Nishi Family › Compare › Computational Mathematics  ·  Frontier Radar  ·  Run-and-Compare Bench

Nishi vs the Field

Computational Mathematics — mechanically measured, liar-killed, sovereign.

Nishi  vs  Wolfram Language · SymPy · SageMath · Maxima

How this is scored. Every Nishi cell is measured: the census reads the real organ source on disk and requires the implementing symbol to exist (no self-grading). Competitor cells record documented capability presence from compute-bar.txt (Wolfram/SymPy/Sage/Maxima docs). Best=exceeds all peers on this axis, Yes=present, Part=partial, No=absent. This measures capability presence, not depth: Wolfram leads depth in nearly every row (10,000+ curated functions + data). Nishi's genuine exceeds are the two sovereignty/determinism axes at the bottom.
CapabilityNishiWolframSymPySageMathMaximaNotes
Arbitrary-precision integersYesYesYesYesYesnx_bigint (bi_add) -- unbounded integers
Exact rationalsYesYesYesYesYesnx_rational (rat_norm/rat_gcd)
Complex numbersYesYesYesYesYesnx_complex (nx_cx_add)
Symbolic differentiationYesYesYesYesYesnx_derive (chain rule); Wolfram deepest
Numeric integrationYesYesYesYesYesnx_integral (Riemann sum)
Symbolic integrationYesYesYesYesYesnx_calc_integrate; Wolfram Risch+ far deeper
Series expansionYesYesYesYesYesnx_calc_series (Maclaurin)
Equation solvingYesYesYesYesYesnx_calc_solve (linear/quadratic); peers solve general
Linear algebra (vectors)YesYesYesYesYesnx_linalg (nx_vec_add)
MatricesYesYesYesYesYesnx_matrix (alloc/ops)
PolynomialsYesYesYesYesYesnx_poly (mul/add/derivative/eval)
Computational geometryYesYesPartYesPartnx_geom (Hough, orientation); SymPy/Maxima partial
Graph algorithmsYesYesPartYesNonx_graphalg; Sage=NetworkX rich; Maxima none
ProbabilityYesYesYesYesPartnx_probability (measure over omega)
Tensors + autodiffYesPartPartPartNonx_autograd_tensor -- reverse-mode autograd; the CAS peers are not autodiff-first
Units / dimensional analysisYesYesYesYesPartnx_units (SI dimensions)
SAT / constraint solvingYesYesPartYesNonx_sat_solver (clauses); Maxima none
Number theory (primality)YesYesYesYesYesnx_triangulation_number_theory; prime proofs
Automated theorem provingYesPartNoPartNonx_atp_to_derive -- ATP to derivation; none is a general ATP
Autoformalization (informal to formal)YesPartNoNoNonx_autoformalize rung 1: words to Peano terms, kernel-checked derivation. Wolfram-Alpha parses NL math (far broader grammar) but emits no checked proofs
Peano / foundational arithmeticYesYesYesYesYesnx_arith (successor/plus/mult from axioms)
Distinguishing axes — where a sovereign integer stack genuinely leads
Sovereign, bits-up (own language+compiler+syscalls, zero third-party runtime)BestNoNoNoPartNishi computes with NO Python/C/Lisp/cloud beneath; Wolfram proprietary+cloud, others large dep stacks
No-float, integer-deterministic by construction (bit-identical)BestPartPartPartPartNishi results are exact-integer + reproducible; peers default to floating point
coverage 1000/1000EXCEEDS 2PRESENT 21ABSENT 0
Honest verdict. Nishi has broad computational presence — a real organ in every core category above, built bits-up in one sovereign language. It is not a depth rival to Wolfram: Wolfram leads depth (curated functions + data) in nearly every row, and the general symbolic engines (SymPy/Sage/Maxima) go deeper on symbolic integration/solving. Nishi's genuine exceeds are exactly two: sovereign bits-up computation and no-float integer determinism. The roadmap is depth, not breadth: deepen symbolic integration/solving and add curated computable data.

Generated by nx_swcompare_compute (sovereign NishiLang organ) — every Nishi cell verified against real organ source on disk; competitor cells from compute-bar.txt. Zero JS, zero trackers. Served bits-up by the Nishi stack.