Wake me up with h.265
H.264 is supported in WebRTC from Chrome 50
31–40 of 49 posts
Re: H.264 is supported in WebRTC from Chrome 50
#32Earlier quoted context omitted.
Note that OpenH264 is terrible and probably isn't viable for anything besides video conferencing and/or screen sharing. A couple of things from their github issues: It's much slower than x.264[1] It only supports baseline profile and the developers aren't that interested in going beyond that[2] IIRC this quality was lower than x.264 even when set to baseline profile... I'll see if I can dig up some statistics. [1] ht…
That's a pointlessly inflammatory statement, particularly in the absence of any explanation of what you're referring to. Are you referring to code quality, the fact that it currently only supports the baseline profile, etc?
The tuning is and will always be far worse than x264, because no commercial team will ever have the motivation to beat the years of video encoding maniacs' free time spent on it.
Plus the Cisco employees probably aren't any good at Touhou games. Presumably they're pretty good at tuning video calls though.
Re: H.264 is supported in WebRTC from Chrome 50
#33I'm surprised I have not come across OpenH264. Why is Cisco eager to pony up the license fees for users of their H264 codec?
Note that OpenH264 is terrible and probably isn't viable for anything besides video conferencing and/or screen sharing. A couple of things from their github issues: It's much slower than x.264[1] It only supports baseline profile and the developers aren't that interested in going beyond that[2] IIRC this quality was lower than x.264 even when set to baseline profile... I'll see if I can dig up some statistics. [1] ht…
You are right, it's not a good idea for anything other than video conferencing because of being Baseline only. I don't think that makes it "terrible" - WebRTC is only video conferencing.
That said, both Firefox and Chrome prefer VP8.
Re: H.264 is supported in WebRTC from Chrome 50
#34Earlier quoted context omitted.
That's a pointlessly inflammatory statement, particularly in the absence of any explanation of what you're referring to. Are you referring to code quality, the fact that it currently only supports the baseline profile, etc?
The tuning is and will always be far worse than x264, because no commercial team will ever have the motivation to beat the years of video encoding maniacs' free time spent on it. Plus the Cisco employees probably aren't any good at Touhou games. Presumably they're pretty good at tuning video calls though.
Yeah, my comment now looks a bit weird after yaur silently edit the original post to add everything after the first sentence.
I still think “terrible” is a bad choice of wording – “slow” would be an objective statement of fact and the fact that it's within a factor of 2-3 of x264 is worth knowing but hardly enough to say nobody could use it.
Re: H.264 is supported in WebRTC from Chrome 50
#35interesting that there's an encoder built in as well. wonder if they will start offloading youtube video compression onto client machines pre-upload.
google keeps original file you upload
There are youtubers that uplaod 20GB extremely high bitrate files for their 1hour shows, and google lets them retrieve original files on demand.
It sounds crazy given amount of video material being uploaded to YT 24/7, no wonder Google wants new type of spinning hard drives, they must pay tens of millions per year for drives alone..
Re: H.264 is supported in WebRTC from Chrome 50
#36What kind of computer would one need to encode H.264 1080p in real time? (I assume it's all done on the CPU, not the GPU)
Depends on bitrate. I mean, my old Lumia 920 with a underpowered dual-core ARM CPU could encode 1080p at 10Mbits/sec from the camera module - really shouldn't be an issue for any modern PC to have video-conference quality video.
Some of the speed comes at a cost of lower compression. Not that big a deal when it only has to got from camera to SD card, but important when it needs to get transmitted across the mobile internet in real-time.
Re: H.264 is supported in WebRTC from Chrome 50
#37Earlier quoted context omitted.
That's a pointlessly inflammatory statement, particularly in the absence of any explanation of what you're referring to. Are you referring to code quality, the fact that it currently only supports the baseline profile, etc?
The tuning is and will always be far worse than x264, because no commercial team will ever have the motivation to beat the years of video encoding maniacs' free time spent on it. Plus the Cisco employees probably aren't any good at Touhou games. Presumably they're pretty good at tuning video calls though.
x264 was and is a very good encoder for H.264, but it sounds a bit extreme to say that no one can ever beat it - real time streaming was never its primary use case. I think the real lack of motivation to improve it is because most of the interest is now in newer video formats, like VP9. Also, many of the developers of x264 were paid by the commercial licensing program, so I wouldn't say it didn't have any financial motivation.
Also, I'm proof that x264, in fact, does not have a monopoly on Touhou-playing video codec developers :)
Re: H.264 is supported in WebRTC from Chrome 50
#38I'm surprised I have not come across OpenH264. Why is Cisco eager to pony up the license fees for users of their H264 codec?
AFAIK they don't. There is a cap on licensing for H264. Cisco is already hitting that limit with their commercial offerings, so they can essentially offer OpenH264 at zero (licensing) cost.
Supposedly offering OpenH264 ended up costing Cisco more money in licensing fees. See Monty Montgomery's blog post from 2013 about the initial announcement of OpenH264:
http://xiphmont.livejournal.com/61927.html
There are a couple of comments in the comments section titled "Cisco's Costs" in which Monty says that someone he knew at Cisco told him that they had been under the licensing cap and that the OpenH264 project would increase their licensing costs.
Re: H.264 is supported in WebRTC from Chrome 50
#39Earlier quoted context omitted.
The tuning is and will always be far worse than x264, because no commercial team will ever have the motivation to beat the years of video encoding maniacs' free time spent on it. Plus the Cisco employees probably aren't any good at Touhou games. Presumably they're pretty good at tuning video calls though.
x264 was and is a very good encoder for H.264, but it sounds a bit extreme to say that no one can ever beat it - real time streaming was never its primary use case. I think the real lack of motivation to improve it is because most of the interest is now in newer video formats, like VP9. Also, many of the developers of x264 were paid by the commercial licensing program, so I wouldn't say it didn't have any financial m…
The commercial licensing for x264 came about well after it was largely "finished" (there's still ongoing work being done, but no major improvements since 10-bit support in 2010). The idea of maybe selling it one day never really came up when all of the work that makes x264 interesting was being done.
Re: H.264 is supported in WebRTC from Chrome 50
#40I'm surprised I have not come across OpenH264. Why is Cisco eager to pony up the license fees for users of their H264 codec?
The blog announcing it can be found here: http://blogs.cisco.com/collaboration/open-source-h-264-remov...
Cisco sell a number of products that use H264 and so were already paying some fees (although not necessarily the max license level) for the use of this codec.