Skip to main content
Verification registry

Measured against published problems, misses included.

44 of 46
closed-form worked examples inside the stated tolerance
238 of 333
finite-element benchmark rows within tolerance of an outside reference
37
outside tolerance, each listed with its measured difference and reason
17
published sources cited by the closed-form registry

The public record of how our engines compare with published problems, in two registries: the closed-form engines against printed worked examples (first table), and the finite-element solver against vendor tutorials, papers and analytical solutions (second table). Each row names the source and page, the value the source prints, the value our engine returns for the same inputs, the difference, and the tolerance with the reason for it.

Measured by our engineers on 2026-11-11 against the published value. Every row is public, misses included, so anyone can check it.

The suite runs each closed-form benchmark in your browser and stamps a reproducible SHA-256 hash. The manual groups both registries and every cited validation suite by analysis, deviations first, as a PDF too.

Showing 20 of 46 rows.

Verification registry: problem, our value, published value, difference, tolerance and status for each published problem; the source, note and runtime open under each row.
ProblemOursPublishedDifferenceToleranceStatusDetails
Bearing capacity — closed form
EN 1997-1 Annex D Nq at φ′d = 38° (M1)
BC-JRC-A33-NQ-M1
48.93348.87+0.13 %± 0.3 %Ours 48.933Published 48.87+0.13 %Pass
EN 1997-1 Annex D Nγ at φ′d = 38° (M1)
BC-JRC-A33-NG-M1
74.89974.8+0.13 %± 0.3 %Ours 74.899Published 74.8+0.13 %Pass
EN 1997-1 Annex D Nq at φ′d = tan⁻¹(tan 38°/1.25) (M2)
BC-JRC-A33-NQ-M2
23.19523.17+0.11 %± 0.3 %Ours 23.195Published 23.17+0.11 %Pass
EN 1997-1 Annex D Nγ at φ′d (M2)
BC-JRC-A33-NG-M2
27.74527.72+0.09 %± 0.3 %Ours 27.745Published 27.72+0.09 %Pass
Terzaghi, square footing — cohesion + surcharge terms (the Nγ term is a convention comparison, not reproduced)
BC-DAS-4.1
968.33kPa968.2kPa+0.01 %± 0.5 %Ours 968.33 kPaPublished 968.2 kPa+0.01 %Pass
General bearing equation, same footing
BC-DAS-4.3
1,374kPa1,373.2kPa+0.06 %± 0.5 %Ours 1,374 kPaPublished 1,373.2 kPa+0.06 %Pass
Inclined load β = 20°
BC-DAS-4.5
273.8kPa273.66kPa+0.05 %± 0.5 %Ours 273.8 kPaPublished 273.66 kPa+0.05 %Pass
Eccentric strip, e = 0.2 m
BC-DAS-4.7
3,288.2kPa3,287.4kPa+0.03 %± 0.5 %Ours 3,288.2 kPaPublished 3,287.4 kPa+0.03 %Pass
Eccentric + inclined strip
BC-DAS-4.12
461.67kPa461.98kPa−0.07 %± 0.5 %Ours 461.67 kPaPublished 461.98 kPa−0.07 %Pass
Case study — Bozozuk silo, φ = 0
BC-DAS-SILO
182.5kPa181.8kPa+0.39 %± 0.5 %Ours 182.5 kPaPublished 181.8 kPa+0.39 %Pass
Terzaghi, φ 30°, B 1.5 m
BC-BOW-4-1
1,641.2kPa1,638.8kPa+0.15 %± 1 %Ours 1,641.2 kPaPublished 1,638.8 kPa+0.15 %Pass
Hansen, submerged tank footing
BC-BOW-4-2-H
1,817.8kPa1,812kPa+0.32 %± 1 %Ours 1,817.8 kPaPublished 1,812 kPa+0.32 %Pass
Meyerhof (1963), submerged tank footing
BC-BOW-4-2-M
2,672.7kPa2,659kPa+0.52 %± 1 %Ours 2,672.7 kPaPublished 2,659 kPa+0.52 %Pass
Hansen, two-way eccentric 1.8 m square
BC-BOW-4-5
5,050.2kPa5,088kPa−0.74 %± 1.5 %Ours 5,050.2 kPaPublished 5,088 kPa−0.74 %Pass
Meyerhof inclined load — total checked against the book-consistent value, NOT the printed 399.9 kPa (documented source error)
BC-BOW-4-7-TOTAL
388kPa388.72kPa−0.19 %± 1 %Ours 388 kPaPublished 388.72 kPa−0.19 %Pass
Meyerhof inclined load — cohesion term c·Nc·dc·ic
BC-BOW-4-7-C
343.29kPa344kPa−0.21 %± 0.5 %Ours 343.29 kPaPublished 344 kPa−0.21 %Pass
Square pad on soft clay, undrained, DA1-C1
BC-DG-6.1-UND-DA1C1
586.79kN586kN+0.13 %± 1 %Ours 586.79 kNPublished 586 kN+0.13 %Pass
Square pad on soft clay, undrained, DA1-C2 (c_u,d = 30/1.4)
BC-DG-6.1-UND-DA1C2
434kN433kN+0.23 %± 1 %Ours 434 kNPublished 433 kN+0.23 %Pass
Square pad on soft clay, undrained, DA2
BC-DG-6.1-UND-DA2
580.11kN580kN+0.02 %± 1 %Ours 580.11 kNPublished 580 kN+0.02 %Pass
Square pad on soft clay, undrained, DA3
BC-DG-6.1-UND-DA3
600.69kN600kN+0.11 %± 1 %Ours 600.69 kNPublished 600 kN+0.11 %Pass
26 not shown yet; search or filter to find a row.

Δ = (ours − reference) / reference. Runtime is one warm call of the engine on the measuring machine, excluding page and network time. A miss stays in this table until the engine is corrected or the source is shown, with evidence, to be in error.

Finite-element verification registry

The finite-element solver keeps its own registry, alongside the table above. Each row compares a solver result with a value from outside the solver where one exists: a value printed in a published document, a closed-form or analytical solution, or a vendor’s published verification result. Self-checks compare the solver with something derived inside this project (for example our own TypeScript solver, a hand calculation or an input); they are labelled as such and counted separately from comparisons with an outside source. The registry was last updated 2026-11-14.

333
rows in the finite-element registry
238
within tolerance of an outside reference
42
self-checks within tolerance, counted separately
35
deviations, outside the stated tolerance
4
pending
0
blocked, waiting on a later solver stage
6
recorded, no published value to compare
8
superseded by newer rows, not open and not passes

By what the row is compared against: 224 published / analytical reference; 42 vendor-published value; 53 self-check; 14 no reference value. Deviations and pending rows stay in the table: a deviation is a result outside the stated tolerance (or a difference the row records as known), and pending and blocked rows are problems the solver is still being extended to run. Recorded and superseded rows are kept as well, with the reason, so nothing is removed from the record. What each status means is listed below.

Pass
Our value is within the tolerance the row states.
Deviation
Our value is outside the stated tolerance, or the row records a known difference; it is not explained away.
Recorded, no published value
We ran the case, but the source publishes no number to compare with, so there is no verdict.
Superseded
An older stand-in or umbrella row, replaced by the newer row or rows named on it. It is not counted as a pass or as open.
Pending
The case can be run with the solver as it is, and has not been run yet.

The notes under each row are our engineers’ working notes. Where a note cites “D” and a number (for example D114) it refers to an entry in our internal defect log; “doc” and a number is an internal design note. Neither is published. Ask us at hello@terrabrains.ai and we will explain the entry behind any row.

Showing 30 of 333 rows.

Finite-element verification registry: quantity, our value, reference value, difference and status for each row; the source, note and comparison basis open under each row.
ProblemOursPublishedDifferenceStatusDetails
total footing force 2π·|Force-Y| at 0.05 m indentation (rigid rough footing, case A)
plaxis-2d-tut1-a-force · vs vendor value
600.77kN—kN+2.17 %Ours 600.77 kNPublished — kN+2.17 %Pass
Force-Y per radian at ΣMstage = 0.3943 (u = 0.0197 m)
plaxis-2d-tut1-a-fy-0.3943 · vs vendor value
-53.974kN/rad—kN/rad+0.34 %Ours -53.974 kN/radPublished — kN/rad+0.34 %Pass
maximum total displacement (the footing)
plaxis-2d-tut1-a-umax
0.05m0.05m0.00 %Ours 0.05 mPublished 0.05 m0.00 %Pass
K0 procedure: σ′v = −Σγh + p, σ′h = 0.5 σ′v, p = 10(2 − y) — worst |error| over every integration point
plaxis-2d-tut1-k0-effective-stress · Self-check
1.00e-12kPahand calculation (γ_unsat 17 above y = 2, γ_sat 20 below, γ_w 10, K0 = 1 − sin 30° = 0.5)0.00 %Ours 1.00e-12 kPaPublished hand calculation (γ_unsat 17 above y = 2… kPa0.00 %Pass
|u| of node (0, 4) vs ΣMstage under 188 kPa on the plate footing (EA 5e6, EI 8.5e3, ν 0), axisymmetric; PLAXIS Figure 1–27: 0.5 → 0.0156, 0.6 → 0.0207, 0.8 → 0.0328, 0.9 → 0.0398, 1.0 → 0.0496 m (±0.…
plaxis-2d-tut1-b-flexible · vs vendor value
0.048745m—m−1.72 %Ours 0.048745 mPublished — m−1.72 %Pass
mesh-refinement study of the case-A force 2π·|Force-Y| (full Newton, tol 1e-8, tension cut-off on)
plaxis-2d-tut1-a-refinement
several values——Ours several valuesPublished ——Deviation
collapse average stress p = 2πF/(πR²) of a smooth rigid circular footing
plaxis-verif-circular-footing-cox
227.284kPa225.6kPa+0.75 %Ours 227.284 kPaPublished 225.6 kPa+0.75 %Pass
settlement at the centre of the tank
plaxis-verif-flexible-tank
74.207mm74.44mm−0.31 %Ours 74.207 mmPublished 74.44 mm−0.31 %Pass
normalised cavity pressure p/c at (a − a0)/a0 = 4 (also 0.4 / 1.2 / 2.8 within 0.15 %)
plaxis-verif-cavity-pressure
7.68127.6846−0.04 %Ours 7.6812Published 7.6846−0.04 %Pass
radius of the plastic zone at a = 5 m (outermost plastic integration point)
plaxis-verif-cavity-rp
28.003m28.28m−0.99 %Ours 28.003 mPublished 28.28 m−0.99 %Pass
normalised cavity pressure p/c at (a − a0)/a0 = 4, updated mesh (worst step over the curve −0.22 %)
plaxis-verif-cavity-large-pressure
5.56285.5668−0.07 %Ours 5.5628Published 5.5668−0.07 %Pass
radius of the plastic zone at a = 5 m, current configuration (outermost plastic integration point)
plaxis-verif-cavity-large-rp
48.74m49.11m−0.76 %Ours 48.74 mPublished 49.11 m−0.76 %Pass
ux / uy of the corner nodes
plaxis-verif-biaxial-le · Self-check
identical to 1e-13identical0.00 %Ours identical to 1e-13Published identical0.00 %Pass
γxy
plaxis-verif-biaxial-shear
0.00250.00250.00 %Ours 0.0025Published 0.00250.00 %Pass
failure σ1 (load control)
plaxis-verif-biaxial-mc
-6.4641kPa-6.4641kPa0.00 %Ours -6.4641 kPaPublished -6.4641 kPa0.00 %Pass
footing force for 0.01 m
plaxis-verif-smooth-rigid-strip
15.225kN/m15.148kN/m+0.51 %Ours 15.225 kN/mPublished 15.148 kN/m+0.51 %Pass
centre settlement
plaxis-verif-boussinesq
0.09033mm0.091mm−0.73 %Ours 0.09033 mmPublished 0.091 mm−0.73 %Pass
mean ultimate stress, smooth / rough (PLAXIS 2D 7.831 / 9.168 kPa)
plaxis-verif-davis-booker
7.8335 / 9.18267.800 / 9.090+0.43 / +1.02 %Ours 7.8335 / 9.1826Published 7.800 / 9.090+0.43 / +1.02 %Pass
settlement under the strip (PLAXIS 2D 0.04845 / 0.04965 m; Gibson 0.05)
plaxis-verif-gibson
0.04951m0.05m−0.98 %Ours 0.04951 mPublished 0.05 m−0.98 %Pass
settlement under the strip (PLAXIS 2D 0.04845 / 0.04965 m; Gibson 0.05)
plaxis-verif-gibson-2011-a
0.04599m0.05m−8.02 %Ours 0.04599 mPublished 0.05 m−8.02 %Deviation
total discharge under the dam (impermeable wall, confined flow)
plaxis-verif-confined-wall-Q
0.81959m³/day/m0.82m³/day/m−0.05 %Ours 0.81959 m³/day/mPublished 0.82 m³/day/m−0.05 %Pass
Dam I: total discharge under the dam
plaxis-verif-concrete-dam-I-Q
1.03e-5m³/s/m1.03e-5m³/s/m+0.15 %Ours 1.03e-5 m³/s/mPublished 1.03e-5 m³/s/m+0.15 %Pass
Dam II (upstream screen): total discharge
plaxis-verif-concrete-dam-II-Q
8.69e-6m³/s/m8.84e-6m³/s/m−1.74 %Ours 8.69e-6 m³/s/mPublished 8.84e-6 m³/s/m−1.74 %Pass
Dam III (downstream screen): total discharge
plaxis-verif-concrete-dam-III-Q
8.69e-6m³/s/m8.84e-6m³/s/m−1.74 %Ours 8.69e-6 m³/s/mPublished 8.84e-6 m³/s/m−1.74 %Pass
Dam I: uplift pressure head at A (dam toe)
plaxis-verif-concrete-dam-I-uplift
2.181m2.25m−3.07 %Ours 2.181 mPublished 2.25 m−3.07 %Pass
Dam II: uplift pressure head at A
plaxis-verif-concrete-dam-II-uplift
2.073m2.13m−2.68 %Ours 2.073 mPublished 2.13 m−2.68 %Pass
Dam III: uplift pressure head at A (upstream side of the screen)
plaxis-verif-concrete-dam-III-uplift
4.004m3.87m+3.46 %Ours 4.004 mPublished 3.87 m+3.46 %Pass
seepage face length along the slope above the tail water
plaxis-verif-seepage-length
25.375m25.5m−0.49 %Ours 25.375 mPublished 25.5 m−0.49 %Pass
seepage face height on the downstream boundary above the tail water
plaxis-verif-muskat
1.5867m1.55m+2.37 %Ours 1.5867 mPublished 1.55 m+2.37 %Pass
total discharge through an unconfined sand layer
plaxis-verif-sand-layer-Q
0.15387m³/day/m0.15m³/day/m+2.58 %Ours 0.15387 m³/day/mPublished 0.15 m³/day/m+2.58 %Pass
303 not shown yet; search or filter to find a row.

Δ = (ours − reference) / reference, in percent, as recorded in the registry; where a row has no reference value it is against the vendor-published value and says so. Each status is the registry’s own. The “Self-check” label follows the registry’s description of the reference.

Bring a job you have already signed.

Run it here and set our number beside yours. 14 days free, no card.