Current formulas in 8.5.3.2.1 are: offsetX = ( ChromaArrayType == 0 ) ? 0 : ( mvCLX[ 0 ] & 0x7 ? 2 : 0 ) (8-106) offsetY = ( ChromaArrayType == 0 ) ? 0 : ( mvCLX[ 1 ] & 0x7 ? 2 : 0 ) (8-107) They seem to enforce using an offset even for 4:4:4 with odd mvCLX coordinates.

These formulae should probably read something like: offsetX = ( ChromaArrayType != 1 && ChromaArrayType != 2 ) ? 0 : ( mvCLX[ 0 ] & 0x7 ? 2 : 0 ) (8-106) offsetY = ( ChromaArrayType != 1 ) ? 0 : ( mvCLX[ 1 ] & 0x7 ? 2 : 0 ) (8-107) to properly account for 4:2:0 and 4:2:2.

Note: there does not seem to be an appropriate version for the SCC text.

The spec says that valid range for offset_len_minus1 is [0, 31]. At the same time, offset_len_minus1=0 means that there is a slice segment of 1 byte size (entry_point_offset_minus1 == 0). But this is not possible coz every slice segment should contain at least 9 bits according to the clause 9.3.2.6.

Thus the valid range for offset_len_minus1 should be described as [1, 31] and valid range for entry_point_offset_minus1 is [1, 2^{32 - 1].}

(on behalf of pavel.evsikov@…)

TEMPORAL_SUBSAMPLE macro in TEncGOP.cpp is leftover probably after cleaning the code for temporal subsampling, since the macro is not defined the code is not active. Macro should be removed.