Icecast dev - Feb 2014 - MPEG DASH

Daniel, I won't make any points regarding DASH's fit with Icecast, but I
wanted to address your statement that there is no purpose for adaptive
streaming.

Cell phone connections are fairly terrible and inconsistent.  Even if you
have a solid signal on a 4G technology, you could walk around a corner and
it will disappear.  What works at 3mbit/s one minute might be 10kbit/s the
next.  Adaptive streaming is most certainly helpful in these situations,
and is one of the reasons I do not exclusively use Icecast for the stations
I work on.

Brad Isbell
brad at musatcha.com
http://www.musatcha.com


On Tue, Feb 11, 2014 at 6:30 AM, Daniel James <daniel.james at
sourcefabric.org
> wrote:
> Hi Thomas,
> > I wrote solid IPR evaluation for a reason.
>
> It might be helpful to ask bitmovin as a starting point, they have
> released the reference code under LGPL:
>
> http://www.bitmovin.net/libdash.html
>
> The LGPL preamble says:
>
> "we insist that any patent license obtained for a version of the
library
> must be consistent with the full freedom of use specified in this
license."
>
> Personally, I don't see the point of adaptive streaming and the implied
> variation in quality. Either the quality you are streaming at is
> acceptable, or you are wasting bandwidth by streaming with too high a
> bitrate, perhaps due to an inefficient codec.
>
> I suppose the reasoning is that drops in quality are preferable to
> playback interruptions, but why tolerate either in a well-designed
> system? Adaptive streaming seems to say "we know the Internet is not
> really cable TV, but we are going to work around that by providing a
> cruddy picture for those times when available bandwidth does not align
> with our business model" :-)
>
> In the Icecast community, we might gain more by promoting Opus as a more
> efficient replacement for MP3 streaming, for example.
>
> Cheers!
>
> Daniel
> _______________________________________________
> Icecast-dev mailing list
> Icecast-dev at xiph.org
> http://lists.xiph.org/mailman/listinfo/icecast-dev
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
http://lists.xiph.org/pipermail/icecast-dev/attachments/20140211/b87efd76/attachment.htm

This document from Qualcomm may help.

http://www.qualcomm.com/media/documents/qualcomm-dash-licensing-commitment

it states :

 Companies will not have to pay royalties or license fees because they have
implemented the DASH Standard using Qualcomm's DASH Essential Patents
within a standalone software application sold and distributed separately
from a product capable of implementing a WWAN standard, even if this
software application is eventually used on a WWAN device.

Yet, I don't know if Qualcomm is the sole patent owner for DASH.

Concerning the purpose of adaptative streaming, just like Brad said, there
is the "unpredictible bandwith drop" use case.

But, like I said earlier, I am more interested in the P2P streaming useage
of DASH : Clients will be able to exchange DASH fragment over WebRTC's
Datachannel, so that less bandwith is consumed on the icecast server.

Y.


On Tue, Feb 11, 2014 at 3:59 PM, Daniel James <daniel.james at
sourcefabric.org
> wrote:
> Hi Brad,
>
> > Cell phone connections are fairly terrible and inconsistent.  Even if
> > you have a solid signal on a 4G technology, you could walk around a
> > corner and it will disappear.
>
> Sure, but if you want to listen to music on a cell phone, you might be
> better off using the FM chip :-)
>
> Cheers!
>
> Daniel
> _______________________________________________
> Icecast-dev mailing list
> Icecast-dev at xiph.org
> http://lists.xiph.org/mailman/listinfo/icecast-dev
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
http://lists.xiph.org/pipermail/icecast-dev/attachments/20140211/8cef7dc4/attachment.htm