Changes to the spektrafilm profiles and LUTs shipped in this folder ==================================================================== The spektrafilm profiles, filter data and LUTs in this folder are by Andrea Volpato (https://github.com/andreavolpato/spektrafilm) and are licensed under CC BY-SA 4.0; see SPEKTRAFILM_LICENSE.txt. This file records how Negs' copies differ from upstream, as that license asks. 2026-10: updated to the upstream experimental branch, commit 28bf883 - profiles/: the colour and B&W profiles are verbatim copies of 28bf883's src/spektrafilm/data/profiles. Unchanged. - filters/: the dichroic, heat-absorbing and lens-transmission data are verbatim copies of 28bf883's. Unchanged. neutral_print_filters.json is 28bf883's with Negs-only entries added (below); upstream's entries are unchanged. - luts/spectral_upsampling/: hanatos2025_irradiance_xy_tc.{npy,toml}, arctic2026beta04_reflectance_xy_tc.{npy,toml}, mallett2019_reflectance_xy_tc.{npy,toml} and hanatos_irradiance_xy_coeffs_250304.lut are verbatim copies of 28bf883's src/spektrafilm/data/luts/spectral_upsampling. Unchanged. Earlier Negs versions shipped the hanatos2025 table as irradiance_xy_tc.npy (same bytes) and the arctic2026alpha02 tables, which upstream has since removed. Negs-only data, not from spektrafilm: - profiles/kodak_trix400_d76_8min.json and profiles/ilford_mgrc_deluxe_grade2.json are new preparations from Kodak and Ilford datasheets, in spektrafilm's profile format. Sources and method: profile-sources/bw-expansion/README.md. - filters/neutral_print_filters.json: entries for kodak_trix400_d76_8min on every colour paper under TH-KG3, derived with upstream's own solver (runtime/print_balance.solve_neutral_filter_shifts at 28bf883) by scripts/derive_neutral_print_filters.py, which first re-derives upstream's entries for Double-X and three colour negatives to check the method. B&W papers (kodak_2302, ilford_mgrc_deluxe_grade2) have no colour to balance, so they get no entries and keep the C0 M50 Y50 fallback.