[x265] using and encoding of h265 parameter set fields

[Steve Borho]

On Mon, May 5, 2014 at 1:06 PM, dave <dtyx265 at gmail.com> wrote:
> Many parameter set fields are encoded in a form that is modified I am
> guessing for the purpose of maximizing the range of possible encoded values
> but are probably more useful for the encoding of frames in their unmodified
> forms.  For example, both the VPS and SPS have a field called
> m_maxSubLayersMinus1 but m_maxSubLayers may be more useful for analyzing and
> encoding of frames.

As far as I know, the +1 and -1, etc are there to make the most likely
values as small as possible to save bits

> When the unmodified form is more useful, I think it would be less confusing
> to use the unmodified form with the variable name without the modification
> and then modify the field at the time of encoding. That is, use
> m_maxSubLayers, then at the time of encoding in TEncSbac, encode
> m_maxSubLayers - 1.

Agreed, it would be best for these +1/-1 offsets to be applied when
generating the bitstream.

> This would break from naming h265 fields after the names used in the
> specification but removes any thought of "Do I need to add one to
> m_maxSubLayersMinus1 before using it.." or "do I need to subtract 1 before
> encoding it.."  I don't think removing MinusX or PlusX from a field name is
> going to hurt too much, especially if it's documented in comments at it's
> declaration.
>
> Of course some fields may be more useful in their modified form. Either way,
> I prefer to have the name of the field match the state of the value it
> holds.  If anyone knows of any particular field that should be treated one
> way or the other please let me know.

I'm not aware of any off-hand.

--
Steve