Li Benchmarks

nbody_gravity greenvalidity pass

Tier 2 · wall_time · tier2 default grid · OS linux

passpass — perf claims allowed when ratio is green

Source: latest.csv:perf_present

Wall-clock row exists only — no explicit correctness signal yet.

Ratio vs best competitor: 0.994 (99.4% of julia speed) (table SOTA: julia — best competitor; charts use julia = 1.0)
Best competitor value: 1.4375 s

No analytical verify rows in ingest yet. Run tier-1 with --verify so latest.csv exports verify_ulps / verify_within_1ulp.

Problem size: tier2 default grid (family nbody_gravity)
Category: physics
Pillar: physics
Package: lic
Li / catalog oracle (mean ± σ): 1.4468 ± 0.000223 s (n=2) / 1.4999 ± 0.008296 s (n=2) (cpp)
Ratio vs catalog oracle: 0.9646×
Best competitor: julia (1.4375 s)
Li relative speed vs SOTA: 0.994 (1.0 = julia speed)
Validity: validity pass
Threshold: 1.2×
Source: lic/benchmarks/workloads/tier2_physics/nbody_gravity

PH ids: PH-5b

Compare oracle: cpp

Relative speed vs best competitor (julia) — SOTA = 1.0, higher is better. Absolute s values are in the table below.

Language comparison
Lang	Mean ± σ	Runs	Unit	Variant	OS
li	1.4468 ± 0.000223	2	s	release	linux
cpp	1.4999 ± 0.008296	2	s	release	linux
rust	1.4632 ± 0.015275	2	s	release	linux
julia	1.4375 ± 0.010852	2	s	release	linux
Metric: wall_time (value = mean of timed runs)