logo Sign In

1997 Star Wars Special Edition 35mm Project (a WIP) — Page 3

Author
Time
 (Edited)

We think most of the Special Edition was a photochemical restoration, and that only scenes with optical compositing were remastered digitally. It still needs to be scanned at 4K even if it is a 2K digital film-out, as the film-out is to negative anyway and the pixels will not align perfectly to the camera either. The 2K film-out resolution (2048x1556) includes the soundtrack area, once that is taken out it’s more like 1828x1556 or something (I forget the exact size), the soundtrack area is typically left blank and added separately. There’s also blank-space above and below the picture in the samples poita posted, however those are not digital scenes, so it’ll be interesting to see how the digital scenes look by comparison. Hopefully poita can post one for us soon!

poita said:

No, I do not think they were, I think it is a common misconception but we will soon find out. (I could be wrong, but I am pretty sure they weren’t)

The new CGI sequences would have been done in 2K full aperture, but the rest of the film was done traditionally, and even the new wipes were done optically, so apart from the new effects, it should resolve well above 2K.

That’s a good point about the wipes, I’m interested to see about the rest of the re-compositing as well, although I imagine they were also done optically in the main.

EDIT…

Actually they may have done more work digitally than just the CGI, as I understand it these “burn marks” are a tear in the camera negative:


Here’s the same two frames from the 1997 SE (“TB release” of a Scandinavian DVB recording):


They’ve also removed the motion blur in that shot. I could be wrong, it could be the tear was to the o-neg and not the camera negative, but it doesn’t explain why they didn’t just make a replacement section from the master positive (or the camera negative) for the original 1977 release. I’m interested to see how the scan of this shot looks!

[ Scanning stuff since 2015 ]

Author
Time

It wasn’t in the camera negative as the story goes, but again, it will be interesting to find out.
It will be clear from the scan if this was digitally repaired or not.

I think we will learn quite a lot from the SE, even about the OUT.

Donations welcome: paypal.me/poit
bitcoin:13QDjXjt7w7BFiQc4Q7wpRGPtYKYchnm8x
Help get The Original Trilogy preserved!

Author
Time

poita said:

It wasn’t in the camera negative as the story goes, but again, it will be interesting to find out.
It will be clear from the scan if this was digitally repaired or not.

I think we will learn quite a lot from the SE, even about the OUT.

I never liked the special edition but holy shit I am excited

Author
Time
 (Edited)

poita said:

3500 lines aren’t enough? 😉

The final scans will be in ACES colourspace.

Nice; hopefully we can get a P3 gamut version produced.

Haha I was referring to how the apparent resolution of release prints is rather low, which I guess shouldn’t be too surprising given that 35mm negative struggles to resolve as well as a Arri 3K camera. Even moreso the case for older theatrical prints; I’d imagine newer print making processes introduce less degradation vs older methods.

For those who’d like to go down the rabbit hole, the director of photography for The Last Jedi made a very interesting video lecture about how different different cameras and formats influence the apparent resolution of an image… basically arguing that the focus on pixel count is misguided; his comparison showing that a 11k scanned 70MM Imax negative struggles to resolve as clear as a 4k master from an Arri Alexa 65 is rather sobering.

http://yedlin.net/ResDemo/

poita said:
it should resolve well above 2K.

I dunno about that; if a 6k scanned 35mm negative has trouble resolving as well as the Arri 3k camera, and take prob 20% off that figure for generational loss (in reality I think it’d be a lot more), you’re looking at about 2.4k.

Here’s some further reading on the matter that’s rather interesting;: http://www.reduser.net/forum/showthread.php?59036-Resolution-of-35mm-film-and-70mm-IMAX-film

Author
Time
 (Edited)

Actually I do know about that 😃 I’ve scanned many prints that easily resolve above 3K, and negatives that far exceed 4K.

The SE prints may well have been finished at 2K, we won’t know until we look at the scans, but if they weren’t we should get good resolution from them.

There are many details in the IB Tech prints from 1977 that exceed the 2160 lines of UHD, there is no way to scale them down to 4K BD and not lose some of that detail.

The area around the Tantive door is an easy place to see this, at 2K the doorframe is a uniform grey, with a 6K scan, you can see it is ribbed, rather like a vacuum cleaner tube.

Some prints are terrible and struggle to have much in the way of resolution, others have far in excess of 2K, and many, many negatives we have scanned exceed 4K and need to be scanned at 6K or even 10K to keep the resolution evident in the negatives.

We have studied all sorts of theories on this, but it comes down to actual practice, 35mm prints and negs often have details that are lost if you dip below 4K.

Donations welcome: paypal.me/poit
bitcoin:13QDjXjt7w7BFiQc4Q7wpRGPtYKYchnm8x
Help get The Original Trilogy preserved!

Author
Time
 (Edited)

This is what the doorframe around the Tantive door looks like just before it blows


(thanks to MV for the image)

If you scan at 3K, it is just a blur, at 4K you can see the pattern and at 6K the pattern is clear.

This is from a 1977 print, it will be interesting to see if the SE print has this level of detail or not.

The negative would have far more detail again.

Donations welcome: paypal.me/poit
bitcoin:13QDjXjt7w7BFiQc4Q7wpRGPtYKYchnm8x
Help get The Original Trilogy preserved!

Author
Time
 (Edited)

I’m not trying to cast doubt on your expertise here, and I’ll agree that it’s important to scan 35mm at 4k+ for full detail retrieval due to anti-aliasing and the like, but if DP Steve Yedlin can’t get fresh 6k scanned 35mm filmstock to resolve more than a 3k Alexa in a controlled environment, there is no way a release print suffering several generations of degradation is going to outdo that (especially when created via older processes).

Of course these are generalities and the resolving capability of film isn’t a very linear process (it isn’t with digital cameras either).

This is also a good read as well: http://www.motionfx.gr/files/35mm_resolution_english.pdf

EDIT Yes I can see what you mean about the ribbing. That being said, it’s important to remember that just because you need to scan at high resolutions for details to become visible doesn’t mean those details “exist” at that resolution/couldn’t be adequately resolved at a lower resolution (the reason why probably has to do with Nyquist–Shannon).

Author
Time

Yes but the question I have about that door-frame is: is it still a blur when projected in a cinema? I think it would be, even if it’s visible on a scan.

[ Scanning stuff since 2015 ]

Author
Time

The pdf I linked to goes in to enough depth such that it might help you to answer your question; tests were conducted in a theater, using (for the time anyway) top of the line film stocks, properly set up projection, lenses, etc. Even then, the apparent resolution of a release print to the audience member was stunningly low.

Author
Time

No, you can see the ribbing quite clearly when projected, unless the projector had very sloppy registration.

Anyway, I disagree with Steve, he is a capable guy, but he is a DP, and has only really been once since 2002, his experience with actual film is very little.
I’m not a DP, my day job is film restoration, I work with negs that far exceed 4K resolution on a daily basis, we can and do measure it. I certainly wouldn’t argue cameras or lenses with Steve, that is very much his domain, but when it comes to his knowledge of film stocks and their attributes, his experience isn’t extensive.
The ITU did tests back in 2000 that gave a vertical resolution of around 2700 lines for a scope print, and considerably higher for a negative.

Plenty of films, especially release prints look little different at 2K, but others have striking differences, text that is not readable below 3K, but is clear at 5K, the details like those in the doorframe, there is actually no way to keep that detail at 2K, even if you were rendering it directly from a computer model, there just isn’t enough pixels, and again, those are in the release print, not the neg. The neg would have superior resolution again.

With this particular print of the SE, we will do what we usually do, scan it at 6K, 4K and 2K and take a look at the scans, and see how much detail the print actually has. If it was in fact mastered at 2K, I won’t expect to see any difference between a 6K scan and 4K scan at all.
We would still scan it at 4K if it was 2K mastered though, otherwise we would absolutely lose details just due to the nature of scanning.

Projection is up for debate, there is no easy way to measure it, though attempt have been made, but so much depends on the equipment, screen, light and seating position.
It doesn’t really interest me though, I am an archivist, and if it is there on the print, I need it there in the scan, regardless of if that detail can be seen or not by a particular audience member in a particular cinema.

Preservation has different goals to commercial releases, for a commercial release, 2K would probably satisfy just about any audience.

Donations welcome: paypal.me/poit
bitcoin:13QDjXjt7w7BFiQc4Q7wpRGPtYKYchnm8x
Help get The Original Trilogy preserved!

Author
Time
 (Edited)

Fair enough. Quick note though: the ITU paper said that “The highest resolution that the expert assessors could discern in the highest performing movie theater was about 875 [lines]” with the qualification that “a wide range of resolution values will be obtained, depending on the stocks, laboratory, type of printer, and so forth.”

Also: it’s important to clarify that just because you need to scan at 4k+ resolution to extract detail from a print doesn’t imply that said detail can only be resolved at that resolution in a digital master; as I mentioned previously, this seems showcase the necessity of oversampling vs the resolving ability of the film.

Anyway, I digress. Scan looks lovely.

Author
Time
 (Edited)

poita said:

This is what the doorframe around the Tantive door looks like just before it blows


(thanks to MV for the image)

If you scan at 3K, it is just a blur, at 4K you can see the pattern and at 6K the pattern is clear.

This is from a 1977 print, it will be interesting to see if the SE print has this level of detail or not.

The negative would have far more detail again.

This is actually not the case. I have a 2k copy (2048x858) of a 35mm scan, which shows the pattern fairly clearly.

Author
Time

DrDre said:

This is actually not the case. I have a 2k copy (2048x858) of a 35mm scan, which shows the pattern fairly clearly.

Was the scan performed at that resolution or was it done at a higher res then laster down-sampled to 2k?

Author
Time
 (Edited)

Synnöve said:

DrDre said:

This is actually not the case. I have a 2k copy (2048x858) of a 35mm scan, which shows the pattern fairly clearly.

Was the scan performed at that resolution or was it done at a higher res then laster down-sampled to 2k?

It was scanned at 4K. The detail at 4K resolution is actually not all that much better than at 2K. Mike Verta used overlays of multiple prints to obtain the clear edges you see in the image poita posted. So, the edge is more grainy on an actual print, but even at 2K the pattern is clearly visible.

Here’s a blowup of the 2K downscale:

Author
Time

Sorry, but I think that demonstrates that the pattern is absent at 2K.

Donations welcome: paypal.me/poit
bitcoin:13QDjXjt7w7BFiQc4Q7wpRGPtYKYchnm8x
Help get The Original Trilogy preserved!

Author
Time

My point is that we regularly have details when scanning at 4K that are only 1 pixel high.

Donations welcome: paypal.me/poit
bitcoin:13QDjXjt7w7BFiQc4Q7wpRGPtYKYchnm8x
Help get The Original Trilogy preserved!

Author
Time

poita said:

Sorry, but I think that demonstrates that the pattern is absent at 2K.

Well here’s what it looks like at 4K:

The reality is, that it’s not a blur at 2K like on the bluray, and a pattern is clearly visible, and it’s not so greatly defined at 4K either.

Author
Time
 (Edited)

poita said:

My point is that we regularly have details when scanning at 4K that are only 1 pixel high.

Sure, but I felt is was worth noting, that the bluray with it’s digital noise reduction isn’t really a good representation of the detail, that can be resolved at 2K. I’ve used that example myself in discussions, but when examining the frames at 2K, I was surprised to see, that especially in motion, the pattern around the edge is also visible at 2K resolution. Here I’ve averaged a number of frames, which should be a better representation how it looks in motion:

Now obviously it’s pushing the limits of the detail 2K can resolve, but there’s no denying the pattern is visible.

Author
Time

Your scan looks a lot worse than mine at 4K, you can’t really see the pattern at 4K on your scan.
That appears to be the very beat-up IB print that was scanned of a DFT scanity, which is a terrible scanner for prints, and struggles a bit with smearing due to its line scanner, unlike the area scanner of the Arri or the Director.

I don’t want to come across as a dick, its just that all we do is scan and restore film, that is literally our job, and we regularly get detail that is one pixel high on a 4K scan from a print, and at 5K on negs.

The 10K scanner is really only that high for oversampling reasons, but up to 5K there is detail on a good negative that cannot be resolved properly if you resize to a lower resolution.
We often get referred to reports or tests done that ‘prove’ that film cannot resolve this level of detail, yet we see it resolved in our daily work. The tests are often very specific, or done by people with not a lot of film experience, or by companies that make sensors in digital cameras, and test specific film stocks at specific labs and specific lens combinations, which is all great, but it leads to conclusions that are often off-base, or are correct for a specific setup, and they then generalise out.

Yes, generally films resolving power on any given shoot isn’t pixel perfect at 4K, often there is around 3K of detail (full aperture) , it is a lot more complex than that due to grain structures and channel issues, processing and light, but generally if you look at a single frame, often 2K would be enough to represent what you can see on the film as a final delivery (if you scanned at 4K or 6K and down sampled properly).

Not particularly rare though, especially on original negatives, are single frame images that cannot be accurately represented at 2K, or even at UHD resolution, especially cinemascope or full frame images that end up at less than 1500 lines on a 4K UHD Blu-ray. This is not theory, this happens in the restoration process that we do.

When you get something like that door-frame, and the alternating lines of the pattern are about 1 pixel high when scanning at 3500 lines of resolution, no processing will keep that pattern at Blu-ray resolutions.

I know I’m banging on, but it drives me nuts sometimes to be told, repeatedly over the years, that film is not capable of doing something we see it do regularly in our work.

Anyway, my gut feel is that for the SE, if it didn’t come from the neg, which I’m not sure a lot of it did, I think we will not need more than 2K as the final delivery to make it indistinguishable from the film, but to do that we would need to scan at 4K anyway, even if it was mastered at 2K.
So to answer the oringal question, we would want to scan it at least at 4K either way.

I’m really hoping we will get better resolution than the IB scans we have of the OUT, we will know soon.

Donations welcome: paypal.me/poit
bitcoin:13QDjXjt7w7BFiQc4Q7wpRGPtYKYchnm8x
Help get The Original Trilogy preserved!

Author
Time
 (Edited)

Very interesting! Thanks for the clear explanation! I’m not going to argue with your expertise. 😃 Didn’t know the Scanity is actually not that great of a scanner.

Author
Time

Leia

Donations welcome: paypal.me/poit
bitcoin:13QDjXjt7w7BFiQc4Q7wpRGPtYKYchnm8x
Help get The Original Trilogy preserved!

Author
Time

Donations welcome: paypal.me/poit
bitcoin:13QDjXjt7w7BFiQc4Q7wpRGPtYKYchnm8x
Help get The Original Trilogy preserved!

Author
Time

This is the scanning setup window, for anyone that is interested.

Donations welcome: paypal.me/poit
bitcoin:13QDjXjt7w7BFiQc4Q7wpRGPtYKYchnm8x
Help get The Original Trilogy preserved!

Author
Time

Very cool - and cheers for the scanning set-up window pic too - always wondered just what controls, options and layout etc there was 😃

A little patience goes a long way on this old-school Rebel base. If you are having issues finding what you are looking for, these will be of some help…

Welcome to the OriginalTrilogy.com | Introduce yourself in here | Useful info within : About : Help : Site Rules : Fan Project Rules : Announcements
How do I do this?’ on the OriginalTrilogy.com; some info & answers + FAQs - includes info on how to search for projects and threads on the OT•com

A Project Index for Star Wars Preservations (Harmy’s Despecialized & 4K77/80/83 etc) : A Project Index for Star Wars Fan Edits (adywan & Hal 9000 etc)

… and take your time to look around this site before posting - to get a feel for this place. Don’t just lazily make yet another thread asking for projects.