team_negative1 said:
As mentioned previously, the comparison goes through at least 3 consecutive compressions, thus causing the detail loss.
We'll post the standalone video shortly.
Team Negative1
Okay, I would have assumed that the compression would have affected both the before and after samples in much the same way, but I guess not.
The Derann prints?
Derann UK bought the rights to do Super8 releases and ran off quite a few. It was all above board and sanctioned.
Woah, what happened here?
I don't remember what clip was posted on page one, if anyone still has it, I will take a look.
team_negative1 said:
I think all of you do realize we are only releasing the original version.
==========================================
Check out Myspleen right now.
Feel free to do your own color correction, and cleanup. That is only a temporary sample with a blue tint. Some don't like the overcast yellow of the original.
Team Negative1
I can understand the dislike of the yellow, I'm not sure what the purpose of tinting it blue was?
Puggo - Jar Jar's Yoda said:
team_negative1 said:
This is the new version, that should become a standard, especially replacing Puggo's 16mm for film fans of the original.
Being a big fan of your work, the last thing I want to do is sound defensive. And I have never, ever claimed the PG was anything but a grungy curiosity with mild research value. But comparing our respective posted screenshots, I'm a bit confused by your above comment:
I agree completely.
I own the print.
Yes, I know him and he has a German Tech IB print. It is pretty nice.
Nope was just missing on the HDD when I sent the scans to the US, one of the reasons I want to rescan, I'm not at all happy with the original scan.
Just to be clear, this scan had nothing to do with the -1 team, as far as I know they picked up the raw scan from Anctu and decided to do stuff with it (which is fine, that is what it is for, I'm happy for them or anyone else to do whatever they like with it). I have nothing to do with the various versions of it that have appeared/may appear.
As for it replacing Puggo's version in any way, that is just nonsense. It is just a scan of a Super8 release, and not a very good scan, it was really just a test of the scanner to see if the Super8 version was worth archiving, and for others to mess with as they see fit.
Each scan is what it is, I'm all for open discussion as to which versions have more detail/less detail, grain and contrast differences etc. as I find this stuff interesting, but the notion that somehow a scan of a Super8 film somehow replaces someone else's scan of a 16mm film is kind of beyond me.
I'll be having another crack at it when well enough to do so, people seem interested in seeing what the Derann release looked like, and it does have some differences to other film versions, so it seems there is enough interest to do a good job of it.
Just enjoy the scans folks :)
Different labs have different marks and they are not always in the exact same spot on the film.
Up until the late 60s the negative was often punched, in which case the marks will be the same on all prints, then you have to pull the information from surrounding frames. The reel changeover points are often on fairly static shots, so again, this is rarely a problem.
The first step will be to colour match the two sources, there are many ways to do this, but I would use Nuke and work it via a script to get the sources as close as possible.
Once you have them colour matched, then you need to align them, you could use Elastix if you want it exact, or could use the warper in After Effects or similar to get it close.
Then it is a matter of blending the footage using some soft masks, you should be able to extract a mask for the subtitles as they will be a particular 'colour', and then create a soft blend of the PAL material into the Japanese footage.
If you have multiple prints with cue marks in different places, then it is relatively easy to remove them.
If not, then it is a matter of replacing them with data from surrounding frames and ensuring it matches in seamlessly.
No Photoshop required, there are a swathe of film restoration tools for this kind of thing, it is relatively straightforward.
Danfun128 said:
Does team negative one have a pervert department?
No, Team Pervert has a negative one department.
A forum member is already bidding on it.
If anyone has a spare reel2 of Jedi, let me know.
I capture all LDs with a 3DCF and then again with a 2DCF and replace the scene change frames. It really does seem the only way to get the best of both worlds.
In reality though when watching a movie, no one will see it if it is a single frame duration, the scene change is enough to throw the viewer for more than a frame.
But I can't help it, knowing it is there is enough to irk me....
age said:
Wow! This is a very high quality scan.Thanks for sharing.
What wide gamut color profile was used for this?
I am trying to setting the black point on the reel 4,it's possible that the original footage is a bit dark?
I'm actually not all that happy with the quality of the scan, and will be rescanning next month.
That is correct, he has stated a couple of times that the Cyan was out of sync, but other reports state the seps were used for some of the restoration, so they definitely have them, but probably didn't use it for much for the SE.
As to whether it would now be usable is an interesting question.
RU.08 said:
They'd have to use a print to show them the structure if there's no negative (i.e. the o-neg is configured to the 1997 SE).
As far as I know, they still had the three strip seperation masters to work from.
It will be done when it is done. 35mm stuff bumped the priority down, but it will happen.
BTW, could you guys post with a signoff of -A, or -B or -C or something? I really hate answering to a committee! (ittee...ittee...ittee)
Image registration can be done using Elastix, it gives solid results, but isn't fast.
Thanks for the detailed infographic :)
That is basically what I was suggesting, with a bit more work on the image registration side of things. It is a good technique if you have a source that has good colour, the colour resolution can be quite low and still look great overlaid onto a source with high luma resolution.
SilverWook said:
AntcuFaalb said:
SilverWook said:
I wonder if the sort of color video voodoo the Dr. Who restoration team has done with episodes that exist as black and white prints could be possible with this?
Unfortunately, no. IIRC, the BBC team relied heavily on the fact that the video was transferred via a composite medium by using information gleaned from the dot crawl to figure out which color caused the interference in the luma (B&W) channel in the first place.
Actually, I was thinking of how they took the color signal from a fan's Betamax recordings and combined it with the monochrome 16mm.
Yes, this is quite possible to do, basically you would register the 16mm print to the other source of choice (laserdisc, DVD whatever) and then you could take the luma from the 16mm and overlay the chroma from the other source, it works a treat.
Ownership of the 16mm ex service prints is actually not a problem, 35mm, not so much...
Sorry, I misread your post and thought you outed yourself!
