[theora] Ogg index and Skeleton 4.0

[Silvia Pfeiffer]

I am assuming the radix field is part of the index procedure? It would
be good to give a brief spec on what it does in the skeleton spec (I
wasn't able to find out what it does).

Monty mentioned that the predelay would be a small positive integer,
so a 16bit unsigned int may be sufficient. Monty, can you confirm?

Cheers,
Silvia.

On Tue, May 4, 2010 at 1:00 PM, Chris Pearce <chris at pearce.org.nz> wrote:
> Not yet, I've only added a 32bit radix and the Role, and Name message header
> fields.
>
> I assume that a 32 bit unsigned integer would be sufficient for the predelay
> field, and that you want it in the fisbone packet? Or can we get away with a
> 16 bit unsigned int?
>
>
> All the best,
> Chris P.
>
> On 4/05/2010 2:18 a.m., Silvia Pfeiffer wrote:
>>
>> Chris,
>>
>> Monty and I discussed an extra skeleton field for Dirac and Dirac-like
>> content today. I'm pretty sure it's the same field that Conrad and
>> Ralph discussed earlier at FOMS. It's a pre-delay field that signifies
>> the PTS/DTS difference in an integer number of packets and helps to
>> more accurately seek in Dirac-type codecs.
>>
>> Have you added such a field already?
>>
>> Cheers,
>> Silvia.
>>
>> On Thu, Apr 29, 2010 at 6:53 PM, Chris Pearce<chris at pearce.org.nz>  wrote:
>>
>>>
>>> On 29/04/2010 4:47 p.m., Chris Pearce wrote:
>>>
>>> I've updated OggIndex to now output Skeleton 4.0 tracks.
>>>
>>> For your testing pleasure, for the next 14 days you can download Firefox
>>> builds which can seek using the indexes here:
>>>
>>>
>>> https://build.mozilla.org/tryserver-builds/cpearce@mozilla.com-try-d1d289c00a8c/
>>>
>>> A good page to compare the difference makes is here:
>>>
>>> http://pearce.org.nz/video/indexed-seek-demo.html
>>>
>>> If you have an already running Firefox instance, you'll need start the
>>> index-capable Firefox build using the --no-remote command line option,
>>> else
>>> it'll just open a new window of your non-index-capable Firefox instance.
>>>
>>> Chris P.
>>>
>>> _______________________________________________
>>> theora mailing list
>>> theora at xiph.org
>>> http://lists.xiph.org/mailman/listinfo/theora
>>>
>>>
>>>
>>
>>
>
>