Track remaining original-input parity gaps #624

Scope and counting

At the pinned MATLAB WEC-Sim and Applications revisions, the inventory has 63 published inputs. The accepted-baseline validation badges still count 59/63: RM3 21/21, OSWEC 13/13, Sphere 12/12, and other systems 13/17. The status ledger records 52 unchanged independent-motion cases, five adopted corrected baselines, two stationary-force cases, and four qualified cases. The five corrected baselines have independent Python gates but do not pair their unchanged published trajectories. Thus nine original published inputs remain unpaired under this tracker's closeout rule, including some whose case-specific issues closed with a documented corrected-baseline fallback. A passing opt-in MATLAB numerical replay, refined-step variant, prescribed source event/state, or source-pose visualization does not by itself count as original-input physical parity.

Original-input groups

• RM3 End_Stops (1 input): Establish full-duration RM3 End Stops corrected validation baseline #616
• RM3 MoorDyn visualization (1): Close published RM3 MoorDyn visualization coupled-motion gap #617
• RM3 PTO Extension float and spar (2): Close both published RM3 PTO Extension free-decay gaps #618; physical gate #664, ledger #665
• OSWEC nonlinear visualization (1): Close published OSWEC nonlinear visualization motion gap #619
• OSWEC PassiveYawRegression (1): Close OSWEC PassiveYawRegression physical held-heading gap #620
• OSWEC Variable_Hydro/Passive_Yaw irregular mode (1): Close OSWEC Variable_Hydro passive-yaw irregular 600 s gap #621
• Nonlinear_Hydro ode45 Regular and RegularCIC (2): the bounded core delay investigation Investigate core added-mass Transport Delay under published ode45 steps #599 is closed by Diagnose ellipsoid ode45 nonlinear force timing #666; unchanged published-input physical parity remains open here
• OWC OrificeModel (1): Resolve published OWC OrificeModel piston-force gap #622
• WaveStar NMPC (1): Close WaveStar NMPC portable physical closed-loop gap #623

The eight unchecked groups cover all nine unchanged original inputs still awaiting physical parity, including five with adopted corrected baselines. Four remain uncounted in the accepted-baseline badges. Both PTO Extension published inputs now have independent full-duration physical gates under their inherited limits; their source-step sensitivity and separate corrected fine-step baselines remain documented. The Sphere set already has 12/12 gates. Existing issues #393, #392, and #448 track confirmed or suspected core behavior in inputs that already have numerical gates; they do not replace the closeout work above.

Rule for checking off a group

For each constituent published input, preserve pinned core/Application revisions, hydrodynamic data, wave phases, duration, and settings. Independently advance Python through its public physical API without MATLAB motion, force, solver-step, or heading-event histories as inputs. Require explicit full-duration motion/PTO/output gates and a checked MATLAB baseline; run portable gates where the platform matters. If an upstream source correction is required, pair the corrected maintained published input before changing its badge status. Keep a case qualified while only a bounded prefix, derived source variant, or opt-in source-numerical route passes.

Update PARITY.md, the status CSV, and badge endpoints only from measured results. Close this tracker when the nine currently unpaired inputs meet those criteria, or explicitly document a later user-approved change to the parity target. Do not lower numerical gates merely to move the fraction.