Known limitations ================= These are the open physics items a user should know about before trusting a number. Each links to the GitHub issue that tracks it. - **The PSF is an Airy pattern for the full 3 m pupil.** The encircled energy of the photometric aperture and the automatic peak-pixel fraction do not include the Lyot stop, which widens the PSF, or pointing jitter. Both lower the encircled energy, so integration times are optimistic. Whether the focal-plane mask throughput table is a total or an in-core throughput is also still open. `#30 `_ - **The raw contrast is one number.** It does not vary with separation, wavelength or bandwidth, no inner or outer working angle is enforced, and the off-axis throughput is read from the table with the nearest point rather than interpolated. `#32 `_ - **Saturation only considers the planet's aperture.** The brightest pixel on the detector is near the mask edge or in the leaked stellar core, where the contrast is far worse than in the dark hole. For a bright host the detector-safe frame time is much shorter than ``saturation_time()``. `#33 `_ - **Nothing has been checked against measured data.** The detector parameters are digitized from the manufacturer's plots, and the vortex throughput curve comes from the literature. Cross-checks against measured detector, vortex and on-sky data are tracked in `#34 `_.