FARPY LABSVerified compute research and public evidence.
RESEARCH

Published findings from verified experiments.

Research records document what FARPY tested, what the evidence showed, and where conclusions stop.

ACTIVE RECORDS
FRS-2026-001

Octane Pixel Determinism Versus Runtime Instability

Output payloads matched, but runtime variance failed the benchmark acceptance gate.

FRS-2026-002

Blender GPU Runtime Reproducibility

Frozen workload contract validation with bounded historical benchmark evidence.

METHOD

Measure

Run controlled workloads.

Verify

Compare evidence against defined criteria.

Publish

Record findings and limitations.