Info: The Dark Knight - EE Reduction and Original Color Timing — Page 2

TheDarkestKnight said:

Bought the Special Edition a few hours ago. It looks like it’s same scan used for the Blu-Ray without the matte. The colors are still way off from the screener and the dynamic range is crushed compared to the BB disk. There also seem to be a couple of IMAX shots that aren’t on the disk. Mostly establishing shots (Wayne Enterprises, Fox getting off the copter, the Batsignal going off). At least that I didn’t notice in the five minutes I’ve had with it yet.

The shots for Rises are superb. If it turns out to be quick and easy, I may go ahead and add those shots to the movie for a true IMAX version.

Are you referring to the Trilogy Bonus Disc? I recently purchased this myself, but have not yet gotten around to the Bonus disc…

The establishing shots are available in their full IMAX height only on the 4:3 Full Screen DVD of ‘The Dark Knight.’ There are no 4:3 transfers of ‘The Dark Knight Rises.’

The one movie which would have benefited from a 4:3 aspect DVD is ‘Interstellar,’ because it would’ve looked stunning and would’ve preserved all the IMAX scenes in a format almost the same as itself, with negligible side cropping (the projector gate would crop the 1.44:1 scenes at the sides and vertically anyway). But sadly, since 2008, Full Screen DVD production has ceased.

Papai2013 said:

The variables of digital are entirely different and will never reproduce the beauty of an organic/analog medium. This is why most digitally graded movies look fake, even those that are shot on actual celluloid film. The digital grading never feels “natural” to my eyes. It feels “tacked on.” Digital footages in addition to digital grading has robbed the life from films. Both give an image a very synthetic, sterile and flat look. It does not feel lively or natural. because it is “not” natural.

You can get close, but that’s it.

While I totally understand your faible for film material, this statement luckily neither holds true technically nor mathematically, especially not in that generalisation which quite sound like the praise of vinyl to me whose alledged superiority to decent PCM doesn’t withstand any serious scientific comparison either.

A common misconception amongst most people is that something being digital means harm per se while in fact interpreting information digitally doesn’t make it better or worse - it’s just a different way of expression.

Another one is the false assumption that any analog signal would be perfect compared to the oh so evil digital ones. Digitalisation and reconstruction introduce errors and thus digital systems aren’t perfect considering the limited sample rate and word lengths by definition - but neither are analog systems to start with.

There are certain proven (but sadly mostly misunderstood) facts about how to digitalise analog signals and how to reconstruct them, the errors involved and the limitations depending on the desired noise floor or bandwidth.

Having written that, it doesn’t mean that even nowadays electronic cameras or digital based postproduction tools don’t have flaws but this is not due to the fact that digital data is processed as it could be theoretically infinitely precise depending on the effort one wants to take. And if an electronic camera looks “digital”, it’s because something within the analog parts like the CCD went wrong from the start. Digitalising stuff “only” limits two parameters: bandwidth and SNR according to Nyquist and Shannon.

And speaking about naturalism: the human DNA can be considered to be digital as well, so in fact something being digital is counterintuitively very natural.

When I was talking about why digital looks lifeless and why film feels more natural, I was speaking from a subjective perspective. I have seen movies shot and projected on film and I have seen movies done digitally. My eyes tell me that film looks more natural, pleasing and beautiful, with all it’s flaws. Digital almost always feels flat and sterile. In digital one can barely feel depth in a 2D shot, which comes much more effortlessly on filmed footage. I was just watching Satyajit Ray’s “The Coward” and Steven Spielberg’s “Jaws.” In each shot I could FEEL the depth, not just understand the placement of objects. I am speaking from what I have seen and felt. Your inference will most likely be the opposite.

By the way, your argument that our DNA is digital is absolutely wrong. DNA is made up of Adenine, Guanine, Thymine and Cytosine. These are organic elements and don’t exist in cyber space. You are welcome to prefer digital though. Just do not try and convince me.
Finally, no disrespect meant to you. 😃

While I’m all with you when it comes to your subjective perception about the different appearance, I only wanted to point out that these differences essentially don’t lie in the fact that one information is analog on film and the other one is digital but that the differences are caused by other things. The sensors in electronic cameras have quite a different characteristic compared to film chemics, I certainly give you that. But that has nothing to do with the digitalisation process itself.

And yes, the DNA consists of 4 bases which perfectly fits the definition of being digital as long as we’re talking about discreet values. Digital doesn’t necessarily mean binary, another very common misconception - it’s just the most prominent variant.

Other digital sets are:

English alphabet (26 values)
ISDN: (3 values)
Chinese: (several thousand)

Doesn’t matter, by definition any limited character set can be losslessly converted to any other base, hence it can be mathematically proven that digital data doesn’t have to be binary.

To summarise that: your observation in practise is correct but you’re drawing the wrong conclusion about what causes it as digitalising something while sticking to the rules of Nyquist/Shannon certainly doesn’t take any “life” out of anything.

That’s all.

I’ve been watching back my project in Premiere and I’m noticing some duplicated frames throughout. These all appear to be problems with the Blu-Ray rip. As I said, I’m a self-proclaimed newbie at this. Luckily for the IMAX bonus disk I was able to do a simple rewrap and not have to transcode anything.

If anyone knows the best way to go about ripping a Blu-Ray, I’ll be taking this over to the Technical Discussions and How To’s so that this thread stays on more project specific details. I just wanted to keep all of you updated as to how it’s getting along.

You are taking recourse to some theoretical concepts to make your point, while I am just speaking from a very practical experience; that of “seeing” with my eyes. So far, I have not really seen digitally shot or processed movies look better than celluloid shot films. Maybe, in the future, digital image acquisition will bridge the gap. But I have so far not seen anything promising, even on the most advanced Digital cinema cameras with 8K sensors and what not. Most experienced filmmakers also prefer the look of film over digital. I am just an audience and film enthusiast, that’s all.

TheDarkestKnight said:

I’ve been watching back my project in Premiere and I’m noticing some duplicated frames throughout. These all appear to be problems with the Blu-Ray rip. As I said, I’m a self-proclaimed newbie at this. Luckily for the IMAX bonus disk I was able to do a simple rewrap and not have to transcode anything.

If anyone knows the best way to go about ripping a Blu-Ray, I’ll be taking this over to the Technical Discussions and How To’s so that this thread stays on more project specific details. I just wanted to keep all of you updated as to how it’s getting along.

Premiere always does that to me. So, it may not be just a problem with your encode. It may also be as well. Glad to know your project is coming along. Also, don’t forget to insert the standalone IMAX shots that are only available on the TDK Full Screen DVD. PM me if you require any assistance regarding that.

Papai2013 said:

Now the problem is Blurg is gone, Spleen mostly won’t allow such stuff. So, even if you finish, how and where do you plan to share your work? It’s like how Rufus Scrimgeour says in Deathly Hallows Part 1 “These are dark times, there is no denying.”

Hmm, these are Dark Knights indeed. I don’t have a Spleen nor have ever used trackers (this being my first foray into fan edits), so my knowledge here is limited. I’ll be sure to research other options as the project continues. Would be a pity to finish this only to wait to find a way of distribution.

nafroe said:

Are you referring to the Trilogy Bonus Disc? I recently purchased this myself, but have not yet gotten around to the Bonus disc…

Yes, these are the shots included on the Special Edition bonus disk.

@TheDarkestKnight, I sent you a PM.

The screen cap that you posted in the first post was shot under Earls court in London. I used to work there years a go and I have been in that kitchen in the basement.

The Blu-ray looks the closest (1st picture) to what the kitchen used to look like. But Earls Court has now been knocked down so anyway Just in case you were looking for what looks like the real thing the first picture looks like that kitchen because it was not turquoise it had white / off white tiles.

Even hough I am not the addressee of your original comment, I would like to add that the intent of TheDarkestKnight’s project is to re-create as much as possible, the colour timing of the original 35mm/70mm prints. How the wall or floor of the actual building looks to the naked eye is not really significant in this case. But it was nice to know from you anyway. Thanks. 😃

Thanks for sharing Ronster!

I did a drive by of a lot of the filming locations while going through Chicago, and ultimately would like to visit all the filming locations for the movie. Sad to hear that Earls Court has been knocked down. I’d have to loved to visit that piece of cinema history.

I remember trying to transfer the colors from the Screener to the Blu Ray a year ago. I used a similar approach as DrDre does for his program. However, I ran in the same issues that you described in your initial post. The main problems are the crushed whites and blacks of the source. I cancled the project out of frustration.

Your approach of combining the lumina of the Blu Ray with the chroma of the Screener got me motivated again. The colors of the screenshot are really good. The only issue is the color smear. You can see it around the suit of the Joker.

I wrote a simple program to fix this issue. The parameters of the program are BluRay_EE.jpg (input) and Blu_Ray_w_Screener_Chroma.jpg (target). I compared the rgb value of each pixel. For each rgb value of the input I got a list of rgb values of the target. For most input rgb values, the size of the list is just one. However, even if the size is greate than one, the rgb values are pretty similar. I calculated the average red green and blue value based on the list of target values. This resulted in a map that maps each input rgb value to exact one rgb value. Based on this map I recolored the input image, which resulted in this image:

Link https://s32.postimg.org/4xeg76khv/mapped.png

As you can see the color smear is almost gone. There is also a slight color change, which in my subjective opinion doesn’t really hurt.

I am really pleased with this approach. The next step is to find a way to estimate color pairs that aren’t in the color map. Once this is accomplished, you are able to calculate a color map based on a few frames and can use it for other similar frames. By doing this you would be able to change the colors of the IMAX scenes.

Excellent! The smearing definitely is reduced, and I agree the color change is slight enough to get away with. I have no technical know-how in writing programs or scripts, so this is very much appreciated.

One note though, I’ve noticed most of the shots have different luminance from the screener, which is affecting the saturation of the colors. Look at the chap on the right’s face and the back wall, both are too dark and saturated in the merged version compared to the screener. Would there be any way to adjust this with the program given the crushed values of the Blu-Ray?

I have implemented a simple algorithm to tackle the dark colors of the blu ray. It goes like this: Calculate the average red green and blue value of both the input and the target. Subtract the average input values from the target values. Add the differences to the red green and blue values of each pixel in the input. I used the result image of my previous post as the input and the screener image from your initial post as the target.

The first result was this: https://s31.postimg.org/81lhm6g89/adjustedoriginal.png

If you compare this with the original screener you can see that the result is to bright. If i just add half the difference I get this result: https://s32.postimg.org/7s3ffif8l/adjusted.png

This is already really close to the screener. But I think it makes sense to introduce a parameter value in order to control the difference to add.

Looks good! It’s worth noting the screener doesn’t have a consistent luminance throughout frame though. The left always appears a little too dark and the right a little too bright, which may be why adding in the total difference looked too bright.

WhateverWorks said:

I have implemented a simple algorithm to tackle the dark colors of the blu ray. It goes like this: Calculate the average red green and blue value of both the input and the target. Subtract the average input values from the target values. Add the differences to the red green and blue values of each pixel in the input. I used the result image of my previous post as the input and the screener image from your initial post as the target.

The first result was this: https://s31.postimg.org/81lhm6g89/adjustedoriginal.png

If you compare this with the original screener you can see that the result is to bright. If i just add half the difference I get this result: https://s32.postimg.org/7s3ffif8l/adjusted.png

This is already really close to the screener. But I think it makes sense to introduce a parameter value in order to control the difference to add.

The black levels seem to be a bit off in the last result, being too bright, as they are for the screener.

Some of you noted, that if you use color matching tool to match the bluray to the screener, the skin tones don’t match, and there are undesired artifacts in the resulting frame.

I decided to have a look myself. First of all, when using the color matching tool, make sure that both images are cropped in the same way. Secondly, if artifacts appear try increasing the stabilization parameter. This usually also requires an increase for the number of color spaces needed to get a good color match. I did some experiments, and found that I get the best results, if I use 100 color spaces, and set the stabilization to 10.

Bluray:

Screener:

Bluray matched to screener:

Note, that the black crush of the bluray, and the off-set in the black levels of the screener results in some undesired noise in the dark areas. The solution is, to simply correct the black levels:

This way the screener color grading is still fully intact, and no fringing.

Here’s a comparison between the bluray and the regrade:

http://screenshotcomparison.com/comparison/179940

Excellent results, DrDre. The more that time passes, the weaker The Dark Knight BD looks. I didn’t know that the current master could be enhanced, as you have done.

It’s subtle, but a definite improvement.

In my opinion, the corrected black levels are too dark and makes the image look harsh. A middle ground between the screener and your corrected version might be the proper way.

Papai2013 said:

In my opinion, the corrected black levels are too dark and makes the image look harsh. A middle ground between the screener and your corrected version might be the proper way.

The problem is, that the crushed blacks for the bluray don’t give you much latitude. Here’s another version, where the black levels were set in GIMP, using the darkest point in the frame:

I think setting black to the darkest point is the best bet for most scenes. While it might appear a little harsh at times, it’s no doubt better than raising the blacks only to add back no information.

I am just speaking from my own observation - We don’t see dark areas like pitch or tar, we see slightly grayish blacks with our eyes. From most 35mm film frame stills I’ve seen, blacks are almost always slightly gray. This keeps the image soothing to the eyes as pitch black feels harsh. This is why I suggested that the black levels be somewhere between the screener and the corrected version. Also, the “darkest blacks” draw too much attention towards those portions. Just my feeling.

This is true. I want to say in film I see blacks on the waveform from around 5-25 on a 10-bit scale. If it is possible the blacks will be raised to around there depending on what the darkest point is. If there is a considerable amount of noise though, it will need to be kept closer to 0.