H2 dissociation on a fixed Cu substrate

SystemCuH2

Atoms—¹

Runs4

Cap10000 force calls. The pinned GPRidge input sets no force-call cap. This acceptance cap copies max_iterations from archive/entries/minimization/lj38/eon-cg/config.ini.

¹ The problem file does not state an atom count.

largest Cartesian force component (the L-infinity norm) at or below 0.01 eV/Angstrom. The pinned input inputs/gpridge/cuh2/params.toml sets finalConvergenceForce to 0.01 and convergenceForceNorm to linfNorm.

4. The pinned input inputs/gpridge/cuh2/params.toml sets sepDists to 0.02, 0.3, 1.0 and 3.0.

The main metric is Force calls. Scoring rules.

Leaderboard

A shorter bar is a lower cost on the log scale for that problem. Select a column heading to sort.

Rank	Code	Best	Median	Success	PAR-2
No ranked row.

The tables below list every ranked row, by method family and overall.

No run is committed for this problem.

Overall

One row is one code, one version, and one method. A lower cost is better. PAR-2 counts a finished run as the calls it used and a failed run as twice the cap. An archived row shows the average it published, and its PAR-2 applies that rule to the failures the entry records. Hessians is the mean analytic Hessian count. FD calls is that count times one force call per coordinate.

Row	Family	Family rank	Unit	PAR-2	Published	Median calls	Mean calls	Hessians	FD calls	Failures	Runs
No ranked row.

Success on this problem: the native search converges, the Cu atoms stay fixed, and the client force calls stay under the cap.

The maximum-norm hydrogen force meets the stop on an independent remote procedure call (RPC).

The Hessian index over the six hydrogen coordinates is 1.

A hydrogen mode is unresolved when its absolute curvature is at most the largest of the force stop, the roundoff floor on the Frobenius norm, and the critical-point floor.

The check links the dimer verification.