4K Movies Releases

[Geoff D]

Originally Posted by D2Girls

I've a question about UHD blu rays.
How exactly can they fit movies on a single disc?
Isn't UHD essentially 4x the information/data? So... will UHD blu ray utilize a new form of disc with 4 50gb layers or something? I just don't see how a 2 or a 2 and a half hour movie can fit on a single disc @ 250Mb/s bitrate.

Huh? They won't be studio master quality, no matter what some people might say.

Most Blu-rays nowadays are delivering superb video quality with, say, 25Mb/s (average) AVC encodes, so if we take your x4 metric (which AFAIK isn't accurate for the reasons dubious states, it's nearer 2.5x) then that's 100Mb/s which fits comfortably inside the maximum 127.9Mb/s bitrate outlined in the UHD Blu spec for two-zoned 66/100GB discs.

Add to that a newer compression codec which is up to 64% more efficient than AVC at compressing 4K material and it's easier to appreciate where the bitrate savings will come from so 100GB capacity discs should be fine. (The problem is more the availability of large-scale 100GB production lines in general, though I should hope that the BDA have sorted that out by now!)

[spectre08]

Originally Posted by D2Girls

how substantial is substantial..? it'd have to be 4x better at compression would it not?

the HEVC is about twice as efficient as current codecs. This means that a 4K video which has 4x as many pixels only requires twice the data to achieve the same quality.

What the quote you're looking at is saying is that they can also apply it to 1080p or lower sources and maintain the same quality with half the bitrate, or maintain the same bitrate with twice the quality.

[Penton-Man]

Originally Posted by D2Girls

how substantial is substantial..? it'd have to be 4x better at compression would it not?

As other members have already indicated, the answer is no. People are free to try it on some content for themselves….
https://hevc.hhi.fraunhofer.de/svn/s....2+SCM-3.0rc1/

The most cutting-edge work involving the implementation of the codec known as HEVC (not revealed in that otherwise fine wikepedia article contributed by Spike ) is that tests on UHD HDR content have shown unexpected new video artifacts such as noise and banding which might become noticeable to viewers if the encoder isn’t specifically adapted.

[Penton-Man]

Originally Posted by spectre08

...

Well where the heck have you been? You missed contributing to our whole discussion pages back about Spectre…. https://forum.blu-ray.com/showthread...e#post11481822

[jono3000]

So after watching The Hobbit Battle of the Five Armies Extended Edition I'm left wondering if we will see HFR versions on UHDBD?

Also reading that Toy Story is 20 years old this week, does anyone know what resolution it was originally rendered at? And when they revisited it to create the 3D version did they keep a 4K version handy for a future release?

[Penton-Man]

^ anyone care to help the man out? Geez guys, the first question isn’t that difficult

[Penton-Man]

Originally Posted by Penton-Man

dvdmike is slowly coming around but still in need of HDR maintenance dosing.

ergo, a new ad by Technicolor/Hollywood in a trade magazine…

[spectre08]

Originally Posted by Penton-Man

Well where the heck have you been? You missed contributing to our whole discussion pages back about Spectre…. https://forum.blu-ray.com/showthread...e#post11481822

Mods put me on another vacation, I don't even remember why.

[d514]

Originally Posted by jono3000

Also reading that Toy Story is 20 years old this week, does anyone know what resolution it was originally rendered at? And when they revisited it to create the 3D version did they keep a 4K version handy for a future release?

According to several sources (Google):
"Toy Story 1 was 24 frames per second at 1536 × 922 pixel resolution, 48 bits per pixel."

[Geoff D]

It's 1.5K then, so much for keeping a 4K version handy...

[singhcr]

Originally Posted by Geoff D

It's 1.5K then, so much for keeping a 4K version handy...

That is surprising that such a low output resolution could hold up to 35mm projection. How good were the film out processes and film recorders back then?

[Geoff D]

It's not that low mate, not when you consider that 2K has basically been the DI standard all this time. It holds up just fine on the average cinema screen, and always bear in mind that this stuff isn't being rendered out with sub-sampled 8-bit colour (e.g. Blu) but with full colour resolution at 10, 12, 16 bits and no compression, I know of a VFX supervisor who says that uncompressed 1K looks better to him than Blu does at 1.9K.

[singhcr]

Originally Posted by Geoff D

It's not that low mate, not when you consider that 2K has basically been the DI standard all this time. It holds up just fine on the average cinema screen, and always bear in mind that this stuff isn't being rendered out with sub-sampled 8-bit colour (e.g. Blu) but with full colour resolution at 10, 12, 16 bits and no compression, I know of a VFX supervisor who says that uncompressed 1K looks better to him than Blu does at 1.9K.

I wasn't aware 10+ bit color was used for so long. So the film recorder could handle such color depths even though displays can't show it? Even now, Dolby is just coming out with 12 bit displays. So even in the 1990s people were rendering at such high color depths? I understand that film has a very wide color gamut compared to digital systems, but I wasn't aware that recorders could take advantage of that during the film out process. And by "full color" you mean 4:4:4?

[Geoff D]

Don't measure what they did behind the scenes 20 years ago by what consumer displays are only just starting to do, generally there's been very little correlation between the two.

As soon as people started moving into the digital domain for processing film (late '80s) they realised that they needed a wide dynamic range (high bit depth) and logarithmic gamma to accurately capture what film is capable of, and I think I'm right in saying that Snow White was restored in 4K back in 1993!

I don't have exact timetables of what changed to what and when because I honestly don't know the specifics, but 10-bit has been around for many years already, that's why just when we're starting to finally get 10-bit software for the home the industry's ceiling for dynamic range has already shifted to 16-bit.

[Geoff D]

In other words: there was no point moving into the digital realm if they couldn't actually replicate the basics of what film looked like (grain, response curve etc) when it was laser recorded out at the other end, so a lot of work went into replicating such things using the tech of the time, even 25 years ago. And that then evolved into the end-to-end DI pipelines that we have today.

[singhcr]

Originally Posted by Geoff D

In other words: there was no point moving into the digital realm if they couldn't actually replicate the basics of what film looked like (grain, response curve etc) when it was laser recorded out at the other end, so a lot of work went into replicating such things using the tech of the time, even 25 years ago. And that then evolved into the end-to-end DI pipelines that we have today.

That makes sense, although you wonder why that same mentality wasn't applied to digital projectors in movie theaters vs 35mm ones. Thank you very much for the info.

This is why I have much more respect for the movie industry than the music industry from a technical standpoint. Digital recording in the 1980s wasn't even close to what analog tape could do fidelity wise, and CDs weren't close to LPs. Even now, when high quality PCM and DSD options are available, they simply aren't used. We get 96/24 recordings and 44.1/16 CDs and even LPs use the CD quality files. Sad.

[Penton-Man]

Originally Posted by Geoff D

the end-to-end DI pipelines that we have today.

Aside….

Since we’re fast approaching the Christmas and New Year’s holidays, there is an anomaly forthcoming in the motion picture business to the traditional DI process in which power windows are typically used, be it for the HDR grade or just an SDR grade.

In the case of Tarantino’s The Hateful Eight, the 4K DI (Baselight 2) utilized no power windows. Think of it being a straight (with no tweaking) digital conversion of the film print.

[Penton-Man]

Wishing a Happy Thanksgiving to all readers (members and lurkers).

Signing off for a couple of days as the Chicago and East Coast relatives are soon to arrive at the airport and I must do 3 separate runs to LAX before we sit down tomorrow for the eats. Thank goodness at least some of our family are local.

Later.

[Geoff D]

Originally Posted by Penton-Man

Aside….

Since we’re fast approaching the Christmas and New Year’s holidays, there is an anomaly forthcoming in the motion picture business to the traditional DI process in which power windows are typically used, be it for the HDR grade or just an SDR grade.

In the case of Tarantino’s The Hateful Eight, the 4K DI (Baselight 2) utilized no power windows. Think of it being a straight (with no tweaking) digital conversion of the film print.

And why not? Just because the filmmakers have all these tools there doesn't mean they *have* to use them. (I presume no such trickery went on with the DSMs of Nolan's movies either?) Digital finishes would look more 'organic' if people didn't keep relying on such things and actually tried to capture the image as best they can first time, I had to laugh when reading Richardson's comments about having to adjust to that way of shooting again!

[Geoff D]

Originally Posted by Penton-Man

Wishing a Happy Thanksgiving to all readers (members and lurkers).

Signing off for a couple of days as the Chicago and East Coast relatives are soon to arrive at the airport and I must do 3 separate runs to LAX before we sit down tomorrow for the eats. Thank goodness at least some of our family are local.

Later.

You Americans and your Thanksgiving, I think I threw up a little in my mouth when I read a yahoo article saying that we should try it in the UK, it's bad enough that the full horror of Black Friday has been unleashed on the rest of the Western world.

Have a good one nonetheless.