[zerocrates]

I don't really understand what the point of EVC is. From my understanding it's really two standards, one royalty-free that's slightly worse than HEVC/H.265, and one patent-encumbered/licensed that's slightly better than H.265.

The post here says "EVC is promising because it provides a quality that is comparable with or better than AV1, although less than VVC. EVC may have a chance if a licence will be published. However, this has not happened yet." I can only assume, from those numbers above, that "comparable with or better than AV1" applies only to the encumbered/enhanced variant.

It's hard to see why anyone would bother to implement the "base" standard vs. the already widely-deployed AVC/H.264, nor the "enhanced" one which seems to be roughly comparable to AV1 but with licensing costs attached (and as the post points out, no certainty at all about what that licensing would actually entail). Apart from those companies that hold the relevant patents, of course.

[ksec]

The widely deployed AVC is still AVC Main Profile, which is what Youtube and many other Streaming / Broadcasting uses due to devices compatibility. EVC Baseline is expected ( or claims ) to be 20% better than AVC High Profile. So you should expect ~30% better than current AVC all while being royalty free.

The beauty of EVC baseline is that it is quite power efficient. Considering AVC already require the least computation in modern codec, EVC baseline offer 30% reduction in bitrate while requiring NO increase in decoding complexity and is actually 40 to 50% less in encoding complexity.

Note: None of these has been tested outside of its members as it does not provide a reference encoder due to "new" patents arrangement with this codec. So we have no way to verify those claims.

Edit: Will add the link to document with those claimed figures later.

Edit2: https://www.youtube.com/watch?v=0Itt0cOvgXU

[zerocrates]

My understanding is that YouTube uses at least some High Profile, and only Main for the lowest resolutions (though who knows how up to date that information is), along with of course VP9 and AV1 where they figure they can.

Anyway, if those claims bear out then it would seem like EVC has a possible future as an "AVC killer," though the computation complexity factors are complicated by the wide deployment of hardware acceleration for AVC encode/decode. Of course, hardware support could materialize for EVC as well, but it does rather complicate possible adoption for what we'll call a "sub-generational" improvement, in a world coming around right about now to AV1 hardware support.

Perhaps reduced computational demands could lead to a niche of usage in unaccelerated environments. I don't know if there's an example of significant uptake for an "in-between" step like this that one could point to as a possible model for success, though.

[drmpeg]

A quick check shows 360p = Baseline, 720p = Main, 1080p = High.

[ksec]

>My understanding is that YouTube uses at least some High Profile

Oh yes. I just double checked. I stand corrected. I was under the impression they only use high profile with MutiView Videos.