The index of the first "raw" pass, if bypass mode is on.

The maximum code-block area (width x height). The surface covered by a nominal size block should never be larger than this. It is 4096

The maximum code-block dimension. No code-block should be larger, either in width or height, than this value. It is 1024.

The minimum code-block dimension. The nominal width or height of a code-block must never be less than this. It is 4.

The number of empty passes in the most significant bit-plane. It is 2.

The number of most significant bit-planes where bypass mode is not to be used, even if bypass mode is on: 4.

The number of coding passes per bit-plane. This is the number of passes per bit-plane. It is 3.

The flag bit to indicate that selective arithmetic coding bypass should be used. In this mode, the significance propagation and magnitude refinement passes bypass the arithmetic encoder in the fourth bit-plane and latter ones (but not the cleanup pass). Note that the transition between raw and AC segments needs terminations (whether or not OPT_TERM_PASS is used).

The flag bit to indicate that error resilience info is embedded on MQ termination. This corresponds to the predictable termination described in Annex D.4.2 of the FDIS

The flag bit to indicate that the MQ states for all contexts should be reset at the end of each (non-bypassed) coding pass.

The flag bit to indicate that an error resilience segmentation symbol is to be inserted at the end of each cleanup coding pass. The segmentation symbol is the four symbol sequence 1010 that are sent through the MQ coder using the UNIFORM context (as explained in annex D.5 of FDIS).

The flag bit to indicate that a termination should be performed after each coding pass. Note that terminations are applied to both * * arithmetically coded and bypassed (i.e. raw) passes .

The flag bit to indicate the vertically stripe-causal context formation should be used.

The stripe height. This is the nominal value of the stripe height. It is 4.