I am not sure where to begin. But let's start with EVC.
EVC Baseline is expected to be 20% better than AVC High Profile, and EVC Main Profile is expected to be 30% better than HEVC. The Codec is backed by Huawei, Samsung and Qualcomm. If you are unfamiliar with the Smartphone market, The three would represent roughly 80% of marketshare excluding Apple.
Now on to VVC. I am surprised by the sentence
MC-IF has 31 members, 7 of which are licensing entities (i.e. a little less than ¼ of all members). The “industry” members account for just ½ of the HEVC patent holders. It is hardly believable that VVC will fare better than HEVC. It could very well fare worse because VVC adoption in broadcasting will take years.
Well Yes. That is because 7 of those licensing entities already covered most if not all of the industry. ( Excluding a few Open Media Alliance Member of course ) The MC-IF actually includes ALL of the current HEVC patents holders. That is HEVC Advance and Velos Media, which is basically Samsung and Qualcomm, along with many others that were not in any HEVC patent pool due to disagreement in the first place.
( And If you notice the removal of the infamous Technicolor, they sold their patents to another entity that is inside MC-IF, but I cant remember which one on top of my head )
VVC is expected to be 50% more efficient than HEVC. And judging from its reference encoder, this is the first time since AVC / H.264 era a video codec that might actually live up to its claims. ( Normally marketing likes to use unrealistic claims ) It really is the state of the art Video Codec, at a decoding complexity that is quite manageable. ( Lower than AV1 )
So what does all that means? Samsung, Huawei and Qualcomm are also inside MC-IF ( represented by different groups ). My guess is that EVC is basically a backup plan or a gesture to MC-IF, if the licensing deal can be agreed upon, they will go with EVC.
As a video codec enthusiast, I am extremely excited with both VVC and EVC.
As to MPEG ( Not to be confused with MPEG-LA ), I am not quite sure why he said it is dead. I reread the article a few times and still dont quite understand it. May be I am missing some context?
> It really is the state of the art Video Codec, at a decoding complexity that is quite manageable. ( Lower than AV1 )
I am afraid this has not been substantiated by any of the public decoder demonstrations I've seen. Please see the most recent VVC technical update presented at the MC-IF meeting on March 2nd of this year:
On slide 10 is a graph of VVC performance showing the VTM (VVC) decoder at 2.0x the complexity of HM (HEVC). On slide 12, the Ittiam production decoder boasts 1920x1080 @ 24fps on a 4-core Cortex A75 @ 2.5GHz.
Compare that with this recent study of dav1d (AV1) decoder complexity on a broad set of mobile SOCs, where 1920x1080 @ 24fps was easily reached by a Google Pixel 1 from 2016! Using just the two LITTLE cores! Even higher frame rates were achieved with more modern devices:
Yes. But I think it should be noted the HM and VTM are reference encoder and decoder, they are not meant to be used for production nor in any way optimised.
> But I think it should be noted the HM and VTM are reference encoder and decoder, they are not meant to be used for production nor in any way optimised.
Sure, but my comparison was to the Ittiam production decoder.
I think that is a very partial reading of the forces at play.
Apple may be 20% of unit sales, but an overwhelming majority of industry profits and has a lock on the best demographics (the rich and the young). AFAIK they haven’t decided, even if historically they tended to prefer ITU/MPEG standards (but boycotted Blu-Ray).
AV1 has Netflix and Google/YouTube behind it. Not sure about Amazon, but they own Twitch which is on the AV1 side. In an era where the content providers also control the app through which it is delivered, they have the control over which codec gets chosen, not the hardware manufacturer. They may tactically choose to use hardware-accelerated standards like HEVC when trading off the royalties they have to pay vs. using less battery life on their customers’ devices, but in the long term it seems to me AV1 will prevail
They want VVC. After all MC-IF was co-founded by Apple. But that doesn't mean MC-IF will succeed. Having everyone on the table does not necessarily mean positive results will come.
EVC Baseline is expected to be 20% better than AVC High Profile, and EVC Main Profile is expected to be 30% better than HEVC. The Codec is backed by Huawei, Samsung and Qualcomm. If you are unfamiliar with the Smartphone market, The three would represent roughly 80% of marketshare excluding Apple.
Now on to VVC. I am surprised by the sentence
MC-IF has 31 members, 7 of which are licensing entities (i.e. a little less than ¼ of all members). The “industry” members account for just ½ of the HEVC patent holders. It is hardly believable that VVC will fare better than HEVC. It could very well fare worse because VVC adoption in broadcasting will take years.
Well Yes. That is because 7 of those licensing entities already covered most if not all of the industry. ( Excluding a few Open Media Alliance Member of course ) The MC-IF actually includes ALL of the current HEVC patents holders. That is HEVC Advance and Velos Media, which is basically Samsung and Qualcomm, along with many others that were not in any HEVC patent pool due to disagreement in the first place.
( And If you notice the removal of the infamous Technicolor, they sold their patents to another entity that is inside MC-IF, but I cant remember which one on top of my head )
VVC is expected to be 50% more efficient than HEVC. And judging from its reference encoder, this is the first time since AVC / H.264 era a video codec that might actually live up to its claims. ( Normally marketing likes to use unrealistic claims ) It really is the state of the art Video Codec, at a decoding complexity that is quite manageable. ( Lower than AV1 )
So what does all that means? Samsung, Huawei and Qualcomm are also inside MC-IF ( represented by different groups ). My guess is that EVC is basically a backup plan or a gesture to MC-IF, if the licensing deal can be agreed upon, they will go with EVC.
As a video codec enthusiast, I am extremely excited with both VVC and EVC.
As to MPEG ( Not to be confused with MPEG-LA ), I am not quite sure why he said it is dead. I reread the article a few times and still dont quite understand it. May be I am missing some context?