Linked by Thom Holwerda on Mon 25th Mar 2013 21:09 UTC
Legal Late last week, Nokia dropped what many consider to be a bomb on the WebM project: a list of patents that VP8 supposedly infringes in the form of an IETF IPR declaration. The list has made the rounds around the web, often reported as proof that VP8 infringes upon Nokia's patents. All this stuff rang a bell. Haven't we been here before? Yup, we have, with another open source codec called Opus. Qualcomm and Huawei made the same claims as Nokia did, but they turned out to be complete bogus. As it turns out, this is standard practice in the dirty business of the patent licensing industry.
Thread beginning with comment 556656
To view parent comment, click here.
To read all comments associated with this story, please click here.
RE[2]: Big picture...
by lemur2 on Tue 26th Mar 2013 10:56 UTC in reply to "RE: Big picture..."
lemur2
Member since:
2007-02-17

here's a bigger picture for you.

VP8 is and has been inferior to H.264 in every single way. Performance, power savings, adoption, you name it.


VP8 is not inferior in performance to h.264 except for just one factor: encoding speed. In every other respect VP8 can match or exceed h.264 performance.

VP8 is actually less computationally expensive to decode than h.264. This is because VP8 puts much of the "hard work" into the encoder process rather than the decoder.

If by "power savings", you actually meant a hardware decoder versus a software one, be advised that VP8 is a part of the Android Multimedia Supported Formats:

http://developer.android.com/guide/appendix/media-formats.html

Here is a list of ARM SoCs (which are used in mobile phones and tablets) from different manufacturers showing which support VP8 and which do not:

http://wiki.webmproject.org/hardware/arm-socs

Unless you buy Apple gear then your (recent) mobile device is more likely than not to support VP8 decode in hardware. VP8 is getting quite prevalent in terms of adoption, just about every current Android device on the market would support VP8 decode in hardware.

Since VP8 is easier to decode than h.264, and since it now has hardware decoding in mobile SoCs, then VP8 is actually likely to out-perform h.264 in terms of power savings.

Good thing VP9 is "open". Look at the endless amount of information on it and the wide spread adoption. Oh no wait, thats H.265 !


"Open" means royalty-free, anyone may implement it. That is most certainly VP9 and not h.265.

Here is an alpha-version implementation of VP9 you may wish to investigate:

http://news.cnet.com/8301-1023_3-57561111-93/googles-new-vp9-video-...

Reply Parent Score: 4

RE[3]: Big picture...
by saynte on Tue 26th Mar 2013 12:26 in reply to "RE[2]: Big picture..."
saynte Member since:
2007-12-10

[quote]
VP8 is not inferior in performance to h.264 except for just one factor: encoding speed. In every other respect VP8 can match or exceed h.264 performance.
[/quote]

What? Didn't you claim this months ago and couldn't provide any evidence for it?

http://www.osnews.com/thread?542644

The conclusion from most comparisons is that x264 outperforms VP8 encoders in quality/bit. So VP8 is inferior in two ways: encoding speed and quality. Maybe the quality is acceptable, but it's still inferior to what a good H.264 encoder can provide.

I'd be happy to see some direct comparison that shows otherwise, but until then maybe you can stop repeating this unfounded claim?

Reply Parent Score: 6

RE[4]: Big picture...
by henderson101 on Tue 26th Mar 2013 15:12 in reply to "RE[3]: Big picture..."
henderson101 Member since:
2006-05-30

Ouch! Nicely played!

Reply Parent Score: 2

RE[4]: Big picture...
by lemur2 on Wed 27th Mar 2013 05:16 in reply to "RE[3]: Big picture..."
lemur2 Member since:
2007-02-17

The conclusion from most comparisons is that x264 outperforms VP8 encoders in quality/bit. So VP8 is inferior in two ways: encoding speed and quality. Maybe the quality is acceptable, but it's still inferior to what a good H.264 encoder can provide.


Your recollection is utterly inaccurate. The conclusion was that, for video on the web resolution and bitrates, if you are prepared to put up with a slower encoding speed, then for a given number of bits you absolutely can make at least as good if not better encoded video with webm as you can with h.264.

You have to go to extreme high resolution/bitrate/quality profiles before there is any noticeable advantage (other than encoding time) for h.264 that webm cannot match reasonably closely. Such videos simply aren't used over the web.

Edited 2013-03-27 05:22 UTC

Reply Parent Score: 1

RE[3]: Big picture...
by lucas_maximus on Tue 26th Mar 2013 13:29 in reply to "RE[2]: Big picture..."
lucas_maximus Member since:
2009-08-18

VP8 is not inferior in performance to h.264 except for just one factor: encoding speed. In every other respect VP8 can match or exceed h.264 performance.


This is rubbish.

http://en.wikipedia.org/wiki/VP8

In a comparison done in May 2011, the libvpx encoder was found to be slow compared to common H.264 encoders and used up to 213% more data for the same quality video, when used in videoconferencing applications.


over twice as much data! So obviously it is not as good.

Edited 2013-03-26 13:30 UTC

Reply Parent Score: 3

RE[4]: Big picture...
by Thom_Holwerda on Tue 26th Mar 2013 13:31 in reply to "RE[3]: Big picture..."
Thom_Holwerda Member since:
2005-06-29

over twice as much data! So obviously it is not as good.


May 2011 - a benchmark two years old.

I'm not saying anything about whether either of you is right or wrong, but we'll need more recent benchmarks than that, especially considering how fast VP8/9 develop.

Reply Parent Score: 2

RE[4]: Big picture...
by galvanash on Tue 26th Mar 2013 14:21 in reply to "RE[3]: Big picture..."
galvanash Member since:
2006-01-25

In a comparison done in May 2011, the libvpx encoder was found to be slow compared to common H.264 encoders and used up to 213% more data for the same quality video, when used in videoconferencing applications.


over twice as much data! So obviously it is not as good.


I won't argue that you are completely wrong - in best case scenerio tests for VP8 and h.264 I find both codecs are within sptting distance from each other in most measurable metrics, but VP8 loses more battles than it wins.

However, your example is ridiculous. Videoconferencing??? You pick a scenario that h.264 was specifically designed for (low resolution, extremely low bitrate, realtime encoding) and because it is better at that you say VP8 is "obviously it is not as good".

Sorry, but there is nothing obvious about that. You cannot pick one edge comparison and make such a broad generalization.

Again, I am not saying VP8 is better than h.264. I will say that for resolutions and bitrates routinely used for web based video distribution (720p and 480p, 500-1200kpbs) it is definitely close enough in most measurable metrics that most people would not notice the difference.

Besides, frankly I think arguments on the technical merits of VP8 are wasted breathe (for or against). No one uses VP8 because it is technically superior - they use it because it is open and royalty free. The fact that it is actually comparable to h.264 when used for its target use case (web video) is just icing on the cake.

Edited 2013-03-26 14:22 UTC

Reply Parent Score: 3

RE[4]: Big picture...
by lemur2 on Wed 27th Mar 2013 05:38 in reply to "RE[3]: Big picture..."
lemur2 Member since:
2007-02-17

"VP8 is not inferior in performance to h.264 except for just one factor: encoding speed. In every other respect VP8 can match or exceed h.264 performance.


This is rubbish.

http://en.wikipedia.org/wiki/VP8

In a comparison done in May 2011, the libvpx encoder was found to be slow compared to common H.264 encoders and used up to 213% more data for the same quality video, when used in videoconferencing applications.


over twice as much data! So obviously it is not as good.
"

Way, way, way out of date.

You can catch up a bit here, if you care to:
http://blog.webmproject.org/

Reply Parent Score: 1

RE[3]: Big picture...
by ndrw on Wed 27th Mar 2013 00:01 in reply to "RE[2]: Big picture..."
ndrw Member since:
2009-06-30

VP8 is not inferior in performance to h.264 except for just one factor: encoding speed. In every other respect VP8 can match or exceed h.264 performance.


Encoder also happens to a part of the system that is exposed to patent issues. Luckily, it's also a part that is the most flexible in dealing with them - all known issues can be worked around (at worst encoder will be slower) and, even if you inadvertently trip on a patent, you can still control the damage by simply updating the code.

OTOH, it is really difficult to infringe on patents in a specification of the data format, unless the specification mandates such infringement. Some do, but it does not happen by accident - you have to force conforming implementations to use specific patented techniques.

Reply Parent Score: 2

RE[3]: Big picture...
by zima on Mon 1st Apr 2013 22:25 in reply to "RE[2]: Big picture..."
zima Member since:
2005-07-06

Here is a list of ARM SoCs (which are used in mobile phones and tablets) from different manufacturers showing which support VP8 and which do not:

http://wiki.webmproject.org/hardware/arm-socs

Unless you buy Apple gear then your (recent) mobile device is more likely than not to support VP8 decode in hardware. VP8 is getting quite prevalent in terms of adoption, just about every current Android device on the market would support VP8 decode in hardware.

You purposefully misrepresent the situation. Qualcomm SoCs don't support VP8 in hw - and Qualcomm is the gorilla in the room, in great many Android handsets.

Edited 2013-04-01 22:32 UTC

Reply Parent Score: 2