mverta

Harmy, by now you know how anal I am about this stuff, so about color let me say that I will absolutely be offering "final/this-is-really-got-to-be-what-it's-supposed-to-be" images, but I'm just not ready to call anything final yet.  It won't be forever.  In fact, probably within the next couple of months I'll have some things together which I'd feel comfortable taking to the bank. 

It's a cluster.  Let's say, theoretically, you had multiple near-pristine Tech prints, and you noticed almost zero, or negligible, color variations between them.  You might be prepared to sum the prints, and take the average value for a given pixel and say "that's it," except the nature of film is such that one frame later, it's different.  So which one is it supposed to be?  Well, let's say you then took the sampling of every occurrence of a particular color (say, a costume) across the entirety of the film, and took the average. Well, that negates the deliberate color-grading which was done scene-to-scene, which intended for the color to appear different each time.  So then perhaps you just take it scene-by-scene, and pick an average value within that scene, and declare a particular color only valid for a certain range.

That's closer to definitive.  But that has a lot of room for error in it as well, such as drifts along the production chain, from negative to IP to IN to Print.  That process is not 100% stable.  And then, if they're prints, then they're old, and there's color fading to consider.  And then, what about at the scanning stage.  Are you sure you're capturing using a light temperature appropriate for the stock?  I only know of one scanner even capable of dialing in the color temperature of its light source on a per stock basis! And then, are you capturing to a colorspace which can faithfully capture the range on the stock?

And then, what about your working/monitor environment?

To really do this right you'd need a slew of things:

1) Reference photos of original costumes, sets, and models taken in 1976 (common; can be found in the private collections of original vfx artists and LFL Archives), and ideally with color charts in them (extremely rare, but exist somewhat). If you have charts, you have, at least somewhere, a record of what each color REALLY was; I mean, in life, irrespective of how they ultimately lit and graded the images. 

2) Data on the various stocks used when shooting Star Wars to determine their inherent biases and/or limitations in capturing color (this is obtainable).  From this you at least know something about what was even possible to capture, compared to the actual color in life.

3) Data on the various stocks used along the transfer process on the way to print to determine their inherent response characteristics (this is obtainable), and response characteristics of your prints (obtainable).

4) Prints which haven't faded, or have barely faded.  A pristine, or near-pristine Tech is about the best you could ever hope for on this front.

5) A scanner which tunes its light source to match the biases of the stock.  Currently the only one I know of is DFT-Film's Scanity.  Negative stock, for example, is orange tinted, so it uses a slightly green light source when capturing, the exact temperature of which is dialed in based on the stock.

6) A calibrated working environment, and flexible colorspace.

So, in the end, it's do-able.  It's just an odyssey of work, research, and asset acquisition.  Oh yeah, and a shit-ton of money. :)

When you consider all this, the fact that any of our fan-based corrections look as good as they do to as many people as they do, is pretty impressive, and something to be grateful for. 

_Mike

Midnight_Trooper said: For years I have looked at pictures of original Star Wars props taken by people at exhibits and the colors differ hugely from one camera to another. 

Yikes.  Cameras don't interpret light the way our eyes do, so unless every one of those photos had a Macbeth color chart in it, whose RGB values you know, and you were viewing them on a professionally-calibrated, color-critical display, you can throw them out.  Given the universe of monitor inconsistency among users, very few of whom have professionally-calibrated color-critical monitors, anyone doing professional color grading has to take most input with a grain of salt.  The total lack of standard in the world is the thorn in every colorist's side, but it's the way it is.  If Harmy - or anyone - manages to find colors which are "pleasing" to 80% of people, that is a freakin' miracle.  But even "pleasing" has little to do with "accurate."  At the level of subtlety being attempted here, there simply isn't a monitoring chain accurate enough to count on, and he can easily end up chasing his tail. Chances are that many people are actually making suggestions which would degrade the actual data in favor of biasing it to register a certain way on their inaccurate monitor.

So take it easy :)

_Mike

There is no one pristine Tech print out there that we know of; not even a rumored one.  The print which showed up at the Senator Theater, for example, is about a C to C+ at best, having a lot of platter damage and a somewhat dodgy Reel 6, though people think that looks great. There are far better ones out there. And, if you were able to acquire multiple Tech prints of superior quality to the Senator, you could probably get about as close to definitive as is likely to be gotten...

_Mike

The edges of an anamorphic image on film are not bowed/pincushioned; it is a square image.  The content of the image itself may have lens distortion, but that is a separate issue.  All my sources are anamorphic, and do not exhibit this behavior when unsqueezed. I've generated anamorphic visual effects for film outs hundreds of times as well, without bowing/pincushioning in the image. A compensatory curve of the screen in a theater should be necessary only when the projector's lens itself adds a bowing to the image, but the actual film frame images are square.

_Mike

Depending on how precise you want to be, your subtitles are either a touch too small or spaced a bit too close.  Check your "I've been looking forward..." shot to my reference at Greedo's right arm.

Hope that helps!

_Mike

You_Too I wasn't actually leveling my post at you as directly as it seemed :)  There's just been this general idea floating around and gaining legitimacy, and it needs to be stopped, that's all.  There are Tech prints out there, and they're just amazing, beautiful sources for restoration when treated properly.  Bottom line, even at their absolute worst, they're better than what we have now.

_Mike

Incidentally, Harmy, those shots are from the extensive library of photos and references from LFL known amongst collectors as the Kodak Photo CD set. There were many discs, and generally collectors have some, and incomplete, versions of them, and are looking to fill in the gaps and holes in their catalog through trades with other collectors.  Because the images have consistent prefixes, suffixes, and shot numbers, it's easy to see what you're missing. 

_Mike

The Technicolor prints of Star Wars are not anywhere near as contrasty as those shots-from-a-consumer-camera-off-a-frigging-movie-screen.  That is NOT what those shots looked like in a theater, and having first-hand experience with Technicolor prints of Star Wars, I can assure you that while Tech stock is "more contrasty" than some other stocks, we're talking about a tiny amount, and well recoverable in a good scan, provided you do it at multiple exposures, as any good scan would.  These screen grabs look practically like bleach bypass prints, they're so exaggerated.

The density of film stock is amazing, and contains data that is rarely accurately realized because of the limitations of bulb brightness and screen reflectivity in a theater.  This idea that Tech prints aren't suitable candidates for restoration is growing into an accepted truth when it's a complete myth, perpetuated by people who are not in possession of proper scans of Tech prints.  Let's end this now.

_Mike

g-force, it's a good cause, but unless you plan on documenting each shot individually, you won't be presenting an accurate accounting.  There is not a universal "thing" plaguing the Blu-Ray, by any stretch.  That many scenes appear to share certain qualities is coincidence, and even within that coincidence, vary wildly.

Most of what you're seeing with the Blu-Ray can be categorized by four classes: 1) Exposure issues, 2) Damaged element recovery, 3) Deliberate color grading, and 4) Lowry.

The exposure stuff I've explained before, in regards to the density of negative/film, and an improper handling of the gamma on the other side of an otherwise good scan.  As for #2, many elements had entire ranges of their color degraded, and you can spot instantly which color/elements were attempted to have things put back for.  The green lightsaber scene is a classic example.  In that sequence, watch the control panel in the foreground's colors, and notice when they suddenly change their entire nature.  Ditto a single shot in the sandcrawler over R2's shoulder where his green light panel suddenly shifts cyan.  That's element damage, where the film had gone green, and wasn't corrected for properly.  There's a ton of that, in various incarnations, all over the film.  You can learn to "feel" when the source had whole colors missing from it - the shots get a sort of sepia undertone to them.  Many Falcon cockpit over-the-shoulder shots suffer from this, where suddenly everything is sort of brown-ish, but other shots are fine. This is how lobster-men came about.  That's a compensation for damage in the elements; an attempt to correct one thing which has consequences elsewhere which nobody was watching or had time to correct for.

#3) is a huge category encompassing things like R2's ridiculously over-saturated blue; an attempt to match his appearance in later films, after which he was repainted using an entirely different process. So simply saturating his blue doesn't work, makes the frame look cartoony, and weirds out other things in the frame.

4) Lowry's process was immature compared to that work today.  LOTS of artifacts as a result.  Watch the Tatooine sand for 3 seconds and you'll see that with every successive frame it turns green, red, blue, yellow, etc.  Convert to LAB or YUV to really get a sense for color instability throughout the film as a result of that process.  It isn't on the negative that way, and it isn't on prints that way.  For an especially rough example, take the "go that way, you'll be malfunctioning in a day or two..." long shot of the two droids and keep an eye on R2's body.  Watch it pop and bounce in luminance and color.  For a real treat, examine the three channels on that shot separately and look how utterly fucked up the blue channel is.

Again, you will not find anything approaching a global set of adjustments which "explain" the blu-ray.  Like You_Too, you may find a setting which really helps, a lot of the time.  Your best bet in this is in regard to luminance.  You can globally correct luminance moreso than any other trait in the image.  But having restored this film from the 2004 DVD once, I can tell you that it ain't that simple.  Every scene has its own "thing."  I still wake up in the dark to the awful shifting tides of Tunisian sand dunes.

_Mike

http://originaltrilogy.com/forum/topic.cfm/Harmys-STAR-WARS-Despecialized-Edition-HD-AVCHD-DVD9-and-NTSC-DVD5-AVAILABLE-see-1st-post/post/548018/#TopicPost548018

msycamore said:

mverta said:

Absolutely possible.  What's curious is I've seen it on 4 4k sources with identical offset.

_Mike

 Hmm, then I don't know what to believe.

I suspect an offset made its way onto an IP or IN at some point and generated a batch of prints like that.  It's the only explanation I can think of outside of it being intentional.  However, if it's NOT an offset, then it IS a glow.  It's definitely not just a straight channel alignment.  However, that glow may be just a byproduct of the technique.

_Mike

Absolutely possible.  What's curious is I've seen it on 4 4k sources with identical offset.

_Mike

All composited elements in Star Wars weave independently of each other; there are no exceptions.

Subtitles move independently of the background.  Starfields move independently of the titles composited on top of them.  The sandcrawler matte painting moves independently of the plate it's composited on to.  If you stabilize one, you exacerbate the motion of the other. This is because optical compositing required multiple passes through the optical printer, and even the most accurate pin-registered gate cannot maintain absolute alignment. Each pass weaves differently. On an exposure area as small as film, the tiniest movement is obvious.  That it worked at all is miraculous, but is also why there are so many edge and fringing errors in composites from that era.

The aggregate motion of multiple items moving independently sort of cancels each other out to a certain degree, and you don't really notice it, but as I said, if you lock one, you suddenly can't miss the other.  I stabilize every single shot in the film as I'm working on it, temporarily, and I have yet to see an exception to this rule.  And every frame of Star Wars moves around in the gate to one degree or another.  There are no stable shots in the film.

Hope that helps.

_Mike

Actually, it's my thread, that's why I might read it.  The question is why are you still reading the thread?

Just can't get enough of me, I guess. 

You get yourself all worked up over somebody that has absolutely zero value in your life; couldn't matter less.  But you just can't help yourself: you have to watch, and you have to shout and stamp your feet. You have to really let me know what a jerk you think I am, 'cause, like, that'll really show me.  And if not, then at least you've done your good deed for the day of just reaching out and hurting somebody you actually don't know at all.  That's what passes for your discourse.

It's delusional to think you know somebody from a handful of posts on a forum, it's infantile to get worked up over somebody who doesn't matter in your life, it's telling that you keep checking the thread and participating, it's malicious that you take all those opportunities just to hurt somebody, and it's naive to think I actually give a crap.

I'd say see ya on my forum, but we don't serve your kind there.

_Mike

[/thread]

I think the mods should close this thread before any aliens who might be reading it give up on humanity entirely.  As the OP, Jay, please give these guys something else shiny to play with.

_Mike

I can explain it:  the loudest guys in this thread are stupefyingly ignorant blowhards.  It's a lot of Walter Pecks shouting about the fakery of it all with all the righteous pomp of a Dunning-Kruger test case.

That said, TServo, help me keep the site goodies for members only; lots of reasons.  We'll use, like, the honor system.  Cool?

You may now return to accusing me of feeling superior, yet spending all your energy constantly talking about me for three straight days after I've left the room.

_Mike

Something you may not have noticed before (actually, this crowd probably has, come to think of it):  Blue drop shadow on the closing credits.

Ladies and Gentlemen, I wanted to return here to let everyone know about the commencement of Legacy 2.0, but now that things are going again, I really need fewer social media sites to check rather than more, so all future discussion will be happening in the forums.

See you there!

MTFBWY,

_Mike

I do not have a film scanner, and I would think it would be very risky for any reputable company to agree to scan illegally appropriated materials.  As far as I know, the only companies who possess the best scanners deal directly with the studios regularly, and many of them are MPAA security-certified, which would essentially total their business if caught.  Certainly, soliciting for digital copies of illegally obtained materials would likely bring down the hammer tout de suite.

_Mike

Nice!  One thing I might caution you guys, though, is that my posted screencaps are WAY preliminary.  Long way to go before I can say with confidence they're right.  I think that Leia/Stormtrooper shot is probably 90%, but that last 10% makes a difference, sometimes a profound one.

Even the screencap copied above was updated minutes later with an improved one:

And that ain't final, either.  It's a long road.

_Mike

Darth Editous said:

Mike, the STAR WARS logo screenshot, on the same page as the colour fidelity one - has that had "stuff" done to it? I ask because the central yellow colour is exactly the same throughout - [255,237,0] - whereas the outer edges show some variation.

DE

Technically, that's an interim shot, but here's the general process with optical titles:

First, that slight fringing at the edges is typical of optical titles and you don't want to kill that.  Tiny as it may be in context, you can actually feel it and miss it when it's gone.  Secondly, for many reasons, color on film can vary from frame to frame in unintended ways.  The Tatooine sand is legendary for this: if you isolate sections and really watch, the sand goes from pink to green to blue to yellow to green to pink to... it's all over the map.  The stormtroopers in the compare shot have the same problem actually.  Now, obviously that's not inherent in the photography or natural, it's a photochemical byproduct, usually of aging.  One tool that I use watches pixel values for a given time and determines the average value.  In the case of the title, it is no surprise that the internal color then stabilizes itself to a consistent value.  But this tool also has the often unintended effect of completely removing the grain pattern.  That happens a lot, actually, as I have another tool which recovers details from within changing grain structure.  As it recovers, it also eats all the grain off.

But of course, the grain element is still known...  and it's easy to put back after the fact, without having to regenerate new grain.  By the way, generating new grain is done all the time in restoration, and even though it is virtually flawless (grain patterns for known stocks can be perfectly replicated), the purist in me likes to see the specific grain pattern on a shot put back.  Of course, that's only the grain pattern for a specific print/source (no two are exactly alike) but that's generally what I do.  Pull it out so I can stabilize color and recover details, and then put it back.  So with the title, you're essentially seeing the grain-free interim image.  The comparison stormtroopers/Leia shot, on the other hand, has its grain intact.

_Mike

Yep.  It's a small internet.

_Mike

There is absolutely no way, logistically, this kind of restoration makes sense.  You can't throw bodies at it, because there aren't those bulk-repetitive tasks like dirt clean-up which comprise the most of it.  And you can't sit down with a pro-colorist and just hammer the thing out even over months because it's not a "color/grading" kind of thing.  It's literally where every element of every frame needs to be individually considered.  In most cases, I'm pulling every single element apart and treating it independently - that means skin, lips, teeth, shirts, pants, walls, etc.  It's like rotoscoping every element of every frame.  Now, that's not always necessary, of course (!), but whoa - lots of times.  That is, you have to if you want to really try a "recover the negative" kind of thing, because as I said, no one print has it all.  So we're talking just years of maybe a handful of people doing that, because the people have to change skill sets with each shot, generally.

I was trained as a generalist - we have to be able to do a broad range of tasks for a job, so that comes in handy.  I've tracked, pulled keys, done digital grading/painting, clean-up, 3D work, compositing... just tons of stuff, all of which I use constantly on this gig.  And I'm also (actually, primarily) a composer, sound designer, and re-recording engineer, so that part of the task is also in hand.  I have facilities and hardware for the gig, and I'm also in software development so I have friends to write tools as I need them.  So really, it's just a lot of things need to happen, and they take a lot of time, and if I was charging pro rates, it'd be millions to do.  Nobody spends that kind of time and money.  Nobody.  I mean, they COULD, sure.  But... they just don't. 

When it's over?  Christ.  That's a long way away.  Ask me in 4 years. :)

_Mike

Don't sweat it, though I appreciate the kind words and support.  There are some people who get their identity by being tiny little kings of tiny little places in irrelevant corners of the internet, and they naturally assume everybody else is, too.  I've said it before: having the world's best Star Wars restoration isn't something to brag about.  Being a good husband, father, and citizen might be, but nothing you did as a pet project for Star Wars counts.  Have some perspective, for Chrissakes.  That said, if you're going to do something, then dammit, do it right.

And I'm as excited about watching Legacy unfold as anyone.  Truly, it's love that drives it, and not the praises or criticisms of others, much as the former is appreciated and the latter is potentially useful.  It's a show: try and enjoy it.

_Mike

I am exceedingly pleased with the good fortune my principles have afforded the project, yes.  Who wouldn't be?  We don't get rewarded in life for the doing the right thing often enough.

I have applauded the efforts of all, generally, but been honestly critical when asked, specifically.  I think you're in the minority, what with getting your panties in a twist when coming across criticism you don't agree with.  Seems to me that the fine members here are mostly aligned in that we all want to do the best we can do.  But if you're looking for a sunshine blower, ready to hail as wonderful merely passable mediocrity, you got the wrong dude. That doesn't help any of us.

And forgive me, but I'm not trying to bring people together; this isn't burning man.  I'm assuming we are a community because of mutual sentiment about the film.  Legacy is a celebration, and those who celebrate will - and do- so in their own way, and plenty, I've noticed. 

Anyway, feel free to hurl pejoratives to your heart's content. I won't tell your mom and dad you're on the computer this late.

_Mike

... and yet that is the very reason mine's superior.  Ironic.  Life's funny that way.

_Mike