Flubly

Do the theatrical film reels also have the crushed blacks problem? Like in the Luke surrender scene in ROTJ where his clothes go below 0 IRE and you can’t see any detail or bring any back. That’s what is disappointing to me about watching any bluray re-grade. You can restore proper color but you can’t bring back the lost dynamic range.

If the film prints don’t have increased dynamic range, then yeah I wouldn’t really see the point of doing a scan of them.

It would be cool to have a scan of all the scenes that were just the original effects but with the elements digitally composited to reduce generation loss. Wide dynamic range and clean composites would definitely be nice to have for stuff like the Hoth battle.

RU.08 said:

Okay let me try to restate it, with Disney animation most characters (including Beast) have one animator that does that character. When they animate Disney cells they are told to follow the contours and movements of live-action reference shots they’re given, not to invent the animation themselves. Many shots even got reused in several Disney animations for example the ballroom dance is a shot reused from Sleeping Beauty. There’s nothing in that process that allows an animator to ever animate a silhouette as a black shadow. Also the animators were not the ones in creative control, it’s a lot easier to get your animator to draw the figure first and then alter its appearance as necessary for the final film. So what the animator for Beast intended and what the director intended for those early Beast scenes could be different.

Makes sense in that more bureaucratic structure, I can see that. It’d be interesting to see the relationship between the painting stage, since that’s a separate position from animation (looking at the credits), and the printing stage. Both of those stages involve decisions regarding tonal values, and it’d be cool to see what the formula that led to the final exposure was. Did the painters make stuff brighter than what they would want so that it would print properly? Maybe contrast ratio was more important than specific lightness so that it could scale to whatever was needed when printing. It’d be interesting to hear interviews with those people.

RU.08 said:

If you like the film I suggest arranging to see it on film some time if you can.

Would love to, though I don’t think I’ve ever seen Disney animation classics in any of the theaters around that do those type of screenings. I’ve seen some older Disney live action around, but not animation. I’m definitely keeping my eyes open.

RU.08 said:

Well it seems logical to me - you can’t animate a silhouette. You animate the character and then turn him into silhouette. Also, I’ve actually seen this movie projected in a cinema, and that’s exactly how I remember it.

And it’s 100% intentional. Don’t forget they showed the work in progress version in New York in 1991, 2 months ahead of the full release.

You can animate a silhouette, it’s just another abstraction of form. If you can animate a stick figure, a distorted long late-afternoon shadow, or amorphous shapes like the pink elephant sequence in Dumbo, you can animate a silhouette. There’s some stylized clean solid black silhouettes in the opening of Batman the animated series.

I don’t doubt that it was intentional though. Honestly, I can see how Disney would want headroom for exposure and have that flexibility. If a projectionist in the boonies shows the print too bright, Beast won’t be a gray blob.

RU.08 said:

Her hair often appears black on film, that’s just how it is.

I’m sure you’re correct. I think the preservation is a great endeavor and believe that original formats should always be available to the public. I’m not petitioning for this to be altered in any way.

Just commenting that I’m surprised that the original intended version (that I agree should be distributed as is) looks strange to me. It’s a weird fall-off, a separate aspect from how dark the film is. Something can be dark and not have a such a steep notch in a very specific area of the contrast curve. I don’t doubt that it’s intentional.

RU.08 said:

As I’ve always said, no home release of this film has ever looked right:

It is possibly the most modified of all the Disney animations on home release. Every home release going back to VHS and Laserdisc has made the beast visible way earlier than he is in the actual film. It changes the entire mood of these scenes where the beast is but a silhouette.

So was the decision to show him totally in silhouette made after the sequence had been colored? I don’t know a whole lot about the animation process, but I’ve seen cels of silhouettes that are painted totally dark before they were printed to film. Seems like kind of an odd decision to achieve a silhouette by fully rendering the figure on every level, including all the different colors between his cloak and fur, and then just print it darker so you can’t see any of that labor. That seems like an extremely wasteful and backwards way to achieve such an effect.

To my taste, I’d probably want a grade that’s a compromise. I feel like the home releases are too bright and this print is too dark (for me, not as a preservation). The line-work in Belle’s hair, at times, is crushed to black. The heavy shadow contrast makes this extremely fast fall-off where the skin tones are even and then the hair (depending on the scene) just disappears quickly into black, like the hair is a black hole sucking away the light around it. It’s a very steep gamma curve.

Again, I’m not disputing that it’s accurate, I was too young at the time to have seen it in theaters. I do think that the VHS retained an analogous mystery in that scene by making him pretty dark, though I do agree its still too bright.

I (roughly) regraded the bluray screencap to something that suits my own subjective tastes:

I tried to maintain some of the red in his cloak while still bringing all the luma information to near total darkness. I think the “almost can’t be seen” feeling has its own kind of mysterious atmosphere. When viewed in the movie as a whole, the shot goes by quick enough that’s it’s not really revealing. It just gives you faint impressions. Then you get those impressions illuminated, literally, when he steps into the light later.

Edit: Forgot that flat pre-grade still from the 35mm was available so I did a rough grade of the same principle to that one as well:

Ronster said:
It knows it is a PC, those options are available depending on what is connected to it. My TV is not on internet I only use it as a monitor.

I used the Nvidia control panel is much better than windows built in one.

Then you can install the firmware using a USB drive. You don’t need to be connected to the internet. Instructions on how to upgrade manually are included on that same link I posted above.

I understand that those options are grayed out for you depending on what is connected. I’m saying that I don’t think that is by design. I believe you are experiencing an error.

Again, what is the model of your TV? You know, the info listed here:

So, Ronster, what is the exact model of your Samsung HDR 4k tv? You linked to a MU6100 calibration page on pg. 3.
Is MU6100 the model listed on the back of your TV?

Grayed out calibration options with different sources is not a feature, that sounds like a bug to me.

The MU6100 has a firmware update as recent as this month:
https://www.samsung.com/us/support/owners/product/2017-uhd-smart-tv-mu6100

Are you up to date on that?

Ronster said:
This is a an association body. It will cost a lot of money to be an associate and what that grants you access to clients who work within that association.

It’s another way of shrugging off competition from smaller companies. Just because they have standards does not mean the actual association or even members of that association created those standards. They merely recognise a set standard and to be a member and put an SMPTE stamp on your company will also mean you have to meet that standard no doubt.

They are also part of drug company imaging devices big pharma contracts.

There is not anything wrong with an association body per se but like I said, this is a way of shutting out smaller competitive companys out of contracts because they can not afford the membership fees of being part of the association. And contracts are probably rarely given outside SMPTE membership in certain spheres of influence.

Not all bad but not all good either totally un-competitive though.

$13,500 dollars good client access or $20,000 for better client access per year.

I don’t think much of what you said here is true. Let’s say it is true, though. What does that have to do with what I asked?

If we all found out tomorrow that SMPTE is involved with human trafficking, it wouldn’t change the fact that the entire industry adheres to SMPTE standards and that using those standards achieves a relative consistency of image.

I asked about the science of video standards. If calibration of monitors to SMPTE standards isn’t a good method of attaining image accuracy and consistency, what is? I already said why I think your idea of having media tailored to TV tech wouldn’t work. A rebuttal to that would make more sense than a rumination on SMPTE’s business ethics.

Ronster said:

Caliberate a monitor to somone elses monitor is nonsense.

Sorry, forgot to mention this in my last post:

Calibrating a monitor to someone elses’ monitor IS nonsense. Calibration is done to an external set of standards, a studio doesn’t have a particular monitor they’re trying to match. They use tools like the Spyder 5.

A list of image sets for calibration and clear instructions can get you good results too, though. Again, not hitting absolute perfection is not an excuse for not trying at all.

Also, SMPTE was founded in 1916. Their standards influenced the manufacturing process of the film Star Wars was shot on. Their standards have also influenced how cameras are built today, how TVs are built today, how editing programs are coded, the specifications of DCPs projected in theaters, the specifications of TV broadcast since TV’s birth, etc.

If you think these standards are bunk, I’d like to know why and what the science would be behind a new standard.

A television can be calibrated such that no amount of color correction could make a source look correct. If I shifted the hue on a TV so that the green subpixels aren’t on, you can boost green in your source file all day and still not see an iota of it on the screen. At the end of the day, the monitor/tv is the one physically emitting light towards your eyes, not the source media.

Ronster said:

4k HDR samsung.

It is a cheaper model though.

the point is anyway if you were working in a studio or something of course everyone would have the same monitor.

If you are doing dual or more projection screens you make sure you have the same make and model to match.

Caliberate a monitor to somone elses monitor is nonsense. I even read a forum on why companies do not calibrate before sending out to customers and how depending on the light in the room all this bollocks. You also have specification to contend with now like HDR color depth resolution and so on.

You calibrate to match for a purpose but in an environment where everyone has a different make / model and specification or type OLEd / LED / LCD / Plasma / projector whatever utterly pointless it is the content that needs to be tailored.

you try calibrating an anti aircraft gun to be more like a machine gun never ever through calibration will you get the 2 things to match.

This is basically the truth.

In my field I deal with calibration a lot and I think you’re mistaken on how matching screens works.

Matching brand and model will help you get very close AFTER calibration. The same make and model can differ in chroma and luma values out of the factory. Even calibration won’t get you 100% similar. It’s like with asymptotes in analytic geometry, you are heading infinitely towards 0 on X or Y and never getting there fully.

Calibration is always approximate, it’s just that you can get that percentage of difference very very low depending on how complete and rigorous your methods are. Just because you won’t exactly match baseline, does not mean you shouldn’t try. Matching 98% is much better than matching 40%. I’ve had high-end professional monitors differ quite wildly despite matching make and model. I’ve seen monitors shift red over time without any settings having been changed and they need to be re-calibrated.

So, making media specific to screen tech is absurd. Two brands of OLED aren’t going to be the same out of the factory. Certain brands lean towards different color bias because of that company’s individual research and development. There are many factors that are involved besides just the base marketed tech of OLED, LCD, QLED, etc. There’s a lot of differing sub-design underneath that title. And like I implied before, there can just be uncontrollable variables in the manufacturing process that make the same models differ. You’d have to have media that’s specific to the serial number of your TV.

Screens are not calibrated in the factory, yes, partially because the lighting will different but mostly because automating that process would be difficult and expensive for very little return. Out of the millions of TVs sold, how many customers truly care about chroma/luma accuracy? The lighting would mostly just account for perceived luma shifts because of lighting reflecting off your screen or how closed/open your iris is because of the ambient light.

Any color differences would not be affected by differing lighting. You might see warm colors as stronger if you’re in a cooler lit room but that would shift to seem normal as you acclimate over 10 minutes or so.

The reason we have calibration is because that is the only available way to achieve consistency. The manufacturing process cannot guarantee matching images. It’s a tool decided upon because of no alternative methods available. There may be innovation in the future, but calibrated media is not part of that future. You cannot tailor home video to every unit because every unit is different.