A recent thread about "Hi Res" audio got me to thinking..."What is the reference standard for video?"
For audio, in general, it's the soundstage acoustics of a live performance. Even completely synthesized music sounds different in a club versus the home. But can be closely approximated.
However what is the reference standard for video? Until we get Star Trek level holosuites, we're still far from the immersion of a live stage performance. So it seems that the reference would be a theatrical presentation. But what theatre? IMAX, Cinerama or whatever equivalent is today, or a multiplex?
Also, unlike audio, where it's possible to get a true copy of the original digital master, consumer video is still a highly compressed, color graded for home version of the original (which we'll almost certainly never see offered).
So What happens in the future when theater screen size (which is shrinking) and acoustics are matched by consumer level equipment? Does that become the standard?
+ Reply to Thread
Results 1 to 4 of 4
"Is it live, or is it Memorex" went the campaign from that old tape company. And, unfortunately for them, nearly everyone could tell the difference.
But a reference serves dual purposes: as a "standard" to compare against, but also as a set of stress tests.
Because these signals have to also maintain their robustness.
Audio has been able to achieve the equivalent of "you are there" facsimile because its data requirements are much more modest. Remember, it has to fully (and beyond!) accommodate the capability of our own human systems.
For audio, (using, for expediency of demonstration right now, LPCM) this means a sample rate of 96kHz or greater (192 is even better when dealing with transient interaural differences), a dynamic range of 144dB (which should be covered by 24bits, though more is preferred during capture and especially during processing), and enough channels to cover the full spherical 3D spatiality (at least 2 - if binaural/hrtf, but likely 4 -ambisonic- or more).
For video, because of the sheer magnitude of data, we still haven't completely passed that mark of realism.
But it would likely be something like:
Beyond existing HDR - like 10000nit displays, using 32bit (float?)/color primary.
ACES-type color spectrum (which is beyond rec2020 and P3), and using 6 color primaries (RGBYCM)
Beyond autostereoscopic - more like holographic or integral imaging
8k or greater resolution
Again, uncompressed LPCM.
This would probably be in the range of terabits/second.
Thus, we are still doing best-effort compromises in this area.
I have no idea what actual question is being asked here. Should also point out that many of the movies you watch in theaters are actually processed in 4:4:4 or 4:2:2 1080p, and then upscaled to 4k, as a to reduce the cost of the process chain.
Last edited by KarMa; 21st Oct 2019 at 09:39. Reason: no instead of not
Don't believe audio is that pat. You are still getting a pre-mixed, mastered version, much like the video is composited & graded.
For copies of consumer-distributed commercial material, the IP of film & video investment is so much more intensive, it is very likely that corporations will not venture beyond UHD-BD for quite some time, especially as the format has NOT been the resounding success they hoped it would be (neither was 3D, but both investments were expecting too much too soon, so they have hastened their own downfall with bad expectations).
There will be better "straight from camera" reference material, but that will NOT be compatible or easily playable on most consumer gear.