Star Wars: The Complete Saga (1977-2005)

[octagon]

Originally Posted by Norbie

We're the "mattes" visible on the original releases when watching them in the cinema circa 77-83?

I very clearly remember one set of fairly noticeable mattes during a shuttle launch in Empire. Other than that *maybe* one or two around tie fighters immediately following the Death Star escape but I can't be sure I'm not misremembering that from a home video viewing.

[Norbie]

Originally Posted by dvdmike

In Empire on a release print, nope not really

If they were not seen in the original experience why would people want to see them now?

This sort of issue also come up in older films with "special effects" such as wires being used. In the original releases the makers knew that because of the lack of detail at the time the wires would not be seen when watching at the cinema.

So going back to Star Wars, why would people want to see such things as "matte" lines when on the original release they weren't seen?

[Thomas Guycott]

You're confusing "matte lines," an inherent compositing artifact of optical printing, with "garbage mattes," which are the transparent boxes visible around models.

It's the garbage mattes which are claimed shouldn't technically be visible under properly calibrated conditions (although I honestly have a grain of doubt to that claim).

[BillieCassin]

Originally Posted by Thomas Guycott

It's the garbage mattes which are claimed shouldn't technically be visible under properly calibrated conditions (although I honestly have a grain of doubt to that claim).

I have yet to find a truly thorough explanation of it, lots of assumptions and theories but no solid, definitive answer - but, I can't deny visually it's true.

While we know that native 35mm film has far greater resolution than 1080p HD video, and that the 35mm film is projected far larger than we are watching at home, somehow, defying that knowledge, things like that are much more obvious when watching on the inferior, 1080p. It makes no sense, but it's true.

There is something about the nature of digital images that somehow captures those details and highlights them. You can see a similar effect taking still digital pictures yourself. I've taken a picture of an object on a table before, a table that to my human eye (with obviously higher resolution viewing than a digital camera) was clear of debris, but the digital camera picks up specs of dust that I can't see there (and I have 20/20 eyesight).

I can't explain it, and as I've said, I've yet to find a satisfactory explanation of the effect - but it definitely exists.

[dvdmike]

Originally Posted by BillieCassin

I have yet to find a truly thorough explanation of it, lots of assumptions and theories but no solid, definitive answer - but, I can't deny visually it's true.

While we know that native 35mm film has far greater resolution than 1080p HD video, and that the 35mm film is projected far larger than we are watching at home, somehow, defying that knowledge, things like that are much more obvious when watching on the inferior, 1080p. It makes no sense, but it's true.

There is something about the nature of digital images that somehow captures those details and highlights them. You can see a similar effect taking still digital pictures yourself. I've taken a picture of an object on a table before, a table that to my human eye (with obviously higher resolution viewing than a digital camera) was clear of debris, but the digital camera picks up specs of dust that I can't see there (and I have 20/20 eyesight).

I can't explain it, and as I've said, I've yet to find a satisfactory explanation of the effect - but it definitely exists.

The grain from a duped and duped again release print covered a ton of issues, also consider the bulbs used.
The best restorations have perfect ideals applied to them, and it was almost never the case in real life.
There is a push for the "Grindhouse" style of look, with all the issues left and not restored, an honest look from a small part of the Internet.
As this is how it was seen theatrically, but they want to make it look pristine now and that is where matched grain comes in.
This is an issue in most films as movies with special effects shot on 35mm and then popped in will have header grain patterns as most likely the film would have been printed multiple times.

[Cook]

Originally Posted by BillieCassin

I have yet to find a truly thorough explanation of it, lots of assumptions and theories but no solid, definitive answer - but, I can't deny visually it's true.

While we know that native 35mm film has far greater resolution than 1080p HD video, and that the 35mm film is projected far larger than we are watching at home, somehow, defying that knowledge, things like that are much more obvious when watching on the inferior, 1080p. It makes no sense, but it's true.

There is something about the nature of digital images that somehow captures those details and highlights them. You can see a similar effect taking still digital pictures yourself. I've taken a picture of an object on a table before, a table that to my human eye (with obviously higher resolution viewing than a digital camera) was clear of debris, but the digital camera picks up specs of dust that I can't see there (and I have 20/20 eyesight).

I can't explain it, and as I've said, I've yet to find a satisfactory explanation of the effect - but it definitely exists.

Brightness and contrast of home tvs makes them more visible I think.

[MechaGodzilla]

Maybe an option for Disney/Lucasfilm would be to offer two versions via seamless branching, one that "fixed" all the little issues like the garbage mattes and the other with all the warts remaining. But that'd probably be too much trouble for relatively minor differences.

[dvdmike]

Originally Posted by MechaGodzilla

Maybe an option for Disney/Lucasfilm would be to offer two versions via seamless branching, one that "fixed" all the little issues like the garbage mattes and the other with all the warts remaining. But that'd probably be too much trouble for relatively minor differences.

That would kill the bitrate, a decent restoration would sort any issues anyway and it would be a release print anyway

[Ernest Rister]

Originally Posted by Thomas Guycott

You're confusing "matte lines," an inherent compositing artifact of optical printing, with "garbage mattes," which are the transparent boxes visible around models.

It's the garbage mattes which are claimed shouldn't technically be visible under properly calibrated conditions (although I honestly have a grain of doubt to that claim).

No, that's true. If your contrast and brightness settings were too high, you'd see the garbage mattes even on the "Deifinitive" laserdisc release box set. A lot of people just don't know how to calibrate their TVs, especially back in the 80's and 90's.

[Geoff D]

Originally Posted by BillieCassin

I have yet to find a truly thorough explanation of it, lots of assumptions and theories but no solid, definitive answer - but, I can't deny visually it's true.

While we know that native 35mm film has far greater resolution than 1080p HD video, and that the 35mm film is projected far larger than we are watching at home, somehow, defying that knowledge, things like that are much more obvious when watching on the inferior, 1080p. It makes no sense, but it's true.

Consumer video is not film. I'm not trying to sound facetious, just stating a fact. The response of things like gamma, colour and sharpness are very different between the two; video generally has a very linear response to such things, whereas film is more non-linear. Its properties can still be quantified from stock to stock, otherwise people would never have any idea what their images would look like until they'd shot & printed them, but it behaves very differently to home video and it's that translation which has been found wanting over the years IMO. (Thomas, FYI, Dennis Muren has long lamented the overly bright video transfers of Star Wars, see this link, about 3/4 of the way down the page: http://www.gizmodo.com.au/2012/01/gi...-dennis-muren/)

And also bear in mind that although we were watching film prints back in the day, the 'native' resolution is eaten into by the generational loss (a 35mm print is a 4th gen element) which obfuscates the finest details like monofilament wires and takes the edge off the grain, whereas most video transfers were done from IPs, never mind the modern penchant for going straight off of the camera negative and scanning at 4K or higher (GWTW and Wizard of Oz were scanned at 8K, which was beyond overkill for the smaller Academy aperture IMO, although the final masters were 4K).

In other words, we really ARE seeing things look sharper and brighter than they would have ever been in the theater originally (well, at least where 35mm is concerned because large format is a different story). There has been a call for the BT.1886 video standard to become more widely used precisely because it responds in a more non-linear fashion, and some people say it's in common use now for home video mastering but I'm skeptical as to how widespread it actually is.

And as much as I witter on about having a calibrated display blah blah blah (who, mesa? ), it won't do a damned thing to hide garbage mattes and other artefacts if they're baked in to the source, i.e. if the master itself is too bright with too low a gamma but with the contrast cranked up to bring back the blacks.

[Geoff D]

(It's also worth pointing out that the telecine transfers of yesteryear often resulted in a noisy image and not an authentically grainy one, leading to a coarsely exaggerated sense of texture which frightened the content providers so much in the earlier days of HD media and even before that, as with the dreadful DNR on the '93 transfers of the OT).

[dvdmike]

Originally Posted by Geoff D

(It's also worth pointing out that the telecine transfers of yesteryear often resulted in a noisy image and not an authentically grainy one, leading to a coarsely exaggerated sense of texture which frightened the content providers so much in the earlier days of HD media and even before that, as with the dreadful DNR on the '93 transfers of the OT).

And they had a huge pink push usually

[HeavyHitter]

Originally Posted by Geoff D

Consumer video is not film. I'm not trying to sound facetious, just stating a fact. The response of things like gamma, colour and sharpness are very different between the two; video generally has a very linear response to such things, whereas film is more non-linear. Its properties can still be quantified from stock to stock, otherwise people would never have any idea what their images would look like until they'd shot & printed them, but it behaves very differently to home video and it's that translation which has been found wanting over the years IMO. (Thomas, FYI, Dennis Muren has long lamented the overly bright video transfers of Star Wars, see this link, about 3/4 of the way down the page: http://www.gizmodo.com.au/2012/01/gi...-dennis-muren/)

And also bear in mind that although we were watching film prints back in the day, the 'native' resolution is eaten into by the generational loss (a 35mm print is a 4th gen element) which obfuscates the finest details like monofilament wires and takes the edge off the grain, whereas most video transfers were done from IPs, never mind the modern penchant for going straight off of the camera negative and scanning at 4K or higher (GWTW and Wizard of Oz were scanned at 8K, which was beyond overkill for the smaller Academy aperture IMO, although the final masters were 4K).

In other words, we really ARE seeing things look sharper and brighter than they would have ever been in the theater originally (well, at least where 35mm is concerned because large format is a different story). There has been a call for the BT.1886 video standard to become more widely used precisely because it responds in a more non-linear fashion, and some people say it's in common use now for home video mastering but I'm skeptical as to how widespread it actually is.

And as much as I witter on about having a calibrated display blah blah blah (who, mesa? ), it won't do a damned thing to hide garbage mattes and other artefacts if they're baked in to the source, i.e. if the master itself is too bright with too low a gamma but with the contrast cranked up to bring back the blacks.

Good post.

Regarding BT.1886, the biggest issue is hardly anything mastered on Blu has been calibrated to such - as a result, watching such especially on catalogs makes the image look too contrasty and a bit artificial from my experience with it. I have heard 4K Blu might use BT.1886, but I still wonder how that's going to look even mastered to such - I am not sure.

Regarding theater projection and prints, I recall a study that was done several years back when analog projectors were still the standard in U.S. theaters and given the generational film-print loss, lack of projector calibration, lack of ideal focus, etc., etc. most movies were being projected at approximately around 720p equivalent resolution at best.

[dvdmike]

Originally Posted by HeavyHitter

Good post.

Regarding BT.1886, the biggest issue is hardly anything mastered on Blu has been calibrated to such - as a result, watching such especially on catalogs makes the image look too contrasty and a bit artificial from my experience with it. I have heard 4K Blu might use BT.1886, but I still wonder how that's going to look even mastered to such - I am not sure.

Regarding theater projection and prints, I recall a study that was done several years back when analog projectors were still the standard in U.S. theaters and given the generational film-print loss, lack of projector calibration, lack of ideal focus, etc., etc. most movies were being projected at approximately around 720p equivalent resolution at best.

720p! Maybe at your la-di-da upmarket cinemas!
I have seen way worse in my time

[moviebuff75]

I worked at a cinema from 1993-2001. You are right...a lot of the time, the prints were very soft, some almost blurry.

[Geoff D]

Of course, the only thing about talking up what older 35mm movies looked like on the big screen versus home video (especially now we're moving into the 4K era) is that it begs the question: if we are seeing them in greater quality than what was originally envisioned, then surely a case can be made for digitally ironing out the seams - but how far do you go? Do you remove the old-school analogue artefacts but leave the continuity errors, or do you do both as with Blade Runner Final Cut?

[HeavyHitter]

Originally Posted by Geoff D

Of course, the only thing about talking up what older 35mm movies looked like on the big screen versus home video (especially now we're moving into the 4K era) is that it begs the question: if we are seeing them in greater quality than what was originally envisioned, then surely a case can be made for digitally ironing out the seams - but how far do you go? Do you remove the old-school analogue artefacts but leave the continuity errors, or do you do both as with Blade Runner Final Cut?

Honestly, I easily could live with either approach at the end of the day. If they release the original movies with the original color timing and a quality restoration going back to the ON with a 4K scan (which no notable jackassy processing), I will be very, very happy whether or not they digitally remove the mattes, etc. You will not read a single negative remark either way from me.

Ideally, you do what Ridley Scott and Warner did with Blade Runner: release all of the versions with their respective characteristics. But, I am don't believe that will happen in the case as Warner is very open to working with directors on releases (still amazing we got the Superman II Donner Cut) and Ridley Scott took such an active role.

I could be wrong, but I don't think Lucas would be too involved if Disney releases the originals since he hates the originals with a deep rooted passion ---and I don't think Disney would put together something Warner did for Blade Runner which was a rare fan 'dream' set.

[Cook]

What is BT.1886?

[HeavyHitter]

Originally Posted by Cook

What is BT.1886?

http://www.spectracal.com/Documents/BT.1886.pdf

[Cook]

Originally Posted by HeavyHitter

http://www.spectracal.com/Documents/BT.1886.pdf

Thank you.