmological simulations can be generated on a spherical patch without periodic boundary conditions, eliminating the toroidal-topology artifact that forces the Ricci scalar to vanish somewhere.
  • Because the system is evolutionary, the solution is unique for given free data, sidestepping the branch-selection failure mode of iterative elliptic solvers.
  • The method extends directly to spatially hyperbolic backgrounds and to ΛCDM (absorbing Λ into the background energy density), as the author argues; S3 topology needs separate treatment at the equator.
  • The generated data can serve as a reference to measure the systematic bias of periodic boundary conditions by comparing evolutions of the same physical configuration with different boundary treatments.
  • The Taylor-series first step near the origin yields 4th-order convergence, making the numerical scheme robust for later curvature growth studies.

Where Pith is reading between the lines

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

  • The author's outward-integration strategy suggests a concrete protocol for quantifying boundary-condition bias: evolve the same initial data with a periodic-box solver and with a boundary-condition-free evolution code, then compare curvature invariants; this comparison is not performed in the paper but is the stated motivation.
  • The regularity-at-the-origin construction may carry over to other foliations with a single focal point (e.g., hyperboloidal or null foliations), where 'origin regularity' would replace outer boundary conditions analogously.
  • A testable extension: use the outward-generated data as input to a free evolution code and monitor constraint violations; if violations grow at the outer edge, the absence of boundary conditions at the data level does not automatically guarantee a well-posed Cauchy evolution, since the outer boundary of the computational domain will still truncate the evolution.
  • The method's reliance on a spherical foliation limits it to topologies with a distinguished origin; for genuinely toroidal or compact slices, boundary conditions cannot be eliminated in the same way, so the no-boundary claim is specific to the single-origin patch.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit.

Referee Report

3 major / 5 minor

Summary. The paper applies the parabolic-hyperbolic formulation of the Einstein constraint equations to construct cosmological initial data on a spherical patch. The freely specifiable geometric data are taken from a flat FLRW slicing, and anisotropy is introduced through localized perturbations of the matter energy density and flux. Regularity conditions at r=0 are derived by l'Hôpital-type limits and a Taylor expansion, and are then used as the starting data for an outward integration of the constraint system. The construction is implemented numerically with the ConstraintSolver code, using a spectral angular representation and a Taylor first step followed by a 4th-order Runge-Kutta integrator. The paper reports 4th-order radial convergence and spectral angular convergence, and claims that the method requires no boundary conditions and always yields a unique solution.

Significance. If the uniqueness and well-posedness claims are substantiated, this would be a useful complement to elliptic conformal methods for cosmological initial data, particularly for studying systematic biases introduced by periodic boundary conditions. The paper's strengths are its explicit regularity analysis at the origin, the clear numerical convergence tests (Fig. 4), and the release of the code under an MIT license. The construction is transparent and the numerical results are reproducible in principle. However, the advertised general uniqueness guarantee goes beyond what is actually proven: the analysis establishes a formal Taylor expansion under smoothness and sign assumptions, but does not prove well-posedness of the degenerate origin problem or global existence/uniqueness. The numerical demonstration, while convincing, is for a specific FLRW background with positive ⋆K, so the broad claims in the abstract and Discussion need to be tempered or supported by additional analysis.

major comments (3)
  1. [Abstract and §4, vs. §2.3 and App. A] The claim that 'solving the constraints as a well-posed evolutionary system always yields a unique solution' is not established. Well-posedness of (4)-(6) is known for regular foliations with a fixed-sign ⋆K, but at r=0 the foliation degenerates and ⋆K=2/r in the FLRW example. The l'Hôpital/Taylor construction determines a finite set of Taylor coefficients under smoothness assumptions; it does not prove existence of an exact C∞ solution, convergence of the Taylor series, or uniqueness among all admissible branches. The paper should either prove local well-posedness in an appropriate weighted or regularized setting, or clearly restrict the uniqueness claim to the demonstrated FLRW case and state the regularity assumptions explicitly.
  2. [Sec. 2.3, Eqs. (22)-(23)] The branch selection at the origin is underdetermined. In spherical symmetry, Eq. (22) admits bN=0 and bN=-1 as well as bN=+1; the paper chooses +1 after assuming bN|0>0. The maximum-principle argument rules out other positive solutions, but not negative or zero branches. If the bN=0 branch were regular, conditions (34)-(36) would change. The statement that the data are 'uniquely determined by our smoothness assumptions' is therefore too strong: positivity of the lapse (or an equivalence argument excluding other branches) is an additional assumption that should be stated and justified.
  3. [App. A, Eqs. (43)-(45)] The Taylor expansion around the origin is truncated at fourth order and used for the first radial step. No bound is given for the remainder terms w_bN, w_k, w_K, and the smoothness class of the free data is not specified beyond heuristic parity statements. The agreement with the numerically integrated solution shown in Fig. 3 is good evidence that the truncation is adequate for the tested cases, but it does not close the existence gap. If the uniqueness/well-posedness claim is retained, this gap needs to be either addressed by a convergence proof or explicitly identified as a regularity assumption underlying the method.
minor comments (5)
  1. [Sec. 2.2, Eq. (14)] It would be helpful to state explicitly that ⋆K=2/r is a consequence of the chosen FLRW slicing and is not a generic property of the parabolic-hyperbolic system; this clarifies why the example is favorable and why the general claim needs separate support.
  2. [Sec. 2.3, Eq. (35)] The paper assumes p=0 before deriving the displayed simplification. This assumption should be stated as part of the perturbation setup in Sec. 3.2 rather than appearing as a later restriction.
  3. [Sec. 3.1, Fig. 4] For the convergence measure Q_f in Eq. (42), the paper should specify how solutions at different resolutions are aligned or interpolated before computing the L2 norms, especially near the origin where the Taylor step changes.
  4. [References] The DOI for Ref. [17] appears malformed ('10.1103/9nsk-jy7f'); please verify it. Also, the Zenodo link for raw data is stated as 'will be available'; for reproducibility, the data or a persistent DOI should be provided at submission or at final acceptance.
  5. [Notation in App. A, Eq. (33)] The mode formula for ∂_r^2 k|0 is hard to parse because of the combined (−1)^m factors and the use of ℓ-m indices. Please spell out the spin-weighted spherical harmonic conventions and the action of ð/ð̄ in this expression.

Circularity Check

0 steps flagged

No significant circularity; the constraint integration is self-contained and no fitted quantity is relabeled as a prediction.

full rationale

The paper's construction is an initial-value integration of the constraint equations (14)-(16). The free data—FLRW geometric background plus user-specified matter perturbations—are inputs, and the constrained variables (bN, k, K) are obtained by outward integration from regularity conditions derived in Sec. 2.3 and App. A. No parameter is fitted to a target quantity and then reported as a prediction. The regularity conditions at r=0 are derived in-paper using l'Hôpital's rule, a maximum-principle argument, and mode-by-mode Taylor analysis; they are not imported as an unexplained ansatz. The self-citations ([22], [23]) concern a Taylor-expansion technique and the author's own publicly available numerical code; neither carries the central derivation. The abstract's uniqueness statement rests on the cited evolutionary formulation [11-13,19] rather than on a proof in this paper, but that is an external well-posedness assertion, not a circular reduction; if unsupported it would be a correctness risk, not circularity. Accordingly, no circular step is present.

Axiom & Free-Parameter Ledger

2 free parameters · 5 axioms · 0 invented entities

The paper introduces no new physical entities. It relies on the established parabolic-hyperbolic constraint formalism, smoothness at the origin, and the freedom to specify matter sources. The main invented content is a set of chosen demonstration perturbations, not fitted parameters.

free parameters (2)
  • Background energy density e0 = 3
    Chosen in Sec 3.3 to set the Hubble scale at r=1. It is a normalization input, not fitted to any target, and the method does not depend on its value.
  • Perturbation amplitudes and bump/Gaussian parameters = Table 1: delta_e amplitude 1e-2; delta_bp amplitude 1e-3; bump centers/widths and Gaussian widths
    These specify the freely chosen matter perturbations. They are demonstration inputs, not fitted to data, and do not affect the validity of the method.
axioms (5)
  • domain assumption The parabolic-hyperbolic constraint system (4)-(6) is well-posed for the chosen foliation, with the Hamiltonian constraint parabolic and the momentum constraints symmetric hyperbolic.
    Taken from Racz [11-13] and Racz-Winicour [19]; the paper relies on this rather than re-proving it.
  • domain assumption The singularity at r=0 is entirely a foliation artifact and all fundamental variables are regular there.
    Sec 2.3 explicitly assumes this; it is the basis for deriving the initial data via l'Hopital's rule.
  • domain assumption Matter sources e and S_a are freely specifiable on Sigma and need only satisfy the constraints and origin regularity (e even, bp odd).
    Sec 2 states 'the matter content is unconstrained and we are free to specify e and S_a'; this is a standard ADM constraint decomposition input.
  • standard math For the chosen flat FLRW slicing, *K = 2/r > 0 for all r>0, so the integration can proceed outward.
    Direct computation from the FLRW metric (11)-(13); it fixes the direction of evolution.
  • standard math The maximum principle on the unit sphere selects bN|0=1 as the unique positive solution of Eq. (22).
    Standard elliptic maximum principle invoked in Sec 2.3.

pith-pipeline@v1.3.0-alltime-deepseek · 12719 in / 19376 out tokens · 186526 ms · 2026-08-01T07:50:32.088381+00:00 · methodology

0 comments
read the original abstract

We apply the parabolic-hyperbolic formulation of the Einstein constraint equations to generate cosmological initial data. The freely specifiable geometric data correspond to flat Friedmann--Lema\^itre--Robertson--Walker background, while the matter sector contains localized anisotropic perturbations of a perfect fluid. Unlike the standard approach, our method evolves the constraints outward from regular data at the origin and therefore requires no boundary conditions. This makes our method well suited as a starting point in investigating systematic biases introduced by commonly adopted boundary conditions, such as periodic boundaries. A second advantage concerns uniqueness: standard elliptic solvers may fail when multiple solutions exist, whereas solving the constraints as a well-posed evolutionary system always yields a unique solution. To demonstrate our method, we implement it numerically and generate cosmological initial data with localized anisotropic perfect fluid perturbations.

Figures

Figures reproduced from arXiv: 2607.21308 by K\'aroly Csuk\'as.

Figure 1
Figure 1. Figure 1: Perturbation in the φ = 0 plane. Panel 1a shows the energy density perturbation δe. The perturbation consists of 5 well-localized rings—although the effects of spectral truncation are still visible. Panel 1b shows the energy flux perturbation pb. Since the purpose of this perturbation is merely to induce some non-trivial structure in K, pb is just a simple spherically symmetric perturbation. 3.3 Results Al… view at source ↗
Figure 2
Figure 2. Figure 2: The constrained variables in the φ = 0 plane. Panel 2a shows the perturbed Nb. Due to the perturbation the variable slowly increases, but the increase is not isotropic. We can clearly see the trace of the perturbation. Panel 2b shows the perturbed K. The effect of the spherically symmetric perturbation δpb drives K away from the homogeneous background. Unlike Nb here we see that after reaching a maximum K … view at source ↗
Figure 3
Figure 3. Figure 3: Panels 3a and 3c show that the applied Taylor series expansion agrees well with the numerical data to significantly larger r than where it was applied (shown by dashed vertical lines). Panel 3b shows the non-vanishing ℓ > 0 modes of Nb. First, we see why we had to use ℓmax = 22 with perturbation truncated at ℓ = 20. The modes with ℓ ≤ 20 are excited by the perturbation which yields a relatively large trunc… view at source ↗
Figure 4
Figure 4. Figure 4: Convergence properties. Panel 4a shows the local convergence rate as we vary the step size in r. Apart from spikes due to roundoff errors we observe the expected 4th order algebraic convergence rate. Panel 4b shows error estimates as we change the angular resolution. The consistent shift in the error is the hallmark of spectral convergence. The convergence properties of the solution are shown in [PITH_FUL… view at source ↗

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

29 extracted references · 5 canonical work pages · 2 internal anchors

  1. [1]

    ‘L’int´ egration des ´ equations de la gravitation relativ- iste et le probl` eme des$n$corps’

    Andr´ e Lichnerowicz. ‘L’int´ egration des ´ equations de la gravitation relativ- iste et le probl` eme des$n$corps’. In:Journal de Math´ ematiques Pures et Appliqu´ ees23 (1944), pp. 37–63 (cit. on p. 3)

    1944

  2. [2]

    James W. York Jr. ‘Conformally Invariant Orthogonal Decomposition of Symmetric Tensors on Riemannian Manifolds and the Initial-value Prob- lem of General Relativity’. In:J. Math. Phys.14.4 (1973), pp. 456–464. doi:10.1063/1.1666338(cit. on p. 3). 18

    work page doi:10.1063/1.1666338(cit 1973

  3. [3]

    Aurrekoetxea, Katy Clough and Eugene A

    Josu C. Aurrekoetxea, Katy Clough and Eugene A. Lim. ‘Cosmology Us- ing Numerical Relativity’. In:Living Rev. Rel.28.1 (June 2025), p. 5. doi:10.1007/s41114- 025- 00058- z. arXiv:2409.01939 [gr-qc](cit. on p. 3)

    Pith/arXiv arXiv 2025

  4. [4]

    Steinhardt

    David Garfinkle, Anna Ijjas and Paul J. Steinhardt. ‘Initial Conditions Problem in Cosmological Inflation Revisited’. In:Phys. Lett. B843 (June 2023), p. 138028.doi:10.1016/j.physletb.2023.138028. arXiv:2304. 12150 [gr-qc](cit. on p. 3)

    arXiv 2023

  5. [5]

    ‘The Anisotropy of the Power Spectrum in Periodic Cosmological Simulations’

    G´ abor R´ acz et al. ‘The Anisotropy of the Power Spectrum in Periodic Cosmological Simulations’. In:Mon. Not. Roy. Astron. Soc.503.4 (Apr. 2021), pp. 5638–5645.doi:10.1093/mnras/stab874. arXiv:2006.10399 [astro-ph.CO](cit. on p. 3)

    Pith/arXiv arXiv 2021

  6. [6]

    ‘Periodic Boundary Conditions and$G 2$ Cosmology’

    Alan Coley and Woei Chet Lim. ‘Periodic Boundary Conditions and$G 2$ Cosmology’. In:Class. Quant. Grav.41.1 (Nov. 2023), p. 015009.doi: 10.1088/1361-6382/ad0b9f. arXiv:2312.03285 [gr-qc](cit. on pp. 3, 16)

    Pith/arXiv arXiv 2023

  7. [7]

    J. C. Niemeyer and K. Jedamzik. ‘Dynamics of Primordial Black Hole Formation’. In:Phys. Rev. D59.12 (1999), p. 124013.doi:10 . 1103 / PhysRevD.59.124013. arXiv:astro-ph/9901292(cit. on p. 3)

    Pith/arXiv arXiv 1999

  8. [8]

    East and Justin L

    Maxence Corman, William E. East and Justin L. Ripley. ‘Evolution of Black Holes through a Nonsingular Cosmological Bounce’. In:JCAP09.09 (Sept. 2022), p. 063.doi:10 . 1088 / 1475 - 7516 / 2022 / 09 / 063. arXiv: 2206.08466 [gr-qc](cit. on p. 3)

    Pith/arXiv arXiv 2022

  9. [9]

    East et al

    William E. East et al. ‘Spacetime Dynamics of a Higgs Vacuum Instability During Inflation’. In:Phys. Rev. D95.2 (Jan. 2017), p. 023526.doi:10. 1103/PhysRevD.95.023526. arXiv:1607.00381 [hep-ph](cit. on p. 3)

    Pith/arXiv arXiv 2017

  10. [10]

    Starting inflation in asymptotically flat spacetimes

    Sam E. Brady, Thomas W. Baumgarte and Katy Clough. ‘Starting In- flation in Asymptotically Flat Spacetimes’. In:arXiv:2607.06441 [gr-qc] (July 2026).doi:10 . 48550 / arXiv . 2607 . 06441. arXiv:2607 . 06441 [gr-qc](cit. on p. 3)

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.2607.06441 2026

  11. [11]

    ‘Is the Bianchi Identity Always Hyperbolic?’ In:Class

    Istv´ an R´ acz. ‘Is the Bianchi Identity Always Hyperbolic?’ In:Class. Quantum Grav.31.15 (Aug. 2014), p. 155004.doi:10.1088/0264- 9381/31/15/ 155004. arXiv:1406.1016 [gr-qc](cit. on pp. 3, 4)

    Pith/arXiv arXiv 2014

  12. [12]

    ‘Cauchy Problem as a Two-Surface Based ‘geometrodynam- ics’’

    Istv´ an R´ acz. ‘Cauchy Problem as a Two-Surface Based ‘geometrodynam- ics’’. In:Class. Quant. Grav.32 (2015), p. 015006.doi:10.1088/0264- 9381/32/1/015006. arXiv:1409.4914 [gr-qc](cit. on pp. 3, 4)

    Pith/arXiv arXiv 2015

  13. [13]

    ‘Constraints as Evolutionary Systems’

    Istv´ an R´ acz. ‘Constraints as Evolutionary Systems’. In:Class. Quantum Grav.33.1 (Jan. 2016), p. 015014.doi:10 . 1088 / 0264 - 9381 / 33 / 1 / 015014. arXiv:1508.01810 [gr-qc, physics:math-ph](cit. on pp. 3– 5). 19

    Pith/arXiv arXiv 2016

  14. [14]

    Numerical stability of the Hyperbolic Formulation of the Constraint equations for $\mathbb{T}^3$ cosmological space-times

    Alejandro Estrada and Leon Escobar-Diaz. ‘Numerical Construction of Initial Data Sets for Inhomogeneous Cosmological Space-Times with Spa- tial Topology of$\mathbb{T}ˆ3$’. In:arXiv:2309.02946 [gr-qc](Sept. 2023).doi:10.48550/arXiv.2309.02946. arXiv:2309.02946 [gr-qc] (cit. on p. 3)

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.2309.02946 2023

  15. [15]

    Aurrekoetxea, Katy Clough and Eugene A

    Josu C. Aurrekoetxea, Katy Clough and Eugene A. Lim. ‘CTTK: A New Method to Solve the Initial Data Constraints in Numerical Relativity’. In: Class. Quant. Grav.40.7 (Mar. 2023), p. 075003.doi:10.1088/1361- 6382/acb883. arXiv:2207.03125 [gr-qc](cit. on p. 3)

    Pith/arXiv arXiv 2023

  16. [16]

    Baumgarte, Niall ´O Murchadha and Harald P

    Thomas W. Baumgarte, Niall ´O Murchadha and Harald P. Pfeiffer. ‘The Einstein Constraints: Uniqueness and Non-Uniqueness in the Conformal Thin Sandwich Approach’. In:Phys. Rev. D75.4 (2007), p. 044009.doi: 10.1103/PhysRevD.75.044009. arXiv:gr-qc/0610120(cit. on p. 3)

    Pith/arXiv arXiv 2007

  17. [17]

    Baumgarte, Katy Clough and John T

    Thomas W. Baumgarte, Katy Clough and John T. Giblin Jr. ‘Restrictions on Initial Conditions in Cosmological Scenarios and Implications for Sim- ulations of Primordial Black Holes and Inflation’. In:Phys. Rev. D112.12 (Dec. 2025), p. 123528.doi:10 . 1103 / 9nsk - jy7f. arXiv:2509 . 26470 [gr-qc](cit. on pp. 3, 16)

    2025

  18. [18]

    ‘Cosmological Initial Data for Nu- merical Relativity’

    David Garfinkle and Lawrence Mead. ‘Cosmological Initial Data for Nu- merical Relativity’. In:Phys. Rev. D102.4 (Aug. 2020), p. 044022.doi: 10.1103/PhysRevD.102.044022. arXiv:2006.16360 [gr-qc](cit. on pp. 3, 4)

    Pith/arXiv arXiv 2020

  19. [19]

    ‘Toward Computing Gravitational Ini- tial Data without Elliptic Solvers’

    Istv´ an R´ acz and Jeffrey Winicour. ‘Toward Computing Gravitational Ini- tial Data without Elliptic Solvers’. In:Class. Quant. Grav.35.13 (June 2018), p. 135002.doi:10.1088/1361-6382/aac5c5. arXiv:1712.03294 [gr-qc](cit. on pp. 5, 6)

    Pith/arXiv arXiv 2018

  20. [20]

    George F. R. Ellis, Roy Maartens and Malcolm A. H. MacCallum.Re- lativistic Cosmology. Cambridge: Cambridge University Press, 2012.isbn: 978-0-521-38115-4.doi:10.1017/CBO9781139014403(cit. on p. 6)

    work page doi:10.1017/cbo9781139014403(cit 2012

  21. [21]

    Boyd.Chebyshev & Fourier Spectral Methods

    John P. Boyd.Chebyshev & Fourier Spectral Methods. Lecture Notes in Engineering 49. Springer Berlin, Heidelberg, 1989.isbn: 978-3-540-51487- 9 (cit. on p. 10)

    1989

  22. [22]

    ‘Hyperboloidal Initial Data without Log- arithmic Singularities’

    K´ aroly Csuk´ as and Istv´ an R´ acz. ‘Hyperboloidal Initial Data without Log- arithmic Singularities’. In:Gen. Rel. Grav.57.6 (June 2025), p. 96.doi: 10 . 1007 / s10714 - 025 - 03424 - y. arXiv:2503 . 11804 [gr-qc](cit. on pp. 10, 17)

    2025

  23. [23]

    2024.url:https :/ / gitlab

    Csuk´ as K´ aroly Zolt´ an.ConstraintSolver. 2024.url:https :/ / gitlab . wigner.hu/csukas.karoly/constraintsolver(cit. on pp. 10, 17)

    2024

  24. [24]

    ‘Spikes in the Mixmaster Regime of G 2 Cosmolo- gies’

    Woei Chet Lim et al. ‘Spikes in the Mixmaster Regime of G 2 Cosmolo- gies’. In:Phys. Rev. D79.12 (2009), p. 123526.doi:10.1103/PhysRevD. 79.123526. arXiv:0904.1546 [gr-qc](cit. on p. 16). 20

    Pith/arXiv arXiv 2009

  25. [25]

    J. D. Hunter. ‘Matplotlib: A 2D graphics environment’. In:Computing in Science & Engineering9.3 (2007), pp. 90–95.doi:10.1109/MCSE.2007. 55(cit. on p. 17)

    doi:10.1109/mcse.2007 2007

  26. [26]

    Harris et al

    Charles R. Harris et al. ‘Array programming with NumPy’. In:Nature 585.7825 (2020), pp. 357–362.doi:10.1038/s41586-020-2649-2.url: https://doi.org/10.1038/s41586-020-2649-2(cit. on p. 17)

    doi:10.1038/s41586-020-2649-2.url: 2020

  27. [27]

    Version latest

    The pandas development team.pandas-dev/pandas: Pandas. Version latest. 2020.doi:10.5281/zenodo.3509134.url:https://doi.org/10.5281/ zenodo.3509134(cit. on p. 17)

    work page doi:10.5281/zenodo.3509134.url:https://doi.org/10.5281/ 2020

  28. [28]

    ‘Data Structures for Statistical Computing in Python’

    Wes McKinney. ‘Data Structures for Statistical Computing in Python’. In:Proceedings of the 9th Python in Science Conference. Ed. by St´ efan van der Walt and Jarrod Millman. 2010, pp. 56–61.doi:10.25080/Majora- 92bf1922-00a(cit. on p. 17)

    doi:10.25080/majora- 2010

  29. [29]

    2025.url: http://gnuplot.sourceforge.net(cit

    Thomas Williams, Colin Kelley and many others.Gnuplot. 2025.url: http://gnuplot.sourceforge.net(cit. on p. 17). 21

    2025