logo Sign In

1997 Star Wars Special Edition 35mm Project (a WIP) — Page 19

Author
Time

The home video release appears to have a different pixel ratio as well–a bit wider stretched than the print. However it looks more like blooming from the digital transfer process than a rotoscope difference.

It doesn’t hurt to offer help, but it always hurts to disregard those that do.

Author
Time

vexedmedia said:

The home video release appears to have a different pixel ratio as well–a bit wider stretched than the print. However it looks more like blooming from the digital transfer process than a rotoscope difference.

yeah, that is what i thought as well.

Author
Time
 (Edited)

Yeah, I think I can spot differences pretty decently, but I don’t see it here. Sorry to say.

Author
Time

DrDre said:

I think you should also consider DNR was applied to the home video versions, that probably affected some of these shots.

The Jap LD and the PAL Broadcast are two entirely separate telecines though, and I don’t see how DNR can be causing the differences we’re seeing.

It’s all in the wrist baby (poita’s scan):

PAL broadcast:

Comparison:

Another comparison:

[ Scanning stuff since 2015 ]

Author
Time
 (Edited)

I don’t think the differences I’m seeing are simply due to dnr and blooming. It doesn’t explain why the lightsaber gets completely blown out but the other highlights in the frame do not. For example, the tip/point of the lightsaber in this image doesn’t change, but the entire glow does:

The lightsaber effect is an optical composite, and it looks to me like they made two versions. They made one version that they printed on the theatrical prints. Then later they did some telecine transfers (probably for promotion of the special edition) and weren’t satisfied with how the telecine looked and did a re-composite of the effect for telecine where they gave a greater intensity to the effect, resulting in the blade becoming much brighter. I’m quite certain this is what has happened with the lightsaber rotoscoping - it’s the same effect, just re-composited with a different intensity.

This isn’t that unusual, there are dozens of examples where otherwise “unmodified” films had certain parts recomposed specifically for telecine - for example to make text and credits fit 4:3 better, or to move/resize subtitles. And in fact the subtitles are another thing that was changed between the 1997 SE theatrical prints and the the telecine versions:

So it’s actually not at all unusual they would change this between release formats, but it was previously unknown.

[ Scanning stuff since 2015 ]

Author
Time

I don’t buy it, to me it looks more like some kind of clipping effect from the telecine. Some limiation from the tape format.

Author
Time

Why would this same clipping effect be on two entirely seperate telecines and apply only to the lightsaber effect and not the highlights in the rest of the frame?

[ Scanning stuff since 2015 ]

Author
Time

It does. In that comparison gif, look at the highlights on 3POs arm, they exhibit the same sort of hard edges.

It’s also possible that highlights introduced through optical compositing could be at a much higher level than highlights introduced though the original photography.

Author
Time

It looks like the same thing the GOUT has. The blade just looks wider and more intensified.

Army of Darkness: The Medieval Deadit | The Terminator - Color Regrade | The Wrong Trousers - Audio Preservation
SONIC RACES THROUGH THE GREEN FIELDS.
THE SUN RACES THROUGH A BLUE SKY FILLED WITH WHITE CLOUDS.
THE WAYS OF HIS HEART ARE MUCH LIKE THE SUN. SONIC RUNS AND RESTS; THE SUN RISES AND SETS.
DON’T GIVE UP ON THE SUN. DON’T MAKE THE SUN LAUGH AT YOU.

Author
Time
 (Edited)

Chewielewis said:

It does. In that comparison gif, look at the highlights on 3POs arm, they exhibit the same sort of hard edges.

Not really, they exhibit some minor blooming that you might expect from an old telecine transfer. It looks nothing like the intensified lightsaber - and notice that it’s not just the highlights that have intensified on the lightsabler - the entire blue glow has widened, and this shouldn’t happen if it was the same on the telecine prints that it is on the theatrical prints. The usual practise of the day, as far as I’m aware, was to produce prints for telecine separately - you didn’t transfer theatrical prints if you didn’t have to. So telecine prints will have differences to them, however in this instance it appears they re-composited the effect so the blue glow would be more visible on the videotape.

Also, why wouldn’t this be the case with the other shots, like this one:

Poita’s scan:

“TB release”:

Or this scene:

No extreme glowing happening there.

[ Scanning stuff since 2015 ]

Author
Time
 (Edited)

Ok so i took the 35mm scan, and ran it though photoshop with a simple levels adjustment, cliping the highlights Imgur

And I get this result

Imgur

Frame comepare here http://www.framecompare.com/image-compare/screenshotcomparison/K7P7NNNX

Obviously this is not scientific, it doesnt match exactly (due to the grain I think), but it shows how a simple levels transformation could be responsible for the TB saber looking that way. Its likely the super bright, high saturation lightsaber would have been simply clipped off through the telecine process.

As to why the Falcon lightsaber in the wide shot doesn’t do the same thing? You can compare the two sabers. Im Bens hut the glow is much brighter, almost as bright as the core where as the falcon saber the glow is much less bright and the saber has a harder edge. So where that levels transformation clips the soft glow in the hut, it doesnt affect the falcon saber in the same way.

The luke closeup is also different as the glow is almost invisible and the saber edges are very hard. Star wars is very inconsistent, there were probably several different ways they made the saber effects.

Here is the falcon shot with the exact same levels transformation applied. The glow is still present but its not hard clipped like the Bens hut shot.

Imgur

I find this a lot more likely than the shots being completely rerendered for the TV release.

Author
Time

Chewielewis said:

Obviously this is not scientific, it doesnt match exactly (due to the grain I think), but it shows how a simple levels transformation could be responsible for the TB saber looking that way. Its likely the super bright, high saturation lightsaber would have been simply clipped off through the telecine process.

Right, but you’ve brightened the entire background to get there. The point is this brightness difference is limited only to the lightsaber optical element - thus it’s from a 2nd composite, one made for telecine. There are two versions of the Jabba shot, and two versions of the Greedo shot (both with and without subtitles) so why shouldn’t there be two versions of this shot also?

I find this a lot more likely than the shots being completely rerendered for the TV release.

It wouldn’t have taken much effort, and would have been done early on probably before the theatrical o-neg was completed. Keep in mind that directors would often shoot replacement shots for television, so why wouldn’t they do it for composites too? IIRC there’s a shot in “Scream 2” that the MPAA wanted cut from the movie (the scene where Phil is stabbed in the head), but the producers or Wes told them they only had the one version of the shot so they let them keep it and got an R-rating anyway. But, really they had thought ahead and shot a TV safe replacement for the shot.

[ Scanning stuff since 2015 ]

Author
Time

Based on the scans of release prints we’ve seen, Ben’s hut has always been a little dark. I can imagine a telecine operator watching this scene back in 97-98 and thinking: “Ah! Too underexposed. Let’s crank up the brightness a little…” And that’s what is causing this anomaly.

What’s the internal temperature of a TaunTaun? Luke warm.

Author
Time

RU.08 said:
Right, but you’ve brightened the entire background to get there. The point is this brightness difference is limited only to the lightsaber optical element - thus it’s from a 2nd composite, one made for telecine. There are two versions of the Jabba shot, and two versions of the Greedo shot (both with and without subtitles) so why shouldn’t there be two versions of this shot also?

Like I say, this is just a demonstration. I don’t have the exact transformation right and probably cant without the right materials. But it shows that the lightsaber look can be attained by clipping the highlights. Consider how different the transformations are between going from o-neg to low contrast interpositive to internegative to release print to scan to colormatch vs going from o-neg to low contrast interpositive to telecine to SD Tape. Different paths could affect the highlights differently.

And there really is no point in recomposing this shot, as you can see from the colormatch the only real difference is the lightsaber looking WORSE. Why would they do that? Why would they, finish the film, go back and change the intensity on the blades so they are blown out, create new film outs, telecine them, edit them into the video master, all at enormous expense for very little return.

I’m not saying that they CANT do that, its certainly possible, I just think very very very unlikely.

And subtitle shots are a different deal as the subtitles are not present on the interpositive which would have been scanned for this version. Subtitles are either cut or burn into internegatives so there are always clean versions of those shots avaliable. Clean versions aren’t made for the TV version they are made for international film prints.

Author
Time
 (Edited)

Chewielewis said:

But it shows that the lightsaber look can be attained by clipping the highlights.

Right, but that’s not good enough. You need to show the lightsaber AND the rest of the scene can be attained with some LUT. I’m certain it can’t because this was scanned from a different composite than the one used in the 1997 theatrical prints, one designed for telecine.

And there really is no point in recomposing this shot, as you can see from the colormatch the only real difference is the lightsaber looking WORSE.

Yes there is. I had to use DrDre’s tool to colour match - that luxury didn’t exist in 1997. I appreciate you haven’t seen how the shot looks with no correction applied at all so I’ll show you. See this screenshot which shows the dark background areas directly against the black frame borders with no colour correction. Compared to the TB broadcast, and the Jap Laserdisc, the entire reel is a lot brighter. The telecine operator would have needed to dial the brightness down, not up. And again I stress that the TB broadcast and the Jap Laserdisc are two separate transfers - they both look the same, and the new scan looks different.

Chewielewis said:

And subtitle shots are a different deal as the subtitles are not present on the interpositive which would have been scanned for this version. Subtitles are either cut or burn into internegatives so there are always clean versions of those shots avaliable. Clean versions aren’t made for the TV version they are made for international film prints.

My understanding is that they strike prints specifically for telecine, they don’t use the IP. Also, these subtitles were not burned in - they are optical composites, so would absolutely be present in the interpositive.

Why would they, finish the film, go back and change the intensity on the blades so they are blown out, create new film outs, telecine them, edit them into the video master, all at enormous expense for very little return.

Well 1. it wouldn’t have been at “huge expense” and 2. because they wanted the scene to look good on video. It’s the debut lightsaber scene, that’s why it looks the best in the entire film! They put the most effort into that one scene, and they wanted it to look as good on video as it did in the theatres.

[ Scanning stuff since 2015 ]

Author
Time

RU.08, I’m not saying you’re wrong because I have no proof to say otherwise, but you cannot compare your scream 2 nor subtitle comparison with a special effect shot.

Your Scream 2 example falls flat because it was a simple retake for a Made for TV version that was likely spliced in on video.

The subtitle comparison falls flat because it is standard to not have burned-in subtitles for international releases or home video. Multiple IPs are made for this reason as well as safety copies.

I’m with everyone else: this looks like DNR and / or artificial brightening of the highlights. Or maybe the lightsaber was masked off and brightened in the telecine process, but I seriously doubt it’s a recomposite done solely for home video.

What’s the internal temperature of a TaunTaun? Luke warm.

Author
Time
 (Edited)

I’d also like to point out that the GOUT version of that shot has the same clipped lightsaber effect. Sure, in 1997 they COULD HAVE gone back to the shot on the computer, rendered off another version, film out, telecine, edit into tape. But in 1977? no chance.

Imgur

GOUT top 2004 bottom
Imgur

Author
Time

Mavimao said:

The subtitle comparison falls flat because it is standard to not have burned-in subtitles for international releases or home video. Multiple IPs are made for this reason as well as safety copies.

Right, but these subtitles were not burned-in they’re an optical composite that would be present in the IP.

Chewielewis said:

I’d also like to point out that the GOUT version of that shot has the same clipped lightsaber effect. Sure, in 1997 they COULD HAVE gone back to the shot on the computer, rendered off another version, film out, telecine, edit into tape. But in 1977? no chance.

Who said anything about digital? This shot was an optical composite in 1977, and an optical composite when the photochemical restoration was performed for the 1997 SE. They re-composited the shot optically for the 1997 SE like they did most optical composites (the Speeder through Mos Eisley, the optical wipes, etc). What I’ve presented above is two composites, both made in 1997, with one giving a greater intensity to the lightsaber optical.

[ Scanning stuff since 2015 ]

Author
Time

Subtitles can be “burned in” with a special laser. This is how subtitled releases were made.

The fades were done optically for the SE but all of the composite shots were done digitally.

What’s the internal temperature of a TaunTaun? Luke warm.

Author
Time
 (Edited)

Right, but the subtitles in Star Wars are not burned in, they’re a composite. That’s why they have a shadow. Burned-in subtitles look like this. It’s where we get the “burned-in” phrase from - they’re literally burned into the image after the film has been developed.

Mavimao said:

The fades were done optically for the SE but all of the composite shots were done digitally.

I disagree, it was a photochemical restoration, while the optical wipes may have been digital, they re-comp’d other stuff optically.

[ Scanning stuff since 2015 ]

Author
Time
 (Edited)

RU.08 said:
Right, but these subtitles were not burned-in they’re an optical composite that would be present in the IP.

If the subtitles are optically composited they would be done further down the line than the interpositive struck from the o-neg. They would be inserted into an internegative or a dupe positive, there would be several versions for different languages.

Chewielewis said:

I’d also like to point out that the GOUT version of that shot has the same clipped lightsaber effect. Sure, in 1997 they COULD HAVE gone back to the shot on the computer, rendered off another version, film out, telecine, edit into tape. But in 1977? no chance.

Who said anything about digital? This shot was an optical composite in 1977, and an optical composite when the photochemical restoration was performed for the 1997 SE. They re-composited the shot optically for the 1997 SE like they did most optical composites (the Speeder through Mos Eisley, the optical wipes, etc). What I’ve presented above is two composites, both made in 1997, with one giving a greater intensity to the lightsaber optical.

Pretty sure this one was a digital re-composite, as the hilt end of the blade is reshaped to fit with the prop, I don’t think they did that optically, only the dissolves and wipes were done optically.

What i’ve presented is how the lightsaber look can be caused by clipping highlights, how both the 1997 scan and the 1993 gout have the same clipping effect on the lightsaber despite comming from composites made 20 years apart.

You say that the only things in the shot that different are the saber blade, look again at 3PO. Here is a comparison between the TB and your color matched scan, see in the waveform, how the only things as bright as the saber are the highlights on 3PO and the light on the far left (cropped in the TB), and see how they are smeared, just like the bottom right corner of the saber blade.

Note the big chunk of dirt both on the TB and the 35mm. If these were entirely separately printed pieces of film that dirt would not be on both. Edit: It’s not on the 4K77 grindhouse edition, so its likely to be dirt on the SE interpositive.

Author
Time
 (Edited)

RU.08 said:

I disagree, it was a photochemical restoration, while the optical wipes may have been digital, they re-comp’d other stuff optically.

It’s not a matter of debate. They were digitally composited. Go to 3:38 in this video:

https://youtu.be/VhLUdy3RDpM

I don’t want to come off as dickish, but you obviously do not know what you’re talking about.

The 97 restoration was partially photochemical and partially digital. All of the non-special effect shots were restored photochemically and the wipes were done optically. The elements for all the fx shots were all rescanned and recomposited or enhanced with CGI and then outputted to 35mm film. The OCN is now a mix of the original 77 negative (non effects shots) and 97 negative (wipes and sfx shots).

What’s the internal temperature of a TaunTaun? Luke warm.

Author
Time

Chewielewis said:

If the subtitles are optically composited they would be done further down the line than the interpositive struck from the o-neg. They would be inserted into an internegative or a dupe positive, there would be several versions for different languages.

No they wouldn’t. They’d be in the IP.

Pretty sure this one was a digital re-composite, as the hilt end of the blade is reshaped to fit with the prop, I don’t think they did that optically, only the dissolves and wipes were done optically.

No way is it digital. The digital shots have a different framing, and compressed contrast, this one is consistent with the majority of the reel. It’s optical.

What i’ve presented is how the lightsaber look can be caused by clipping highlights, how both the 1997 scan and the 1993 gout have the same clipping effect on the lightsaber despite comming from composites made 20 years apart.

Yes it’s the same optical element in both shots, just at a higher exposure for telecine.

You say that the only things in the shot that different are the saber blade, look again at 3PO. Here is a comparison between the TB and your color matched scan, see in the waveform, how the only things as bright as the saber are the highlights on 3PO and the light on the far left (cropped in the TB), and see how they are smeared, just like the bottom right corner of the saber blade.

No I didn’t say it’s the only thing different, what I said was blooming/smearing does not explain the difference as it only applies to the lightsaber effect. The blooming on the background anomalies does not match that on the saber blade.

Note the big chunk of dirt both on the TB and the 35mm. If these were entirely separately printed pieces of film that dirt would not be on both.

Right, the dirt is something I noted to poita, in fact the very first thing I noticed and I gave him quite a few examples. It’s not on the print it’s in the print. Not just black dirt, but white dirt also (i.e. dirt on the positives). And yes it would still be in both because it’s a composite shot and the dirt is on the film used to make the composite.

[ Scanning stuff since 2015 ]

Author
Time
 (Edited)

Right, how prints were made (to add to the subtitle discussion)

“The process generally goes as follows: The A, B, and C (if necessary) rolls, are all printed onto an interpositive, which has lower contrast than ordinary release-print stock (contrast builds up in the internegative and release print stages).

This interpositive is then printed onto one or more internegatives, which is/are then used (along with a separate soundtrack negative, containing optical tracks and any digital tracks/timecode that might be used for that particular film) to print theatrical prints. If foreign distribution is expected, the C roll (containing titles) is sometimes printed separately on its own interpositive, and then both interpositives are printed onto the internegative(s). This allows for different versions of a film’s titles, which can be made in different languages for foreign prints; subtitles for foreign prints can also be added by splicing them into
the `title’ interpositive.”

Source: http://stason.org/TULARC/movies/production/6-6-What-is-an-interpositive-An-internegative-Film-La.html

What’s the internal temperature of a TaunTaun? Luke warm.