Re: [Linphone-developers] Tr : Incoming video garbled (H264, MP4V-ES), no video with VP8

[Bossiel thioriguel]

@Nathan Stratton

My response was not agressive at all :)

- Just to remember you that your company (BlinkMind) sent me several mails because you were willing to work with us (Doubango Telecom) but we said we're not interested. I still have the mails in my box :)

- For sure it's because our code is mess that you're using it in your reciprocate based MCU for websocket support. Don't ask me how I know it.

- I've just registered to this mailing list because someone told me to check your message. Now I'm unregistering as I"ve nothing else to add. bye bye.

---

De : Nathan Stratton <address@hidden>
À : Bossiel thioriguel <address@hidden>
Cc : "address@hidden" <address@hidden>
Envoyé le : Samedi 22 décembre 2012 4h27
Objet : Re: Tr : [Linphone-developers] Incoming video garbled (H264, MP4V-ES), no video with VP8

On Thu, Dec 20, 2012 at 5:19 PM, Bossiel thioriguel <address@hidden> wrote:

Hello,

This is my first message to this maling list and I don't have the previous messages on the thread.

From Nathan Stratton: "BTW, a LOT of things and broken with sipml5, boghe, and imsdroid. They generally talk to each other because they use the same stack, don't look at any interoperability as a sign they are doing things correctly."

This is not true at all:

May have been worth reading more then just the P.S....

 

- sipml5 uses chrome media stack from Google. As authors of the draft I guess Google know how VP8 packetization should be done.

Yes, guess what, the VP8 packetization is correct. Your right, it has NOTHING to do with sipml5. I was not talking about VP8 as something broken with the sipml5 junk, they had nothing to do with VP8 as you pointed out.

 

- the VP8 implementation in Doubango is fully conform to the latest draft. The last time I tested Linphone the VP8 packetization implementation was not up to date

Correct, as I pointed out in my first post! We ended up supporting it and Bria along with the standard in our MCU. Last time I tested sipml5 sip over websocket implementation was not up to date, just to name ONE issue.

- WebRTC implementation from Mozilla Firefox (Nighly) also work with imsdroid and Boghe

Yes, Linphone has its issues, but I still stand my my comments. The doubango clients have MAJOR issues, the code is an absolute mess. However my point was more of an issues with someone saying something is broken because two clients made with the same stack work, but a 3rd client does not is not a good indication that the 3rd client is the one broken.

--
><>
Nathan Stratton
nathan at robotics.net
http://www.robotics.net