BitTorrent unveils new live-streaming platform | Hacker News Reader

BitTorrent unveils new live-streaming platform

theverge.com

43 pointsvincent_s15 comments

Any good streaming solution needs to transcode multiple bitrates and formats. You either need to do that on the source machine or across the network somehow.

Doing it on the source machine is technically easier, but much more heavy weight on machine. Doing it on the network means that the transcode is happening multiple times, which can result in playback issues if you get a source segment from two different transcodes.

I have thought about doing something like this because I work for a streaming company, but have always come to some roadblock like the one above for a "real" application of this technique. I wonder how they have solved this.

> "Powered by our proprietary and patented peer-to-peer live streaming protocol,"

And now I care about this much less...

source: http://blog.bittorrent.com/2016/05/17/bittorrent-live-multic...

Has BitTorrent (the company) ever developed anything that gained wide traction? (Original client was developed pre-incorporation)

Looking through their Wikipedia page, it seems like a list of flops or freeware. Are they actually profitable?

is bittorrent in any way proprietary? Weren't there extensions?

If you control the client, multiple formats are unnecessary. As we transition to gigabit the need for multiple bitrates becomes diminished, but even now I would argue you could get by just fine with just 2. Most TV shows and movies you torrent today are uploaded with 2-3 quality tiers anyway.

> Doing it on the network means that the transcode is happening multiple times, which can result in playback issues if you get a source segment from two different transcodes.

Not necessarily. H.264 Annex G specifies a way to encode lower quality subset bitstreams in a higher quality bitstream. Drop some packets and bam, you've got a lower quality stream.

AAC specifies something similar for audio.

Link to that spec?

Access to codec specs is unfortunately rather difficult.

Wikipedia has a lot of information and some links that may be useful: https://en.wikipedia.org/wiki/Scalable_Video_Coding

This is really good for the sports streaming community.

The current landscape is either a paid site — which is really only http://hehestreams.xyz right now (but it's awesome) — an AceStream link (quality suffers), one of the really janky sports streaming sites where you get blasted with ads, or a YouTube link which likely gets taken down before you can sit down and enjoy it.

AceStream, while good, doesn't have the backing of someone like BitTorrent.

Interesting read: http://www.atlasobscura.com/articles/the-fascinating-local-c...

So how long until hackers reverse engineer and open source it?

The guy says that BitTorrent is half of the internet traffic "by some estimates". Does anyone have sources for this?

Historically, 5-10 years ago, it'd be between 30-40% of total internet traffic. Its come down close to ~7% in the last few years with Netflix, Hulu, and other over the top services being available.

Why isn't this a licensing deal to established content deliverers? It seems to only alleviate provider cost. If this actually does solve a problem, why wouldn't they just work on getting Netflix to buy into it?

Cost has totally different determinants at different scales. At the smallest scales, fixed costs overwhelm variable costs. As you scale up and architectural needs change, there are steps in the cost function. Just because something is cost effective at one scale doesn't mean it's cost effective at all scales.

Personally, I think peer to peer streaming is most interesting for open source projects. You can host static code for free and then use something like this to do streaming. The developer pays nothing, and bandwidth fees are born entirely by the user.

Yet another.

Reply on news.ycombinator.com