Re: [Cellar] AD Review: draft-ietf-cellar-ffv1

[Michael Niedermayer <[e-mail]>]

On Thu, Jan 02, 2020 at 05:26:07PM -0600, Adam Roach wrote:
[...]

> 
> §3.8.1.2:
> >  exact contexts used are best described by the
> >  following code, followed by some comments.
> 
> I can't find the referenced comments.

I think this refered to the put_symbol() code below and the comments
inside it.

pseudo-code                                                   | type
--------------------------------------------------------------|-----
void put_symbol(RangeCoder *c, uint8_t *state, int v, int \   |
is_signed) {                                                  |
    int i;                                                    |
    put_rac(c, state+0, !v);                                  |
    if (v) {                                                  |
        int a= abs(v);                                        |
        int e= log2(a);                                       |
                                                              |
        for (i = 0; i < e; i++) {                             |
            put_rac(c, state+1+min(i,9), 1);  //1..10         |
        }                                                     |
                                                              |
        put_rac(c, state+1+min(i,9), 0);                      |
        for (i = e-1; i >= 0; i--) {                          |
            put_rac(c, state+22+min(i,9), (a>>i)&1); //22..31 |
        }                                                     |
                                                              |
        if (is_signed) {                                      |
            put_rac(c, state+11 + min(e, 10), v < 0); //11..21|
        }                                                     |
    }                                                         |
}                                                             |

[...]

> ---------------------------------------------------------------------------
> 
> §4.1.17:
> 
> >             +-------+-------------------------------------+
> >             | value | relationship                        |
> >             +=======+=====================================+
> >             | 0     | Frames are independent or dependent |
> >             |       | (keyframes and non keyframes)       |
> >             +-------+-------------------------------------+
> >             | 1     | Frames are independent (keyframes   |
> >             |       | only)                               |
> >             +-------+-------------------------------------+
> >             | Other | reserved for future use             |
> >             +-------+-------------------------------------+
> >
> >                                 Table 14
> 
> I'm having a hard time understanding "Other" in this table. Section 4.3
> defines "keyframe" to be of "br" type. I'm confused about how a boolean
> variable can take on more than two values.

This table does not describe the keyframe flag but the intra value from
the global header. I think this should already be clear as it refers to
that field "4.1.17. intra " and not the keyframe flag. But we could of
course rename the field if that would make it clearer.


[...]

-- 
Michael     GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB

It is dangerous to be right in matters on which the established authorities
are wrong. -- Voltaire