yoda-sama said:
CLV smear shouldn't matter much to Star Wars projects, as the Definitive Edition is CAV (which'd be all around better anyway, what with every frame of the movie being present).
Definitive Collection = GOUT (2006 bonus DVDs)
There's no point in preserving that release IMHO.
The same goes for the THX "Faces" LDs.
captainsolo said:
AntcuFaalb said:
Continued...
The closest to a "100% pure video signal" from an LD that I think we'll ever get is a TooT of 3 captures from 3 different copies of the same movie on LD (pressed by the same pressing factory, preferably) on an LD player with a >51dB signal-to-noise ratio that doesn't smear CLV LDs (if the LD in question is CLV, that is).
What a mouthful!
What player does all that exactly? Silly me would buy one of course. ;)
Plenty. The Panasonic LX-900 is one (affordable) example.
If you get one manufactured in late 1994 or 1995, then it won't suffer from the minor AGC issue that impacted the 1993-manufactured LX-900s.
I have two: a 1995-manufactured one for capturing and a 1993-manufactured one for parts.
I've started a "Star Wars Taxonomy" thread: http://originaltrilogy.com/forum/topic.cfm/Star-Wars-Taxonomy/post/644708/#TopicPost644708
Please weigh-in.
I’m interested in identifying different releases by the intermediate film print master used. This kind of taxonomical information is useful and interesting to preservationists like ourselves.
For now, I’d like to focus on ANH. We can focus on TESB and ROTJ later.
We need to identify a set of easy-to-spot characteristics that will allow us to quickly categorize different releases.
All I can think of at the moment are the “Tantive Orange Errors”: http://fd.noneinc.com/Glitch/starwars-tantiveorangeerrors/starwars-tantiveorangeerrors.html
Using this characteristic as a guide, we can assume that all home video releases of ANH come from one of two IPs: the one without them (e.g., the JSC LD) and the one with them (e.g., the Technidisc LD and all 35mm theatrical release-prints).
Interestingly, the IP without the “Tantive Orange Errors” is also the one without the burned-in subtitles for the Greedo scene.
Please weigh-in, guys. I’d like to come up with a Star Wars Family Tree.
Excellent thread, Doctor M. Keep up the good work!
Considering that your suggestions range from LDs to BDs, it would make sense to choose DVD as the format most appropriate for a preservation series of all of Disney's animated theatrical features.
ww12345 said:
Oh, I reserve my judgement then. :)
Don't. Admittedly, LD is a crap source.
I'm just saying that it's less crap than people think.
ww12345 said:
Hahahaha! :)
I don't mean LD is a bad source, it's just that at a certain point, the material exceeds its usefulness in time or expense - just like it would be foolish to have a 4k scan of an 8mm home movie...
Agreed, but I still believe that this community hasn't seen what can really be pulled from an LD.
Myself, poita, and a few others are working on that right now. :-)
ww12345 said:
I don't know - LD is not a great source anyway (IMHO, not high resolution enough) so I don't know how many people would be clamoring for an electron scan of all of the Star Wars LDs... I could be wrong, but that's just my guess.
~528x480 is enough for me. ^_^
I get more pissed-off by artifacts: aliasing, damaged/dirty print, etc.
ww12345 said:
Here's part of R1, in SD:
https://www.dropbox.com/s/p5he2dgw1ehiuvr/Reel%201.mp4
Awesome!
Did you already get the HDDs from poita?
Also, what constitutes "original"?
The "tantive orange items" were part of every release print of Star Wars, including the IB Technicolor prints and the prints with the spliced-in '81 crawl.
http://fd.noneinc.com/Glitch/starwars-tantiveorangeerrors/starwars-tantiveorangeerrors.html
I'd consider any release that's missing 'em to be non-original, but I'm a pain in the ass. ^_^
If this happens, then let's hope they do it right by restoring an IP rather than trying to work with the negative.
The last thing fans of the OOT need is some kind of botched attempt to restore the film from the negative.
Think about it: How likely is it that the team working on the restoration would know about, e.g., the new optical wipes done for the 97SE by Pacific Titles?
The lowest generation source for the original-original-trilogy is an IP. Period. The. End.
Moth3r said:
AntcuFaalb said:
nightstalkerpoet said:
Excuse my ignorance (as always) but could you simply physically scan a laserdisc surface at a microscopic zoom rate, and use a pc to analyze the lans/pits and pull the information from it? That way you could pull a 100% pure video signal from it?
That would most certainly be possible, but very difficult. Consider that the LD itself might be subject to minor warping (many are) and you'd need one hell of a digital tilt servo to compensate for that while tracking.
Then consider that you'd have to contend with occlusion defects (dirt and other material between the plastic and the aluminum layer), possible laser rot (oxidation of the aluminum layer), ..., the list goes on.
And also consider that you'd need access to an electron microscope:
That's awesome, Moth3r. Thanks for sharing!
rnranimal said:
How sensitive to the disc's surface cleanliness is playback, compared to CDs & DVDs? Some of my laserdiscs play back with various specks (not actual rot). I know that's common for laserdisc, but was wondering if that's something that can be improved by cleaning the disc. My discs mostly look fine, but aren't cleaned to a shine. I'm always afraid of marking up the surface by trying to clean it, so I've always just played them as they are. But I'm starting to transfer some to digital now and wonder if a cleaning is in order.
A cleaning can't hurt (see below), but LDs aren't anywhere near as sensitive as CDs and DVDs are when it comes to surface cleanliness.
If you do choose to clean 'em, I suggest using Novus #1 Plastic Polish and a good microfiber cloth. Also, consider using Novus #2 to buff-out scratches.
Continued...
The closest to a "100% pure video signal" from an LD that I think we'll ever get is a TooT of 3 captures from 3 different copies of the same movie on LD (pressed by the same pressing factory, preferably) on an LD player with a >51dB signal-to-noise ratio that doesn't smear CLV LDs (if the LD in question is CLV, that is).
What a mouthful!
nightstalkerpoet said:
Excuse my ignorance (as always) but could you simply physically scan a laserdisc surface at a microscopic zoom rate, and use a pc to analyze the lans/pits and pull the information from it? That way you could pull a 100% pure video signal from it?
That would most certainly be possible, but very difficult. Consider that the LD itself might be subject to minor warping (many are) and you'd need one hell of a digital tilt servo to compensate for that while tracking.
Then consider that you'd have to contend with occlusion defects (dirt and other material between the plastic and the aluminum layer), possible laser rot (oxidation of the aluminum layer), ..., the list goes on.
Also, keep in mind that the signal stored on a laserdisc is composite and, to the best of my knowledge, there are no free-and-open-source 3D comb filters that are as good as the proprietary ones.
.
.
.
So, it's hard to say what a "100% pure video signal" is when talking about laserdiscs. The information isn't stored digitally and each pressing has its own characteristics.
Definitely make your own thread, Doctor M. My thread, unfortunately, sucks.
I'll gladly put a hyperlink to your thread in the OP of mine to send people the right way...
poita said:
Unfortunately "PNG files with 16-bit color depths (48-bit or 64-bit pixels) are not supported."
:(
I didn't know that! That sucks... :-(
poita said:
Scanning is finished at last!
I am uploading an SD version now, about 900MB per reel, and am going through the s-l-o-w conversion to a .PNG sequence and will be posting the HDD back to the owner of the film to do with what they will.
It has been an interesting experience, I will post some before and after images from the scans. There is still a lot of work to do, but hopefully by scanning with controlled light separations and custom LookUpTables (LUTs) it has reduced the amount of work required to get some colour out of the red mist.
I'm looking forward to seeing what you all do with it.
You might want to consider using PNGOUT (http://advsys.net/ken/utils.htm) to reduce the PNGs to a provably-minimum size.
Darth Lucas said:
poita said:
No, the faces set is the same as the definitive edition discs. Nothing to do with the scans for the SE.
And Darth Lucas, the faces set, and all laserdiscs as far as I know have the Ep IV ANH crawl, not the 1976/1977 crawl.
That's what I said. The 2006 DVD was suggested, and I was saying that I think that version just said Star Wars, and that I was looking for the crawl to say A New Hope. Most people use the theatrical crawl in their preservations.
Unfortunately, our community has yet to produce an HQ version of the 1981 crawl.
DuracellEnergizer said:
TV's Frink said:
Oh...kaaaaay.....
You have no sense of tumour.
It's not a tumor!
Thanks for contributing, drngr!
The DMR-ES15 uses the same in-house-developed A/V chip as the DMR-ES10.
It's the Panasonic C1AB00001979 according to the service manual.
Hey, it's me. said:
I just discovered something curious on Twitter. On George Lucas's official twitter page he has posted a link to a short film (roughly 5 mins long) called 'The Game' and has said that if it gets 100 views or retweets, he may divulge some info about Episode VII.
His Twitter page... wow. :-/
monks19 said:
True. So saying what colours are the right one when you compare all the home video releases is, sometimes, pure speculation.
I think there's value in "how most of us remember it".
If most of us remember and enjoy the theatrically-inaccurate colors of a particular home video release, then I'd say that's the one to try to match.
We'll never be able to get a release to be 100% theatrically-accurate anyway; not on these 8-bit displays.
Spaced Ranger said:
If Zip Doodah has full access to the 35mm files, perhaps he can run an Avisynth script to sample a single frame inside each shot change. That would produce color correction targets for the entire film. I saw such a script once. I'll see if I can track it down, unless you know or have one.
All you need to do is use the SCSelect() function.
This is excellent news. It'll be my first Conan experience as well!
Thanks Team Blu!