Open Broadcast Encoder - VoD

[kieranrk]

Latest Release: 0.21

Current Information
This is an application designed for Video-on-Demand content encoding. It’s a fork of x264cli with a ts muxing library, audio and closed caption insertion.
A few things are removed such as VFR timecodes in order to eliminate confusion with SMPTE timecodes.

It supports VBR and CBR muxing along with DVB and CableLabs standards. It should comply with T-STD (though audio does need analyzer verification).

It can mux audio using separate files: AC3, MP2 and AAC (both ADTS and LATM). There is a special build of qaac, a quicktime API frontend, which can output LATM if necessary.

There are lots of other rare features such as DVB AU_Information, SCC caption insertion and CableLabs 3D signalling.

You can download binaries at http://code.google.com/p/open-broadcast-encoder/

The OBE wiki has a lot more information about analyzer and STB compatibility: http://obe.wikia.com/

There is a “corporate” site at http://www.ob-encoder.com/

It is hoped that where appropriate features can be returned to x264cli and ffmpeg in due course.
Thanks to all who helped in the creation of this.

Features coming soon
MCC caption insertion for EIA-708 captions - Thanks to CPC for providing the spec.

[pandy]

Thx kieranrk! As a tester hope that this tools solve some of my problems!

[kieranrk]

Thx kieranrk! As a tester hope that this tools solve some of my problems!

What are you trying to do with it?

[b66pak]

thanks a lot...
_

[Blue_MiSfit]

This tool is incredimazing. If I have any say (and I do), it will be used professionally :)

Derek

[pandy]

What are you trying to do with it?


Heh, a lot of things - mostly STB tests - functionality, some hardware testing. Now it is bit weird - struggling with some tools limitations.

btw

any chance for some support for PSI/SI? like text/hex/bin table description and inserting in defined period of time at defined time at defined PID?

[kieranrk]

Heh, a lot of things - mostly STB tests - functionality, some hardware testing. Now it is bit weird - struggling with some tools limitations.

btw

any chance for some support for PSI/SI? like text/hex/bin table description and inserting in defined period of time at defined time at defined PID?

Please post any reports in the wiki of STB compliance.

You can set the PMT and PCR PIDs. The libmpegts api lets you set the PAT and PCR periods so I could expose that in the CLI.
Not entirely sure how to have descriptor insertion work but it's in the TODO list for the libmpegts API.

[pandy]

Please post any reports in the wiki of STB compliance.


Maybe not STB but i can report IC compliance - IMO STB reports not very useful due full custom/not for resale model.


You can set the PMT and PCR PIDs. The libmpegts api lets you set the PAT and PCR periods so I could expose that in the CLI.
Not entirely sure how to have descriptor insertion work but it's in the TODO list for the libmpegts API.

Will be nice, i think that full PSI/SI support is not(? i thinking sometimes about making modulator in FPGA - it will allow use any STB for local reception ie small home network ;) ) possible (NIT,SDT,EIT/TOT/TDT) but anyway all small things are useful.
Thanks in Advance!

[kieranrk]

Will be nice, i think that full PSI/SI support is not(? i thinking sometimes about making modulator in FPGA - it will allow use any STB for local reception ie small home network ;) ) possible (NIT,SDT,EIT/TOT/TDT) but anyway all small things are useful.
Thanks in Advance!

Yeah there are code stubs for NIT/SDT etc which will happen as the broadcast encoder reaches maturity.

There is a software dvb modulator called gbdvb. I've been meaning to make a modulator so i can test on real hardware.
It is possible to use some cheap USB out plug to test STBs with.

[pandy]

Yes, now i think that even some USB to some "DVB like modulation eg C or T" interface can be used (DekTek, Alitronika etc) - many of them are simply analog players - modulation is implemented in software and use CPU as modulator DSP.
BTW my idea was to use GPU (IFFT should be one of this areas that can use modern GPU hardware efficiently)

[kieranrk]

Yes, now i think that even some USB to some "DVB like modulation eg C or T" interface can be used (DekTek, Alitronika etc) - many of them are simply analog players - modulation is implemented in software and use CPU as modulator DSP.
BTW my idea was to use GPU (IFFT should be one of this areas that can use modern GPU hardware efficiently)

Well the IFFT in ffmpeg is blazing fast so that's also a possibility. I don't really know enough about the hardware side to make this happen though.
From what I gather it's a relatively simple process.

[pandy]

From what I gather it's a relatively simple process.
At least it should be simple process - quite well described but...
For a short moment I've started thinking on something similar to this: http://bk.gnarf.org/creativity/vgasig/ and http://bellard.org/dvbt/

[kieranrk]

0.21 stable released today.

[pandy]

Thx Kieranrk!

[kieranrk]

Thx Kieranrk!

BTW, the DTG (UK Digital TV Group) are using this for development of new standards and STB testing, which I think is what you were doing.

[pandy]

Quite close, however mostly we use hardware encoder units - but for me OBE is interesting as auxiliary stream source to test non-common functionality in a bit more easier way which should occurs in next 2 months i hope - limitation only to hardware (SDI and similar sources) can be a issue.