yotsuya

I need to check some of the stars against the TESB starfield (to make sure I didn’t include any ghost stars) and I have almost finished stabilizing the flyover. I think my planets need some tweaking. Something just doesn’t seem right. I have a better version that I rendered, but I want to do some of the tweaks before I post it. One I’m confident the flyover is stabilized, I need to attack it for color and detail. On further consideration, having a separate layer for the laser bolts and the engine glow makes more sense. I need to desaturate the star destroyer hull and leave the others bright. I can also usetransparency to better composite the elements. Getting very close to a final version.

I finally have it and have watched it. Very good. Mostly a nice color palet, but in many scenes there is too much yellow. Where the Blu-ray is way to red, this seems to be hit or miss with some scenes overcorrected to a yellow bias (when Ben find’s Luke and when they reach the Alderaan system and find the Death Star) followed by others that still carry a lot of red (inside Ben’s Hut and climbing out of the smuggling compartments). Aside from a yellow cast to a lot of the scenes, it is fantastic. I love the detail in the restored scenes. Definite the go-to version for high quality image. For color, almost but not quite. It lacks the consistency of the old home video telecines and the bad, but consistent, colors of the Blu-ray. Some scenes are stellar but some are way too yellow. Han’s shirt should be a creamy off-white, not a dirty yellow.

Actually, there is more detail in the shadows giving the end image more detail. He said the middle was supposed to be the correct white balance.

Funny, when you look at the three corrected photos, the lower one looks better and it didn’t have the right color balance to begin with. I can’t wait for your tool to be finished. I have a ton of photos that need this kind of help.

Not bad considering the source. Definitely a lot of ghosting on the stars. One of the biggest problems with VHS and LD sources.

I’ve been refining the starfield more. I found that the old pan & scan PAL LD had the intro squished. Althor1138 very graciously gave me a copy. So far, it has the most complete starfield, with the exception of some cut off on the left. I was able to find enough pieces in TESB that I stitched something together to cover that area. The problem is, while these old sources do show the position of the stars, the stars themselves are distorted and unusable. Even the TESB starfields are inconsistent. So, since I was able to pull the brightest stars directly off the SSE bonus 81 crawl, I have gone ahead and recreated the rest of the starfield manually. I could pull most of it from TESB, but the stars really don’t line up (probably a slightly different angle of shooting). Since none of them are usable directly, recreating the starfield seems like the best option.

I also did some preliminary examination of the various flyover sources and while the quality of the SSE 77 flyover and the blu-ray flyover are excellent, there are noticeable differences, especially in the engine flares (and in the case of the blu-ray - the engine cowlings themselves). The blu-ray would also require fixing the laser blasts to match the 77 and 81 versions. So that leaves the SSE bonus 81 sequence as the only one that will work and look right. I still will need to create a steller version of that final frame (in the above video it is adequate). But the rest of the frames are going to require a lot of manual work to get it right. It’s only 410 frames. But if I do it that way, I can use some better tools and I won’t have to rotoscope the elements in Vegas. I also won’t need a separate blue glow layer. The goal will be to get as close to the blu-ray quality as possible. I might even need to use some sections of the blu-ray star destroyer to restore details that are obscured. We’ll see how much effort I want to put into it. But first, stabilizing.

I’ve tried 3D several times. It makes to more obvious that what you are watching is fake. If you try to focus on another part of the image, the edges stand out in a very fake way. I saw TFA in 3D and all the models looked small and fake. Anything that was too close or too far from the camera distracted from the experience. All the midrange stuff looked great. But it wasn’t just TFA. All the trailers had the same issues. All the times I’ve viewed 3D, I’ve noticed pretty much the same thing. I’m not likely to ever pay to see a movie in 3D again or get any sort of 3D home entertainment system. It is a great novelty, but other than than offering a depth dimension to things, it just makes everything seem unreal - the opposite of what a movie experience should be. I know lots of people love it, but I hate it. I know how to put it… watching TFA in 3D was like watching it in GOUT quality at home. Watchable, but constantly distracted by artifacts that a quality viewing wouldn’t have.

Well, I can’t speak for the American market, but the BBC wanted copies destroyed or returned when the contract was up. I suspect that some US stations might have kept the contract going for years so they may have kept a copy that they might air periodically. But one a contract was up and not renewed, the tapes went back or got wiped. With the BBC’s bad archiving, lots of things got lost because of this. The occasional stray turns up that didn’t get destroyed or returned so the policy is well documented in hopes more strays turn up.

So my guess is that tapes sent out for broadcast were wiped afterward.

yoda-sama said:

yotsuya said:

4k is more suited to 3D and gaming than motion pictures.

Which is why they didn’t include 3D in the 4K Blu-ray specs…

3D’s biggest problem is a lack of detail at depth. It needs 4k to look good. Right now it is pretty crappy.

TV’s Frink said:

yotsuya said:

I think 4k home media is largely a waste. For most titles, DVD is still good.

I read those two things and decided not to read on.

Most movies weren’t made with such detail in mind. Most don’t need that level of detail to tell the story. New movies look great in 1080p, but 720p for HD broadcast for fine for most. For those who sit far across room from the screen and are more interested in the story, 480p is plenty. I still watch DVD’s And I don’t bother with Blu-ray from Netflix. Now a movie like Interstellar… that is a Blu-ray movie. When you watch for the visuals then more pixels becomes important. Vital. I’d love to see Lawrence of Arabia in a 4k or higher projection. But for the latest comedy or a typical silver screen classic… DVD works.

I have extracted the Title fade and the crawl from TESB and ANH blu-rays respctively. I manually removed the excess black background from the Title after using some video processes to isolate the color portion and forcing everything else to black. It would have been much simpler if the title wasn’t black text with a yellow border. There was just no easy way to do it. I matched the fade out to the LD source. I used almost the same process to get the crawl, but I had to add a layer to bring out the area where it fades. I didn’t need to edit the individual frames since there was nothing black that I needed to keep so it is a simple lighten layer. I need to still tweak the starfield, but I have the alignment between the starfield a moons. This rendering is just a draft of where it is now. I’m checking the alignment, star brightness, star placement, and pan for issues and overt differences from the original. The flyover at the end is just a place holder and is from TN1’s SSE bonus 81 Crawl sequence.

https://vimeo.com/165022953
password: OT

Mockingjay PT 2 had such a track. I think all home video should default to a matrixed stereo track and full encoded surround should be an alternate.

I think 4k displays are a great thing. I think 4k (or higher) preservations of of titles is awesome. I think 4k home media is largely a waste. For most titles, DVD is still good. For high detail films, Blu-ray is pretty much all you need. Going higher produces diminishing returns and it won’t help most titles. Most titles aren’t 4k ready. I believe that 4k maxes out what 35mm film can capture. Higher is good for things shot on larger film stock, but those aren’t as common. I don’t think the average person is going to be able to tell the difference between 1080p and 4k. I think the extra display pixels will do more for the content than increasing the resolution of the content.

I know there are many people out there who will have a setup that enables them to really tell the difference, but they are a minority (though I’m not so sure that this site doesn’t lean the other way). 4k is more suited to 3D and gaming than motion pictures.

towne32 said:

So until we find out what he has, if anything, might as well keep improving our skills by working on a remaster.

Well we’ve already seen a list of at least some of the scans he has done, so I don’t think the “if anything” is really valid. But you’re correct that you’re proceeding how you best can, given that you don’t have those scans or know when it would be possible.

What he has scanned doesn’t really tell us what shape it is in. What we need to do a proper restoration is a scan that contains a proper starfield and the final frame. If he has something like that and someone with the right skills restores it (stabilizing and removing damage) then what I’m trying to accomplish can be scrapped, but if it isn’t complete than my project might serve as a guide to fixing it. I really hope he has something. That is what I really want.

It looks like these are a gamma correction from being awesome. Looks like they may have been scanned a bit dark.

We are all anxious to see what Poita has been up to, but I have heard of no time estimates so far. Besides, using similar techniques as Harmy uses would create a version that is appropriate to tack on to the Despecialized Edition. So until we find out what he has, if anything, might as well keep improving our skills by working on a remaster.

You could do that if you have access to those sources and know how to align the images. I still don’t know if that would be high enough quality or if you could recover enough detail. I’m more comfortable remastering it from other sources. That is something I can do and I have access to everything I need. At the moment the flyover is the issue, but there are many ways to address that. I need it to match in quality to the GOUT, the Despecialized Edition, and the 35mm film scan (I hear we are soon to see a second version of it).

I made a new topic for the 81 crawl since it has come up in other topics as well as here - http://originaltrilogy.com/topic/Remastering-the-1981-Episode-IV-TitleCrawlFlyover/id/49668

The 1981 Title/Crawl/Flyover is the version seen by everyone on broadcast and home video between 1983 and 1996. It exists out there in many forms, but not one of them is up to even the standards of the 2006 GOUT. Some come close, but the 2006 GOUT was transferred from the master video tape (with the 1981 crawl replaced with the original 1977 crawl) to DVD, making it much higher quality than the 400 lines of video stored on any of the LD versions. Since this is the version I watched so much at home (such as the SciFi Channel premier in 1992), it is the version I want if I am going to watch the original cut of Star Wars. To get it at an acceptable quality, it will have to be remastered. Now, if we were just making it from scratch, you couldn’t say that. The goal here is to take existing elements and recombine them back into an accurate version of the 1981 sequence. The other option is a scan of the print, but the only one currently available suffers from a variety of issues, not the least of which is a lack of a starfield to speak of.

First up, the starfield.

Here is the first place where TESB comes in. When the created the 1981 sequence, they used the starfield from TESB. But it isn’t the straight starfield. It is offset so you can’t just lift the TESB starfield. Add to that it is at a sligltly different angle and there is some lens distortion and it becomes very difficult to use the TESB sequence starfield. Darth Lucas has noted that they reuse the same starfield over and over and found that most of the missing left side can be found. I thought about that, but after examining what is at hand, I have taken the route of pulling what I can from the best sources. Team Negative 1’s bonus with the SSE release isn’t completely lacking in stars, but it has only the brightest and you have to manipulate the image and pick the right frames to get what few are there. I started with that as one layer. Then I turned to the LD sources. The Japanese Special Collection is eliminated right away because of how it is cropped. The Definitive Edition LD (or the Faces version which uses the same source video) is the next best source. I could have gone with one of the newer archived versions, but my goto version for the DE is my TR47 set. I aligned the two starfields and painted out the stars that were on the 35mm scan. A more accurate version could use that for the left side and the TESB to recover even more stars on the rest of the screen, but that would create an imbalance so I stuck with just that. I did find a way to create a diffuse background layer of stars. The result is this:

A careful examination reveals that the moons do not move in sync with the starfield, meaning they were composited in later. A remaster must take this into account, which is why I created the starfield and moons on different layers. The moons and planet are the same in all versions, but the placement was different in the 1981 crawl. They are easy to recover from whichever version of the sequence you thing they look the best in. I used a layering of the SSE and the blu-ray.

Then we come to the other reason the TESB crawl is important. If you watch the video I posted, you will notice that each of the ANH crawls has the title receed at different rates. However, if you compare the 1981 sequence to TESB (something I just noticed this morning), you will see it exactly matches until it fades out (it is a simple fade out while the TESB title continues to receed for a few more frames (until the crawl starts) before vanishing. That means isolating the title from the TESB stars will give an accurate element for the title. The 97 SE (and subsequent DVD and BR releases) reused the crawl from the 1981 sequence, so separating that from the stars will give a good element to remaster that. Not quite as simple as it sounds, but not impossible.

That takes us to the flyover. That is probably the hardest part of the entire sequence. The 77, 81, and 97 versions all seem similar, but the 77 sequence has a noticable black line around the Star Destroyer as it crosses over the moon. They recomped it in 1981 to eliminate that artifact and used the same elements in the 1997 SE (but with the 1977 starfield, moons, and planet behind them).

I have not decided how I want to proceed with the flyover. The BR version has issues with the lasers. The broadcast version of the SE isn’t very clear (similar DNR to the GOUT and LD’s) and the two SSE versions are grainy. Perhaps some combination, but I have not decided how I want to proceed.

This has come up in a number of other threads and since I am determined that someone needs to do it (I’m trying, but it is taxing my skills), I thought there should be a thread to discuss it.

What better way to start than with a video that shows 4 title sequences together - the original 77 sequence, the 1981 sequence, the 1997 SE sequence and the TESB sequence. Why include the TESB sequence? well, that will become clear later. It shares to things in common with the 1981 sequence that are important for comparison. The versions I have used are the 2006 GOUT, the TR47 version of the Definitive Edition LD, the broadcast version of the 97 SE, and the 2006 GOUT of TESB.

https://vimeo.com/160390349
password OT

AntcuFaalb said:

yotsuya said:
From what we have seen from the Technicolor scans of Star Wars, there is some green tint, and DrDre has noticed a additional red tint, but in terms of contrast and saturation, the colors seem to very much match our other sources.

The two 35mm IB Technicolor scans poita has had done for our team have some considerable differences from our '80s Eastman dupe LPP scan, even when each is corrected to their respective print. As noted earlier, the techs tend to be more contrasty and are more saturated.

Well, the clips and images I’ve seen don’t really show that. I’m also comparing them to what I’ve been doing to restore the colors to both the GOUT and BR as well as reference photos. I see the color difference, but not the contrast or saturation. And from what I’ve seen, the other prints have some pretty serious issues that arise from the photo duplication process. You may be quite correct, but I haven’t seen it.

It’s all what you are familiar and comfortable with. I like Windows and Android, Mac is OK bit a bit confusing, but Linux is just a maze (or was the last time messed with it). My wife is Mac all the way and my best friend is a Linux guru. I’ve tried them all and I stick withe the one I know the best. I also use legacy software so switching isn’t really an option.

AntcuFaalb said:

yotsuya said:

For our beloved 1977 Star Wars, this does mean we can’t be 100% certain of the quality of the extant prints. Lucas obviously had the best one.

No, Lucas had to borrow a 35mm IB Technicolor print from a collector to use as a reference for the 97SE. This particular collector was fortunate enough to receive his print back from LFL when they were done.

Serves me right for listening to what GL had to say. I was under the impression that he had a print in his personal collection that they used.

yotsuya said:

the Technicolor prints are a bit off in a variety of ways. I’m thinking, since they were shutting down production in 77, that they are probably print control issues rather than fade issues.

The Technicolor IB process introduces color changes independent of quality control. IB Technicolor prints tend to have greater contrast, saturation, and are shifted toward green. They may be fade-free, but this doesn’t mean that they’re truly accurate to their source.

Well, the early Technicolor process had many color issues. They refined it as time went on to the point where next to 70mm, a Technicolor print was probably the highest quality image that audiences could see. From what we have seen from the Technicolor scans of Star Wars, there is some green tint, and DrDre has noticed a additional red tint, but in terms of contrast and saturation, the colors seem to very much match our other sources.

yoda-sama said:

You Windows users just make me sad…

There are users and power users. All operating systems have them. As a 24 year Windows vet, there isn’t much I can’t do. On this case, all you need is CCCP.

One thing to note, with a Technicolor print, you do not get very much added grain. That is a huge benefit in clarity. As a frequent viewer of TCM, I can tell you that they usually just telecine the prints they have on file in the archive for broadcast. You can really tell that the quality of some of the early Technicolor prints was not that great as even at 720p there is some obvious misalignment. It doesn’t diminish the fantastic colors or the fairly sharp image. Not as sharp as the high quality black and white, but sharper than some of the other color formats. Though I think with access to the original negatives and the color separations, modern computer technology could achieve a higher alignment. From what Mike Verta has done, I think you could even scan one of those old Technicolor prints and realign the color and adjust for any fading or printing issues and achive a decent (for home viewing or broadcast) HD picture quality.

For our beloved 1977 Star Wars, this does mean we can’t be 100% certain of the quality of the extant prints. Lucas obviously had the best one. They are a good reference, but the images that DrDre has been showing us tend to show that the Technicolor prints are a bit off in a variety of ways. I’m thinking, since they were shutting down production in 77, that they are probably print control issues rather than fade issues. I think if we combine information we are gathering about the color film stock prints with what we have on the Technicolor prints we will be about as close as we can get. We need to identify how the colors are different and apply that to a technicolor print scan to get as close to the original as is possible at this point.