V-Nova a new video codec 50% more efficient than hevc

[roo1234]

http://www.v-net.tv/perseus-does-not-disrupt-compression-ecosystem-or-hevc-upgrade-plans

HEVC proved unable to compete in video quality

News from May?

It seems all these articles are the same, a lot of biz talk and nothing concrete to show. Since Apr 1st there is no advance in proving what is being claimed. I wonder why.

[Stears555]

http://www.v-net.tv/perseus-does-not-disrupt-compression-ecosystem-or-hevc-upgrade-plans

HEVC proved unable to compete in video quality

News from May?

It seems all these articles are the same, a lot of biz talk and nothing concrete to show. Since Apr 1st there is no advance in proving what is being claimed. I wonder why.

Because you don't read the news.

[vhkuho]

Always again amazing when groundbreaking new technologies are announced around the 1st of April.

[Stears555]

Always again amazing when groundbreaking new technologies are announced around the 1st of April.

A new video:


https://www.youtube.com/watch?v=zYaeJAk0WXo

[LigH.de]

Are you still feeding Gremlins after midnight?

[badyu17]

Interesting comment mentioning V-Nova here (watch from 22:20)

After listening to the AAC talk, is there something in the video world comparable to SBR (Spectral band replication) that could be used?

So it seems like what V-Nova is doing is fairly similar to SBR. People definitely are playing with the idea of downscaling the video and having a guided upscaler, which is essentially how SBR works in audio.

[LigH.de]

Finally, a little bit of substance. Now we can possibly even predict academic samples when it will be suitable or when it may fail...

[KarMa]

I've already invested millions into the new "Shadow" codec that is said to be able to compress HD movies into a few kilobytes. So I don't have time for this subpar V-Codec or the ancient HEVC. Can't wait to keep all my movies on a floppy.

http://www.cyborg.co/#nicolas-dupont
http://www.cyborg.co/shadow/

[Cornucopia]

They just keep coming out of the woodwork!

Scott

[Stears555]

Here is the reality

https://www.youtube.com/watch?v=TmAMAZzJWDk

[Cornucopia]

So...?

Boring, nothing new except "partnership annoucements", no "HOW". No meat, just fluff.

And it wasn't shot well - cameraman had auto iris turned on or something (exposure fluctuates). Hope that wasn't the PERSEUS codec!

Don't send me PR links anymore via PM unless it is one that actually provides scientific explanations.
<edit>For those who haven't figured it out: This is a repeat of a reply to an unsolicited PM sent to me from @Stears555, who I have now put on my ignore list.</edit>

Scott

[Stears555]

So...?

Boring, nothing new except "partnership annoucements", no "HOW". No meat, just fluff.

And it wasn't shot well - cameraman had auto iris turned on or something (exposure fluctuates). Hope that wasn't the PERSEUS codec!

Don't send me PR links anymore via PM unless it is one that actually provides scientific explanations.

Scott

Scientific explanations? Most of the professional cameramen , video editors and video technicians have no clue about higher mathematics....
So why is the scientific or higher mathematician level is important for you? Can you understand the higher math?

[smrpix]

Scientific explanations? Most of the professional cameramen , video editors and video technicians have no clue about higher mathematics....
So why is the scientific or higher mathematician level is important for you? Can you understand the higher math?

I suspect you're challenging the wrong guy on this. He probably can.

[roo1234]

Scientific explanations? Most of the professional cameramen , video editors and video technicians have no clue about higher mathematics....
So why is the scientific or higher mathematician level is important for you? Can you understand the higher math?

I suspect you're challenging the wrong guy on this. He probably can.

this stears person thinks this repeated mkt bs can convince anyone in this forum.
show specs, compression techniques, and most important real comparison tests with bitrate X subjective quality, for example with HEVC side by side.

[Stears555]

Scientific explanations? Most of the professional cameramen , video editors and video technicians have no clue about higher mathematics....
So why is the scientific or higher mathematician level is important for you? Can you understand the higher math?

I suspect you're challenging the wrong guy on this. He probably can.

this stears person thinks this repeated mkt bs can convince anyone in this forum.
show specs, compression techniques, and most important real comparison tests with bitrate X subjective quality, for example with HEVC side by side.

I must repeat: Most of the professional cameramen , video editors and video technicians have no clue about higher mathematics....

[Cornucopia]

That's all you seem to do @Stears555: repeat BS ad nauseum.

Your pronouncements have no basis in fact. Or you are hanging around the wrong sort of video people. Most video people I know (including myself) have bachelors or masters levels in communications, engineering, math & physics. And we completely understand ALL that is described in the specification(s). In fact, some of them are the very same people that DESIGNED the systems, using "higher mathematics", that you lamely attempt to align with.

If you have something important to contribute, do it. Put up or shut up. I'm waiting...
Otherwise, you are just amounting to a wannabe delusional complaining troll.

Scott

[jaggy]

Germany’s IRT (Institut für Rundfunktechnik) is a serious institution. Their initial findings "confirm a substantial improvement" carry some weight. The link at the top of this page offers good insight into V-Nova. I don't understand the emotional charge regarding this.

http://www.v-net.tv/perseus-does-not-disrupt-compression-ecosystem-or-hevc-upgrade-plans

[KarMa]

Germany’s IRT (Institut für Rundfunktechnik) is a serious institution. Their initial findings "confirm a substantial improvement" carry some weight. The link at the top of this page offers good insight into V-Nova. I don't understand the emotional charge regarding this.

http://www.v-net.tv/perseus-does-not-disrupt-compression-ecosystem-or-hevc-upgrade-plans

I noticed your exceptionally selective quote. Also your website does not speak of the Rundfunktechnik study. So I found their press release talking about it.

"The IRT evaluated the V-Nova’s technology against benchmarks set by JPEG2000 in Contribution and production codecs such as AVC Intra Class 100 and XAVC-Intra100 at comparable bit-rates. The results, based on a mixture of objective and subjective evaluations, confirm a substantial improvement compared to JPEG2000 and a quality that challenges H.264-intra-based production codecs.The tests were performed on HD interlaced content (1080i/25), which represents the large majority of content being acquired today. "

“The tests by the globally respected IRT technology institute shows that PERSEUS provides commercially valuable increases in efficiency over contribution compression codes AVC-Intra and JPEG2000 at the most widely adopted HD format today, 1080i”, commented Guido Meardi, V-NOVA CEO and co-Founder.“Due to PERSEUS hierarchical nature, we know that the benefits are evenlargerat higher resolutionssuch as UHD/4Kand higher frame ratessuch as 1080p50/60. We will continue supporting any independent validation to prove PERSEUS many benefits.”

http://www.v-nova.com/en/press/2015-09/VNL%20Press%20Release_IRT_FINAL.pdf

So in short this IRT test shows it to be a decent Intra-codec, meaning that it can compress single images individually well. But much of data savings of video compression come from I frames being used for reference in a GOP, to remove large amounts of redundant picture information over potentially 100s of frames. Also noticed they did not even talk about HEVC, which not only has superior Intra-compression but also Inner-frame compression (GOPs). Would be interested in seeing go up against FFV1 or Lagarith, as intra compression certainly has its purposes.

As usual, the provided info about the PERSEUS codec is vague at best.

I don't understand the emotional charge regarding this.

They promise the world by saying they can beat HEVC and all they do to prove this is to say they can beat JPEG2000 at Intra-frame compression, and can vaguely compete with H264 Intra. Forgetting HEVC completely. Intra-frame compression would almost be an after thought if I had this miracle codec.


https://en.wikipedia.org/wiki/Snake_oil

[jaggy]

Both news items were linked to previously in this thread. It would appear that being too engaged in the discussion few people actually read these. The CEO of V-Nova has emphasized that "in the real world, PERSEUS is not an alternative codec to H.264 or HEVC but will be used with encoders that also have these compression capabilities." "...From a theoretical technology point of view, PERSEUS can be used as an alternative to HEVC but it is not an alternative in practice. It is something that will be built on top of whatever you were already using.”

[KarMa]

Both news items were linked to previously in this thread. It would appear that being too engaged in the discussion few people actually read these.

You just made it seem like your quote was in the website you linked to. As there was no other description for that website link. As now I know that link was just naked and unrelated to your text.

The CEO of V-Nova has emphasized that "in the real world, PERSEUS is not an alternative codec to H.264 or HEVC but will be used with encoders that also have these compression capabilities." "...From a theoretical technology point of view, PERSEUS can be used as an alternative to HEVC but it is not an alternative in practice. It is something that will be built on top of whatever you were already using.”

Well considering the provocative name of this thread along with the BBC quotes.

"Movie streaming firm Netflix currently requests users of its 4K Ultra High Def service to have a steady 25Mbps broadband connection, with analysis of their video stream showing between 12 and 16Mbps is actually required. V-Nova says it can deliver the same quality picture using just 7-8Mbps. http://www.bbc.com/news/technology-32140732

I don't seem to be the only one that has been mislead. And V-Novas website also seems to forget to emphasize this point of being "theoretical" and "not an alternative in practice". I'm also failing to understand how you go about using PERSEUS if it is to be used in conjunction with AVC/HEVC, and when you do or don't use it and for what reason.

Pretty sad when a brand new Image format called FLIF is already more forth coming with comparison charts, source code, and it's free. And with PERSEUS we have to take all of the news articles and boil them down to the actual substance, which is very little.



Pay no attention to that man behind the curtain

[LigH.de]

If Perseus is a kind of "noise modelling" technology in the video area, vaguely comparable to mp3PRO and HE-AAC in the audio area, it could have beed stated so right away, and we would have understood and complained less about possible "vapour ware"; this approach would be understandable, and the reports about improvements more comprehensible.

[Stears555]

That's all you seem to do @Stears555: repeat BS ad nauseum.

Your pronouncements have no basis in fact. Or you are hanging around the wrong sort of video people. Most video people I know (including myself) have bachelors or masters levels in communications, engineering, math & physics. And we completely understand ALL that is described in the specification(s). In fact, some of them are the very same people that DESIGNED the systems, using "higher mathematics", that you lamely attempt to align with.

If you have something important to contribute, do it. Put up or shut up. I'm waiting...
Otherwise, you are just amounting to a wannabe delusional complaining troll.

Scott

http://www.pocket-lint.com/news/129953-4k-is-only-half-the-story-bbc-also-to-trial-liv...onwealth-games


Perseus proved its ability for British viewers. So handle the quality of Perseus as a fact.

Your argumentation is very weird. Do you believe that the Earth has geoid (or roughly ORB shape) ? Or following your logic: Is the Earth flat, because you have never been on the board of a spaceship to check it....

[poisondeathray]

http://www.pocket-lint.com/news/129953-4k-is-only-half-the-story-bbc-also-to-trial-liv...onwealth-games


Perseus proved its ability for British viewers. So handle the quality of Perseus as a fact.

This article doesn't say anything about perseus or v-nova

If it's a noise modeller - the streams probably won't be very high quality. It won't replace HEVC or AVC. The CEO says so - that's a fact. It looks like it's aimed at a different usage scenario. Low bandwith, lower quality scenarios. Not high quality. So it still might have a place in the world, eg. for bandwith restricted rural areas it could be very useful

[Stears555]

http://www.pocket-lint.com/news/129953-4k-is-only-half-the-story-bbc-also-to-trial-liv...onwealth-games


Perseus proved its ability for British viewers. So handle the quality of Perseus as a fact.

This article doesn't say anything about perseus or v-nova

If it's a noise modeller - the streams probably won't be very high quality. It won't replace HEVC or AVC. The CEO says so - that's a fact. It looks like it's aimed at a different usage scenario. Low bandwith, lower quality scenarios. Not high quality. So it still might have a place in the world, eg. for bandwith restricted rural areas it could be very useful

q
Yes it is important in such a country, where the telecommunication systems are generally backward, like the USA.

[KarMa]

q
Yes it is important in such a country, where the telecommunication systems are generally backward, like the USA.

Not sure what you mean by backward.

But a from what I hear this is an intra-only codec, meaning that it can't remove redundancies across a large number of frames. Like MotionJPEG. Along with the fact that they don't even beat H.264 intra, it makes me wonder why we are even talking about perseus.

[poisondeathray]

But a from what I hear this is an intra-only codec, meaning that it can't remove redundancies across a large number of frames. Like MotionJPEG. Along with the fact that they don't even beat H.264 intra, it makes me wonder why we are even talking about perseus.

It's not intra only - the only released closed testing results by a 3rd party happened to test intra only (or at least didn't reveal the results any of the other tests if they did others )

[KarMa]

It's not intra only - the only released closed testing results by a 3rd party happened to test intra only (or at least didn't reveal the results any of the other tests if they did others )

Apologies as I must not have found all of the eggs in this Easter egg hunt of information. Can someone link to an article where they claim more than just intra. As this would be the first type of test I would do and show off to the press.

[poisondeathray]

It's not intra only - the only released closed testing results by a 3rd party happened to test intra only (or at least didn't reveal the results any of the other tests if they did others )

Apologies as I must not have found all of the eggs in this Easter egg hunt of information. Can someone link to an article where they claim more than just intra. As this would be the first type of test I would do and show off to the press.

Let me rephrase - they never said either if it's intra or inter to my knowledge. The only test results presented (that weren't really even revealed fully, or open, just hearsay really) were against intra compression codecs, AVC-Intra and MJPEG. A 20+ year old codec and a 9 year old codec. Yeehaw. I'm just saying it would be wrong to conclude that it's ONLY capable of intra one from that one test result.

[PuzZLeR]

Ahem...

I interrupt these broadcasts of V-Nova (of which looked "ghosty" to me on their screens) to once more remind its "High Fives" that some of us are still waiting for a decoder/encoder.

Otherwise, it's permanently good bye from me, and hopefully from many others.

No decoder/encoder, no interest.

[david55]

V-Nova’s PERSEUS Powers First UltraHD Broadcast of Real Madrid vs. Barcelona .

https://translate.google.com/translate?hl=en&ie=UTF8&prev=_t&sl=es&tl=en&u=http://www....baja-latencia/

https://promotedstories.com/story/440897

http://www.enhancedonlinenews.com/portal/site/eon/permalink/?ndmViewId=news_view&newsI...st-Real-Madrid