lightcraft/docs/arw6-compression.md

Sony ARW6 (Compressed RAW 2)

LightCraft decodes the raw sensor data of Sony's Compressed and Compressed (HQ) ARW files (TIFF Compression 32766, SonyRawFileType 6, "ARW 6.0") written by the A7R VI (ILCE-7RM6) in pure Rust. The shared RAW decoder serves import, desktop previews, editing, CLI rendering, web and export. Before this decoder these files failed the full decode and silently rendered from their embedded JPEG. The ILCE-7M5 (A7 V) writes the same codec; no sample of it was available, so it is unverified.

Decode timings: see the end of this document (measured in the performance pass).

Format and implementation

The codec is Sony's use of SMPTE RDD 34 (LLVC, Low Latency Video Codec). crates/raw/src/llvc.rs is the codec core (bit reader, variable-length decoder, dequantiser, inverse 5/3 wavelet); crates/raw/src/vendor/arw6.rs is the Sony container, geometry and colour reconstruction; crates/raw/src/vendor/arw6_curve.rs is the companding table; arw.rs dispatches Compression 32766 to it. A test-only encoder (arw6_testenc.rs) builds synthetic tiles for round-trip tests.

Container

• Raw SubIFD: Compression 32766, 14 bits per sample, BlackLevel 512 x4 (0x7310), WB 0x7313, crop 0x74c7/0x74c8 (full frame: origin 12,8, size 9984x6656; APS-C: origin 16,10, size 6528x4352), CFA RGGB. The Sony tag-based white level (15360) is not read; the DNG WhiteLevel is (16383).
• The single strip starts with a tile table: u32 count, u32 0, then per tile a u64 offset (relative to the strip) and u32 x, y, width, height. Seen: full frame 4 tiles of 5008x3336 (HQ) or 5024 / 4992 wide (Compressed); APS-C (6592x4372) one tile. A 480-byte block follows the table (not interpreted, not needed).
• A tile is a sequence of 16-byte words. Word 0 is RDD 34 Picture_info (HS = tile width, VS = tile height / 2, BBD = 16, NC = 3, NW = 3, NP = 1000000000b; other values are refused as unsupported). Word 1 holds five u24 group byte-sums, word 2 is zero, words 3-7 describe the groups (a count byte, then u24 stream sizes in words). Groups: g0 = LL3 (3 streams, one per component), g1 / g2 / g3 = level-3 / 2 / 1 bands (3 streams each), g4 = green residual (1 stream). Streams follow in that order from word 8.
• A stream is a header word (a u16 count of index words, a u24 count of data words, then 11 bytes that are not interpreted), an index table and data, each padded to a word. The index has one entry per transform unit (TU): A(16) X B(16) X, where A and B are the byte lengths of the TU's two halves and X are quantiser-index (QI) nibbles (one for g0 and g4, three for g1-g3: HL, LH, HH). Halves per tile = ceil((VS - 2 - s) / 8) + 1 with s the first-half shift (below), TUs = ceil(halves / 2): 105 and 137 for the sample sizes. Halves are numbered n = 2 * TU + h; each is byte-aligned and consecutive halves are contiguous.
• QI values seen, identical in Compressed and HQ files: level 1 component 0 (1,1,2), components 1-2 (2,2,3); level 2 (0,0,1); level 3 (0,0,0); LL3 0; residual 2. Compressed and HQ differ only in the coefficient data.

Entropy coding and dequantisation

• Each line of a band is an RDD 34 section 6.3 coefficient-set stream: ceil(width / 4) sets of four coefficients driven by the DPT / ABS / SGN state machine, with zero runs and the "0^lim" end code. Values past the width are discarded. Lines of a half follow one another bit-contiguously; the state resets per line.
• Two deviations from the RDD 34 text were established by black-box comparison: in state DPTs = 0 with DPT = 0, a "0" code is a zero set that also enters DPTs = 1; and a run code of value V means V - 1 further zero sets before the post-run "0^n 1" depth code.
• Dequantisation is recon(q) = sign(q) * (((2|q| + 1) << (QI - 1)) - (|q| & 1)), recon(0) = 0, and QI = 0 leaves q unchanged. This replaces RDD 34's plain shift.
• LL3 lines are DPCM: a running sum along the line. The 2048 base (codes are unsigned 12-bit) applies to component 0 (the green mean) only; the chroma planes start at 0.
• Check: every half's lines end inside its last byte. The Phase 0 measurement found budget - 8 < consumed <= budget on every half of all 8 files; the decoder rejects a half that does not.

Frame geometry

• Each tile holds three W/2 x H/2 planes (W, H = tile size): component 0 = M, the green mean; components 1-2 = chroma c1, c2; plus a W/2 x H/2 residual plane from g4. A three-level 5/3 wavelet is inverted (vertical, then horizontal) with whole-sample symmetric extension at every edge and integer arithmetic. Tiles are independent.
• The TU grid and the lifting phases are anchored to the full sensor, not to the tile. Let s = (lines in g4's first half) - 2, read from the stream (residual lines are decoded while at least 8 bits of the half remain, at most 8): 0 on full frame, 3 on the APS-C crop (the crop's sensor-row offset 573 is 5 mod 8). Half n then covers plane rows 8n - 6 + s .. 8n + 1 + s, clipped to the frame. The lifting phases follow from s: full frame (0,1,1), APS-C (1,0,0). Each half holds, in clipped natural order, its rows of HL1, LH1, HH1, HL2, LH2, HH2, LL3, HL3, LH3, HH3 and, in g4, 8 residual rows; edge halves are shorter. With the right s every line of every file is used.

Colour reconstruction

With plane row j, column x (x +- 1 clamped at the edges), in 12-bit code units:

• G1[j][x] = M[j][x] - ((res[j-1][x] + res[j-1][x+1] + res[j][x] + res[j][x+1] + 4) >> 3), with res[-1] := res[0].
• G2[j][x] = ((G1[j][x-1] + G1[j][x] + G1[j+1][x-1] + G1[j+1][x]) >> 2) + res[j][x] (unclipped G1; the last row uses G1[j] for G1[j+1]).
• R = 2*c1 + ((clamp(G1, 0, 4095) + clamp(G2, 0, 4095)) >> 1), B = 2*c2 + (same) (a negative green, as in deep high-ISO shadows, counts as 0).
• Every code is clipped to 0..4095. Per 2x2 cell: R (0,0), G1 (0,1), G2 (1,0), B (1,1).

Output range (companding table)

• Codes map through one monotone table shared by all files: identity up to code 1427, then a smooth curve reaching 39002 at code 4095 (steps grow from 1 to 58). No closed form fits within +-22, so arw6_curve.rs embeds the table (2668 u16 values from code 1428). It is not stored in the file.
• Provenance: the table was read off the black-box oracle's (LibRaw, below) output over the 8 Compressed / HQ files; every code from 1027 to 4095 occurred and no two files disagreed. It is therefore a measurement of that program's output, not an independent derivation.
• Unit: 2 x 14-bit DN. The decoder reports black = 2 x 512 = 1024 and white = 2 x the file's DNG WhiteLevel (16383), i.e. 32766; saturated highlights sit at code 4095 = 39002, above white, and clip as usual. If a file carries no usable white level, the white is 32766 (2 x 16383, the table's knot below), not a data-derived value.
• The table has one knot: at code 3977 it runs 32679 -> 32766 -> 32777 (steps +87 then +11), the same in every file. 32766 = 2 x 16383 is exactly the reported white level, so the knot is the white point of the measured curve (inferred from the arithmetic, not from any documentation): codes 3978..4095 sit above white and clip.

Verification

Method: LibRaw master (commit d1f0dd95, 2026-10-08, run as unprocessed_raw, 16-bit output before any processing) was used only as a black-box oracle: it was built and run, and its source was never read. A Python reference model of the format (the Phase 0 reference model, local, not in the repo) was written from RDD 34 and from the oracle's behaviour on real and synthetic input, then reproduced the oracle exactly on every tile of the 8 Compressed / HQ files below. A test-only encoder, validated byte for byte against real tiles, probed the oracle with synthetic coefficients (dequantiser, phases, edge mapping). The Rust decoder reproduces the Phase 0 reference sums on all 8 files.

Geometry was checked against each file's embedded full-resolution JPEG at the DefaultCropOrigin: our output aligns at offset 0 on full frame and APS-C.

LibRaw's APS-C bug

LibRaw assumes the full-frame grid (s = 0) for every tile. On the APS-C crop that drops the first three plane rows, leaves the last three unfed, and shifts its whole output up by 6 sensor rows; its first rows and last rows are also wrong. Our decoder differs from LibRaw there on purpose. To compare against LibRaw on APS-C files, our row y+6 equals LibRaw's row y for 12 <= y < h - 24 (LibRaw's rows 0-11 and the last 18 rows of the shifted overlap differ); on full frame compare directly, the whole frame is exact.

Corpus

File	Source path (RawDB, ILCE-7RM6)	Size (bytes)	SHA-256	Fetched by
arw6-sony-ilce7rm6-apsc-land-hq.arw	arw6/ILCE-7RM6_APSC_land_RAW-C-HD.ARW	29 720 576	b60a7e752d2a8102ccbe6eebcfdbf533c6b0f2545739090873e9e8eb32cc0637	cargo xtask corpus --download
arw6-sony-ilce7rm6-apsc-land-c.arw	arw6/ILCE-7RM6_APSC_land_RAW-C.ARW	30 900 224	1c803304e63ec508b2b5f9d59d4a7e30ed279315b35fa4c5d75f959bed62198b	curl -L
arw6-sony-ilce7rm6-apsc-port-hq.arw	arw6/ILCE-7RM6_APSC_port_RAW-C-HD.ARW	28 758 016	0b6d782571e86868670f613de9fa1f1364623b24fbfd21b704b2c516d7a1c1cb	curl -L
arw6-sony-ilce7rm6-apsc-port-c.arw	arw6/ILCE-7RM6_APSC_port_RAW-C.ARW	29 044 736	c41674a485ffcf6d873e20e545ae62936192d94e95be3d53548e93bc543ecfd4	curl -L
arw6-sony-ilce7rm6-ff-land-hq.arw	arw6/ILCE-7RM6_FF_land_RAW-C-HD.ARW	63 451 136	559fa26501d7813e56afecde2aea3d9965a99228162b194752d986193abb172c	curl -L
arw6-sony-ilce7rm6-ff-land-c.arw	arw6/ILCE-7RM6_FF_land_RAW-C.ARW	63 844 352	d900d6cbe838be864269ad54de4a33fcac7eb3a3d4f91009739f770b8f85a23b	cargo xtask corpus --download
arw6-sony-ilce7rm6-ff-port-hq.arw	arw6/ILCE-7RM6_FF_port_RAW-C-HD.ARW	62 640 128	81c709633cb8f6340e612c3f53bb6ec6b0cccfe5b713c67efb1e977ece9e266d	curl -L
arw6-sony-ilce7rm6-ff-port-c.arw	arw6/ILCE-7RM6_FF_port_RAW-C.ARW	63 168 512	f4448485629c3f82ff1955fd046404350e9a1f09b346038574b89c074d7c50fd	curl -L
arw-sony-ilce7rm6-apsc-land-lossless.arw	raw_modes/ILCE-7RM6_APSC_land_RAW.ARW	44 167 168	7aaf0b3941509d7364a6f03844783155ebc912d246d1c456e5cdb4693ec09fbc	curl -L
arw-sony-ilce7rm6-apsc-port-lossless.arw	raw_modes/ILCE-7RM6_APSC_port_RAW.ARW	43 692 032	0f011933ee96525a39710daa50b59d39702b4ec5934fa0206742e2128c55f366	curl -L
arw-sony-ilce7rm6-ff-land-lossless.arw	raw_modes/ILCE-7RM6_FF_land_RAW.ARW	100 851 712	5cb3c870bc961cf6b079d56c860d0cdf11acd1afd04e5fb8dbe041bcfd2d6201	curl -L
arw-sony-ilce7rm6-ff-port-lossless.arw	raw_modes/ILCE-7RM6_FF_port_RAW.ARW	100 884 480	b1a4026aa6adbaa27973b8b3426d8b4aa0c1cae1845cf5102359327d3b069691	curl -L

These 12 files are CC0 1.0. RawDB's record of the set, https://rawdb.dnglab.org/api/sets/Sony/ILCE-7RM6 (the data behind https://rawdb.dnglab.org/sets/Sony/ILCE-7RM6, accessed 2026-10-10), says "license":"CC0 1.0" for the set (uploaded by abbradar, 2026-08-29) and again for each of the 12 files, next to the SHA-256 values in the table.

They live in the gitignored corpus. cargo xtask corpus --download fetches two of them, one per geometry and compression mode (full frame Compressed, APS-C HQ), and checks each against its SHA-256, pinned in RAW_SAMPLES (xtask/src/main.rs). Fetch the other ten by hand: the server answers 303 to a signed object URL, so curl -L -o corpus/raw/<file> https://rawdb.dnglab.org/api/download/Sony/ILCE-7RM6/<source path>, then compare with the SHA-256 above. cargo test -p lightcraft-raw corpus_sony -- --nocapture checks the decoder's sensor sum and position-weighted sum of the two pinned files against the Phase 0 reference sums and skips absent files.

Nine further full-frame HQ photos from the maintainer's A7R VI were compared with the oracle locally; they are private and never committed or published. All 9 private FF photos (ISO 100-12800, incl. two near-black night shots) decode bit-exact over the full frame (0 mismatching pixels of 66.8 MP each).

Unconditional synthetic tests (encode -> decode round trips for both geometries, the RDD 34 worked example, the dequantiser, hostile and truncated input) run without the corpus.

Coverage

Coverage	Status
ILCE-7RM6 Compressed and HQ, full frame, landscape and portrait	4 files, exact against the reference sums
ILCE-7RM6 Compressed and HQ, APS-C crop, landscape and portrait	4 files, exact against the reference sums (interior-exact against LibRaw after the 6-row shift; LibRaw's own edge rows differ)
ILCE-7RM6 lossless (existing LJ92 path)	4 files decode; not in the corpus test
Private A7R VI HQ photos	9 files, bit-exact over the full frame against the oracle (local)

Not covered

• The 12-bit Compressed mode (electronic shutter with continuous drive): no sample exists, so it returns Unsupported unless its headers prove identical.
• The ILCE-7M5: same codec, no sample; unverified.
• The x2 scale is an assumption backed by shadow-floor evidence (the darkest pixels of all eight ISO 100 files sit at 1037..1059, just above 2 x 512, and the lossless sibling's shadows match at x2; x2.38 would put them 76 DN below black). A tripod-matched lossless / compressed pair would confirm it. If wrong, only the reported black and white levels change.
• Tiles whose plane width is not a multiple of 8.
• First-half shifts other than 0 and 3 (other crops / bodies): the geometry is generic and round-trip tested, but no real sample exists.
• Memory: one FF decode peaks at about 896 MB RSS (about 6.7 x the 134 MB output; the lossless path peaks at 249 MB). Follow-ups: dequantise straight into bands, place tiles as they arrive.
• Colour: the ILCE-7RM6 bundled colour profile and Sony's embedded lens corrections are separate work.

Sources

• SMPTE RDD 34:2015 "LLVC (Low Latency Video Codec) for Network Transfer", https://pub.smpte.org/latest/rdd34/rdd34-2015.pdf (accessed 2026-10-09): picture / TU structure, VLD, inverse quantisation, 5/3 lifting with symmetric extension.
• LibRaw master d1f0dd95, run only as a black-box oracle (not a product, test or build dependency).
• The embedded JPEG of each file, for the geometry check.
• Not read, by licence: LibRaw, rawspeed, dnglab, RawTherapee and other ARW6 decoder sources. Exact byte layouts, deviations, geometry and the table were established through sample analysis and black-box comparison.

Timings

Raw decode only, 4 CPUs, release build, best of 3 wall-clock (2026-10-10). Method: a small throwaway Rust program linking lightcraft-raw reads the file into memory, then times lightcraft_raw::decode alone with std::time::Instant (no render, no JPEG encode), three decodes in one process; the best is reported. Files were already in the page cache.

File	Size	Sensor	Decode
ILCE-7RM6 FF, Compressed (HQ)	63.5 MB	10016x6672	0.45 s
ILCE-7RM6 FF, Compressed	63.8 MB	10016x6672	0.51 s
ILCE-7RM6 APS-C, Compressed (HQ)	29.7 MB	6592x4372	0.35 s
ILCE-7RM6 APS-C, Compressed	30.9 MB	6592x4372	0.34 s

The 1.0 s budget for the full-frame file is met. Parallelised: rayon over the tiles, over the 13 streams of each tile and over its 3 components.