Specify CENC #827
Specify CENC #827
Conversation
Fix 820
|
@felicialim |
index.bs
Outdated
| url: https://www.iso.org/standard/84637.html#; spec: CENC; type: dfn; | ||
| text: cenc | ||
| text: cbcs | ||
|
|
There was a problem hiding this comment.
| url: https://www.iso.org/standard/84637.html#; spec: CENC; type: dfn; | |
| text: cenc | |
| text: cbcs |
Co-authored-by: sunghee-hwang <97494915+sunghee-hwang@users.noreply.github.com>
According to the ISOBMFF spec, (whole-block) full sample should be used for audio. I've added some text to clarify it, please take a look. |
| @@ -1938,7 +1949,7 @@ NOTE: In a typical case, the OBUs in the first [=Descriptors=] of an [=IA Sequen | |||
| A file conformant to this specification satisfies the following: | |||
| - It SHALL conform to the normative requirements of [[!ISO-BMFF]]. | |||
| - It SHALL have the <dfn export for="ISO-BMFF Brand">iamf</dfn> brand among the compatible brands array of the FileTypeBox. | |||
| - It SHALL contain at least one track using an [=IASampleEntry=]. | |||
| - It SHALL contain at least one track using an [=IASampleEntry=], possibly transformed by encryption as specified in [[#commonencryption]]. | |||
There was a problem hiding this comment.
Strictly speaking an "old" implementation of v1.0 could have assumed that files would never have encrypted IAMF tracks. I wonder if this warrants a new brand to signal the possible encryption.
There was a problem hiding this comment.
We already have box presence to signal CENC presence. We decided to not require new profiles to use CENC. If there are other examples of codecs that use brands to signal encryption presence we can do the same.
index.bs
Outdated
|
|
||
| [=IA Track=]s MAY be protected. If protected, they SHALL conform to [[!CENC]]. | ||
|
|
||
| When the protection schemes [=cenc=] or [=cbc1=] are used, an [=IA Track=] SHALL be protected using full sample encryption. When the protection schemes [=cens=] or [=cbcs=] are used, an [=IA Track=] SHALL be protected using whole-block full sample encryption. |
There was a problem hiding this comment.
This is a lot of options, and therefore a lot of testing, and possible market fragmentation. Can't the group agree on one approach only?
There was a problem hiding this comment.
The problem is CENC is mandatory in ISOBMFF, but CBCS is more widely used. We could pretty easily reduce it to two. Reducing it to one might be more difficult.
Fix #820
Preview | Diff