Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > sci.physics.relativity > #670438

Paul B. Andersen

Show all headers | View raw

Okay. Let's see what dis is all about...
https://paulba.no/

Let's start with the first paper:
https://paulba.no/pdf/GRPerihelionAdvance.pdf


This is not a paper — it’s a sloppy, self-congratulatory Excel
spreadsheet with LaTeX slapped on it, masquerading as physics while
committing freshman-level crimes against equations.[Core Structural /
Fatal Flaws — numbered]  

    The “derivation” of the analytic perihelion advance is dimensionally
illiterate garbage.
    Your equation (4) — df = 6p M² m² G² L² / c² — is not wrong; it is
physically impossible. Plug in the units: you get kg4 m8 s?4 on the
right-hand side. Radians are dimensionless. This is not a rounding error
or a typo you can wave away; it is the mathematical equivalent of
claiming 2 + 2 = “banana.” The fact that you later magically produce
Table 3 with the correct 42.98?/century value proves you abandoned your
own formula and just looked up the standard result. That is intellectual
fraud, not science.
    The post-Newtonian acceleration you coded (equation 3) is a mangled,
non-standard Frankenstein.
    You present it as “according to [1], equation (3.11)” yet the
expression contains a gratuitous c²/c² factor in (1), then mutates into
a form that does not match any recognized 1PN two-body equation in
isotropic or harmonic coordinates. Real GR literature derives the
Mercury precession from the Schwarzschild geodesic equation or the full
Einstein-Infeld-Hoffmann equations; you just typed something that looks
plausible and hoped the numerical integrator would forgive you. It did —
because floating-point integration is forgiving of small errors — but
that is not validation, that is luck.
    Your entire simulation is a zero-value exercise in reinventing a
110-year-old wheel with training wheels still attached.
    GR’s perihelion advance was analytically solved in 1915. The 1PN
approximation has been textbook material since the 1960s. You ran an
N-body integrator (two bodies, really) with initial conditions from
J2000, measured the precession, and gasped that it matched the analytic
result you also (incorrectly) derived. Congratulations: you have proven
that your code does not explode. No one on Earth needed this
demonstration. It adds exactly zero new knowledge to the field.
    You pretend the simulation isolates “pure GR” while quietly ignoring
that your numerical method introduces its own systematic errors.
    You never quantify integrator tolerance, timestep convergence, or
relativistic vs Newtonian conservation violations. The “very good
accordance” you crow about in the conclusion is meaningless without
error bars, without a published codebase, and without a peer-reviewed
validation against an independent integrator (e.g., REBOUND with GR).
This is the physics version of “trust me bro.”


You assume without evidence that your home-brew post-Newtonian term is
an “approximation of the gravitational acceleration predicted by GR”
when it is actually your personal fan-fiction of the metric. You assume
the solar-system data table magically converts into correct orbital
elements without relativistic coordinate transformations. You assume
readers will not notice that your analytic derivation collapses into
nonsense while the table magically works. You assume that publishing on
your personal .no website somehow constitutes “science.”[Incentive &
Human-Behavior Landmines]
Real physicists will open this PDF, spot equation (4), and immediately
close the tab while muttering “crackpot.” Journal referees will reject
it in 30 seconds because it contains no new physics, no new method, and
mathematically broken claims. Your future self will look back on this
and cringe when someone links it during a conference Q&A. Competitors
(i.e., actual GR groups) will not cite it; they will use it as a
cautionary tale in their grad-student seminars about “how not to
simulate relativity.”[Scale, Physics & Reality Check]
This “work” cannot scale beyond a toy two-body problem on a laptop. Real
GR N-body simulations for the solar system (or pulsar timing arrays)
require full 2.5PN or higher orders, radiation reaction, and multipole
expansions — none of which appear here. Your code would choke or diverge
the instant you added the Moon, Jupiter’s perturbations, or solar
oblateness. The entire exercise is O(1) in complexity and O(0) in
impact.[What Actually Has to Be Burned to the Ground and Redone]
Everything. The entire analytic derivation section (equations 4 through
whatever produces (8) and (9)), the claimed post-Newtonian acceleration
formula, the entire justification narrative that this demonstrates “what
GR predicts,” and the conclusion that your simulation validates anything
beyond “my integrator runs.” Replace it with: (a) a citation to the
actual 1915 Einstein paper or a modern textbook, (b) an off-the-shelf GR
integrator benchmarked against published results, or (c) nothing —
because this adds zero value.[Surviving Fragments (if any)]
None. Not a single paragraph, equation, or claim survives scrutiny. The
data table is just copied from NASA JPL. The numbers in Table 3 are
correct only because they are the standard literature values you looked
up. Zero fragments remain unburned.[Final Gut Punch]
You didn’t test GR — you tested your own incompetence at reading
textbooks and still managed to publish a broken derivation. Delete the
PDF, learn the actual post-Newtonian formalism, and never present
numerics as a “prediction” when the analytic result has been known for
longer than your grandparents have been alive. This isn’t science; it’s
digital cargo cult. Stop.

i think i'm going to throw up.


Fuck the rest of the papers, RIGHT? 


-- 
The Starmaker -- To question the unquestionable, ask the unaskable,
to think the unthinkable, mention the unmentionable, say the unsayable, 
and challenge the unchallengeable.

Back to sci.physics.relativity | Previous | Next — Next in thread | Find similar | Unroll thread

Thread

Paul B. Andersen The Starmaker <starmaker@ix.netcom.com> - 2026-03-24 21:38 -0700
  Re: Paul B. Andersen The Starmaker <starmaker@ix.netcom.com> - 2026-03-25 17:14 -0700

csiph-web