Li Benchmarks

sph_dam_break_2d greenvalidity pass

Tier 2 · wall_time · tier2 default grid · variant shared_c_kernel · 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: 0.999 (99.9% of julia speed) (table SOTA: julia — best competitor; charts use julia = 1.0)
Best competitor value: 0.9566 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 sph_dam_break_2d)
Category: physics
Pillar: physics
Package: lic
Li / catalog oracle (mean ± σ): 0.9576 ± 0.00938 s (n=20) / 1.0041 ± 0.001311 s (n=2) (cpp)
Ratio vs catalog oracle: 0.9537×
Best competitor: julia (0.9566 s)
Li relative speed vs SOTA: 0.999 (1.0 = julia speed)
Validity: validity pass
Threshold: 1.2×
Source: lic/benchmarks/workloads/tier2_physics/sph_dam_break_2d

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	0.9576 ± 0.00938	20	s	release	linux
cpp	1.0041 ± 0.001311	2	s	release	linux
rust	0.9704 ± 0.022491	20	s	release	linux
julia	0.9566 ± 0.004883	20	s	release	linux
Metric: wall_time (value = mean of timed runs)