Re: [rtcweb] Interoperability between browsers (MTI Video)

[Hadriel Kaplan <[e-mail]>]

On Mar 16, 2013, at 3:14 PM, Leon Geyser <[e-mail]> wrote:

> --Decide on VP8:
> This would be the ideal choice and there shouldn't be a reason not to implement VP8.

I believe the jury's still out on that.  From what I understand, some of the vendors are concerned that VP8 may still be encumbered by patents outside of MPEG-LA, and besides the details of the RF usage for VP8 from MPEG-LA have still not been disclosed. (e.g., the strings attached to the usage might be unacceptable to some vendors)


> --Decide on H.261:
> This might be old tech, but browsers might be able to use H.261 without paying any royalties and without transcoding.
> The codec can be used at 352x288 resolution from 64Kbit/s to 2Mbit/s
> Remember this is just a fallback for interoperability. Browsers will still implement VP8 and/or H.264.
> ...
> Atleast there is video instead of no Video. Any opinions?

I agree this would be better than nothing, as mentioned on the thread titled "Video Codec MTI & User Experience == 1".
I also asked a few vendors after this week's meetings whether their existing devices support H.261, but they didn't know.  If most existing video devices do, then having H.261 as the MTI would also facilitate interop with legacy equipment.  I.e., it would follow the same rationale for why we chose G.711 as one of the MTI audio codecs.

-hadriel