The 16 real eigenvalues of an overdamped chain, in two groups of 8, with the certificate between them
At its defaults it draws the 16 real eigenvalues of an overdamped chain, in two groups of 8, with the certificate between them. A chain of 8 masses with C = 8K, which is above the critical damping 5.75877 = 1/sin(π/2(n+1)), so the quadratic is hyperbolic and every one of its 16 eigenvalues is real. The vertical line is the μ at which Q(μ) is negative definite — a Cholesky that completes, and the whole proof that the spectrum is real. It also lies in the gap: exactly 8 eigenvalues above it and 8 below, with a gap of 0.6698 between the two groups. One factorisation answers both questions, which is the second thing a boolean would not have carried.
overdamped-gap is one function in lib/figures/hyperbolic.js —
real by class — a cholesky that certifies a spectrum, and a double root that costs half the digits. Everything below came out of it during this build, at
arguments taken from the essays rather than invented for this page. A figure here is the
figure a reader meets in an essay, and if the generator changes, this page changes with it.
At its defaults
Drawn even though every essay passes arguments — which on this site is every essay, at 100% of placements since the standard pass. A default nothing exercises is a trap for the next essay to call this with none, and this is the page where a default that has drifted from the figures around it becomes visible.
A chain of 8 masses with C = 8K, which is above the critical damping 5.75877 = 1/sin(π/2(n+1)), so the quadratic is hyperbolic and every one of its 16 eigenvalues is real. The vertical line is the μ at which Q(μ) is negative definite — a Cholesky that completes, and the whole proof that the spectrum is real. It also lies in the gap: exactly 8 eigenvalues above it and 8 below, with a gap of 0.6698 between the two groups. One factorisation answers both questions, which is the second thing a boolean would not have carried.
show: "definite-errors", n: 8
The arguments are the ones A guarantee paid for in digits passes. A value nobody placed would be a picture no essay asked for and no claim was ever checked against.
Reference: the closed-form spectrum. The companion route inverts the leading coefficient and takes a real Schur form; the definite route takes a Cholesky of the best positive definite combination of a symmetric linearisation and a Jacobi eigensolver. ε 10⁻²: companion 8.88·10⁻¹⁵, definite 2.35·10⁻¹⁴; ε 10⁻⁴: companion 4.15·10⁻¹⁴, definite 3.95·10⁻¹²; ε 10⁻⁶: companion 5.3·10⁻¹³, definite 1.46·10⁻¹⁰; ε 10⁻⁸: companion 1.13·10⁻¹², definite 5.92·10⁻⁸; ε 10⁻¹⁰: companion 1.07·10⁻¹⁰, definite 9.8·10⁻⁶; ε 10⁻¹²: companion 2.69·10⁻⁹, definite 0.002. The definite route's spectrum is real by construction; the companion route returned no complex value at any of these margins.
show: "definite-denominators"
The arguments are the ones A guarantee paid for in digits passes. A value nobody placed would be a picture no essay asked for and no claim was ever checked against.
ε 10⁻²: Crawford number 0.002288, 0.2288 ε; separation 0.09848, 0.9848 times the square root of ε; ε 10⁻⁴: Crawford number 2.279·10⁻⁵, 0.2279 ε; separation 0.009823, 0.9823 times the square root of ε; ε 10⁻⁶: Crawford number 2.279·10⁻⁷, 0.2279 ε; separation 9.823·10⁻⁴, 0.9823 times the square root of ε; ε 10⁻⁸: Crawford number 2.279·10⁻⁹, 0.2279 ε; separation 9.823·10⁻⁵, 0.9823 times the square root of ε; ε 10⁻¹⁰: Crawford number 2.279·10⁻¹¹, 0.2279 ε; separation 9.823·10⁻⁶, 0.9823 times the square root of ε; ε 10⁻¹²: Crawford number 2.284·10⁻¹³, 0.2284 ε; separation 9.822·10⁻⁷, 0.9822 times the square root of ε.
show: "definite-angle"
The arguments are the ones A guarantee paid for in digits passes. A value nobody placed would be a picture no essay asked for and no claim was ever checked against.
Where the curve is above zero the combination is positive definite. ε 10⁻²: peak 0.2288 ε at t = 0.337; at t = 0, 0.2158 ε; at t = 1.5, 0.0901 ε; ε 10⁻⁴: peak 0.2279 ε at t = 0.334; at t = 0, 0.2153 ε; at t = 1.5, 0.0901 ε; ε 10⁻⁶: peak 0.2279 ε at t = 0.334; at t = 0, 0.2153 ε; at t = 1.5, 0.0901 ε. Divided by ε, the three profiles coincide.
show: "definite-pair"
The arguments are the ones A guarantee paid for in digits passes. A value nobody placed would be a picture no essay asked for and no claim was ever checked against.
closed form: -0.347296846423 and -0.347295864245; companion: -0.347296845490 and -0.347295865178; definite pencil: -0.347296355138 and -0.346602323314. The axis is centred on -0.3472963553 and 0.00174 wide.
show: "definite-rotated"
The arguments are the ones A guarantee paid for in digits passes. A value nobody placed would be a picture no essay asked for and no claim was ever checked against.
ε 10⁻⁴: 0 of 20 reading every Schur block exactly, 0 at the default block tolerance; ε 10⁻⁸: 0 of 20 reading every Schur block exactly, 0 at the default block tolerance; ε 10⁻¹²: 0 of 20 reading every Schur block exactly, 0 at the default block tolerance; ε 0: 5 of 20 reading every Schur block exactly, 6 at the default block tolerance, largest imaginary part 5.3·10⁻⁸.
What it checked while drawing
Every figure above checked its own claims on the way to being drawn, and a claim that failed
would have stopped the picture rather than shipped a wrong one. Those checks used to leave
no trace at all: a passing one returned true and the only evidence the figure had
checked anything was that nothing crashed. The list below is what they actually said, collected
by running this generator with an observer installed — not a description of
what it is believed to check.
19 distinct claims across 6 sets of arguments, grouped below by shape — because most of them are one sentence with a different number in it, and how many separate times that sentence was put to the test is the informative part.
a chain length the sweep measures
a chain long enough to have a spectrum and short enough to draw
a Cholesky of the definite combination
a combination of the pencil that is positive definite
a hyperbolic quadratic to split
a linearisation has as many eigenvalues as it has rows
a positive mass
a quadratic the certificate says is hyperbolic
a reduction this file knows
a spectrum that fits along the axis
an entirely real spectrum
damping above the critical value, so the spectrum is real
damping that removes energy rather than adding it
every computed eigenvalue is matched to an unused exact one
Jacobi needs a symmetric matrix
LU is for square matrices
matmul shapes agree
n eigenvalues each side of the certificate
two spectra of the same size
Against the rule
It draws a decomposition and prints its residual. It calls
splitAtCertificate,
and every figure above carries the badge — which residualcheck verifies by looking
for it in the emitted SVG rather than by finding the call that builds one. A badge that is
constructed and then left out of the body is the failure that check exists for.
Across the library: the rule bites on 217
of 397 generators —
199 print a residual and
18 are exempt with a published reason;
180 factorise nothing.
Read from lib/residual-rule.js, which is the same body the gate enforces from,
and the gate's last check fails the build if this page and it disagree about any generator.
Where it is called
Changing this generator changes every figure on this list. That is what makes the list worth publishing rather than keeping in a check script.
A guarantee paid for in digits
A hyperbolic quadratic has a real spectrum, but the route that computed it so far, a companion linearisation and a nonsymmetric Schur form, only happens to return one. The theory offers a route on which it cannot do otherwise: a symmetric linearisation that is a definite pencil, reduced by a Cholesky to a symmetric eigenproblem. Near the class's boundary that guarantee is the expensive route. On a damped chain at damping β*(1 + ε) the definite pencil's Crawford number — what its Cholesky divides by — vanishes like 0.228ε, every combination of the pencil shrinking with it, while the colliding pair the companion route has to resolve separates like 0.98√ε. At ε = 10⁻¹² the definite route keeps two or three digits and the companion route eight or nine. And on twenty rotated chains the companion route returns no complex pair at any margin from 10⁻⁴ to 10⁻¹², only on the boundary itself.
The eigenvalue problem that is not linearEvery eigenvalue real, and a test that says so
A quadratic eigenvalue problem has no reason to have real eigenvalues. One class does, as a property rather than an outcome, and the proof is a Cholesky that completes. The boundary of the class has a closed form, and at the boundary the arithmetic loses half its digits with nothing ill conditioned anywhere.