RU.08

UnitéD2 said:

So, what we call original negative is not the image directly captured by the camera for each shot.

It can be, or it can be a dupe with the camera negative in stored away. Some films even have more than one ON. Profondo Rosso/Deep Red may have had two entirely different ON’s - I don’t know I haven’t looked into it, the alternative is the director had the ON reconfigured for export. My understanding is it would have had an export ON seeing as the 4K restoration was made from the ‘original negative’ implying that the Italian Cut was intact.

Can a 4K scan of a very good print be as sharp as it is ?

Well clearly it can be. Prints do get scanned and released on Bluray when there’s no ON or IP available. Studies will use whatever they can, depending on the condition and their budgets, but they prefer using the ON or the camera negatives where they can - Robocop 4k was scanned from the camera negative for example. Obviously that could be much more expensive and time consuming than simply scanning the ON because you may have to first catalogue all the camera negatives and find all the pieces that were used in the final cut of the film before you can even start work on scanning it.

UnitéD2 said:

From A print ? Is it not a combination of numerous prints, in order to obtain an image as accurate and sharp as the negative ?

Doesn’t matter how many prints you use it’s still not as sharp as the ON (original negative) or IP (interpositive) they’re made from. With SW there are some scenes that even on the ON are a 3rd, 4th, 5th, or even 6th generation copy of the camera negative (optical wipes for example are at the very least 2nd generation copies and possibly 3rd or 4th generation depending on other factors) - for the 1997 SE they scanned every camera negative they could to replace and re-composite material that suffered from generational loss. That is to say that not even the ON was as sharp as some of the stuff that’s in the 1997 SE ON (such as the speeder sequence through Mos Eisley).

theMaestro said:

They’re even less than 1080p? Wow. I wonder if Lucas has regrets about shooting them that way now.

They are 1080p, and were shot in full 1080p, but most AOTC scenes were compressed with 3:1:1 subsampling (i.e. true resolution of 1440x1080 for chroma and for 960x540 luma) as that was the limitation of the HDCAM tapes. It was not shot using an anamorphic lens, therefore it was cropped from 1080 to 817 vertical lines. For the most part the result looked great (even on IMAX), however the low chroma resolution posed problems for compositing, and those are definitely visible in the end result. But they didn’t always use HDCAM for storage, so some scenes were shot and stored at full 1080p with 4:2:2 subsampling. IIRC the cameras were much larger and more cumbersome when compared with 35mm cameras, so they had a lot of disadvantages, beyond the relatively low resolution of the HDCAM tape storage.

RoTS was shot and stored at full 1080p with 4:4:4 subsampling, also it was not shot using an anamorphic lens, and therefore was cropped to 817 pixels as well. So it’s true that AOTC is mostly below full 1080p resolution (for the live-action stuff), but most/all of ROTS is full 1080p for all live action (minus any occasional scene that was shot early or borrowed from leftover AOTC material).

That sounds like fantastic news!

thxita said:

Not that VHS and Laserdisc were faithful transfers either… but I guess better than the DVD and Bluray versions.

They were faithful but limited by the format and the technology of the day.

Yes that’s true, I’m already more than a 10th of the way there with just two pledged donors so far, so don’t hesitate to contact me about this if you’re interested in contributing something, you can of course remain anonymous as well. It’d be great if the money was raised by the end of the year and the print can be sent to the scanner.

But regardless it will be made available once completed (even if you want the 4K Dnxhr file). And then everyone can see exactly what I’ve been talking about as to how different it really is to the home video releases. So far that I can tell it’s really the most different of the Disney animations compared with the current releases available. There’s a cropped HDTV version of LATT for example (which you’ll find on rutracker) that looks pretty accurate to its film counterpart, but for BATB there’s nothing out there that looks right. There are so many great Disney animations, and my aim is to help get them all scanned. There’s a number I’m aware of that other’s have already had scanned, and there are certainly more to do after this one!

I saw it on film in 2014. The night scenes are dark, but the day scenes are bright. When Belle borrows the book IIRC it’s quite dark inside the bookstore (darker than the bluray).

towne32 said:

The star destroyer is most likely red because it is from a faded Eastman print. It’s a nice idea, though. 😃

Yes that’s why I suggested to match the 20th cent Fox logo. The 81 flyover is on Eastman and spliced on to previously 1977 prints (they weren’t newly struck, why bother after just 4 years?) If you match the logos then it should tell you more or less how the rest of it looks. 😃

Yep I just need to build it. I’ve already done all the work really, with just a couple of technical issues to conquer.

Williarob said:

And here is the new Mt2s file (for the '81 crawl only):

https://mega.nz/#!45gFWYCB!0xcOeCA2UrWoH9tVvlXhoTqEmDrS3wB7fVNcObf02MU

Let me know what you think. Also let me know if you can simply swap out the old file of the same name in the iso. if swapping out the file breaks the blu-ray, let me know and I’ll upload all the supporting files…

Looks great, thanks!

Yeah I don’t want to see anything rotoscoped. The real 81 flyover looks significantly different to the recreated one in a number of ways, and partly it’s because the 77 flyover appears to be used as the basis for the recreation. Here’s some examples:

You can clearly see different lens-flare for example, and the matte-line is almost non-existent in the 1981 version when compared with the recreation.

I think the best way is to colour correct the 20th century fox logo, and then leave the rest of it. The 20th cent logo should look the same in both versions, but every other part of the 81 crawl/flyover was newly done and has its own subtleties. Very Nice work Rob, can’t wait to see the upload!

Okay, so there are a few differences between BATB on 35mm and home video. They are quite substantial.

For comparison I like to mention Aladdin. Aladdin has bright vibrant colours, and dark blue nights. The bluray looks about 80-90% accurate in terms of colour and overall look of the film, and there are a few subtle differences in the pallet in some areas, and there is some animation that was tweaked for the IMAX version (though these are more subtle than the animation changes done to Lion King).

BATB looks very different on film to any home video release. It’s a lot darker than the VHS/Laserdisc. I’m told this is because in those days it was difficult to keep dark scenes without loosing the picture entirely, and detail decreased hugely the darker the scene. But the colours and such are at least mostly accurate (although there might be animation changes, that won’t be something I can confirm until we can see them side-by-side). So the VHS is maybe 70% accurate or so. The DVD and Bluray look completely wrong. That is to say maybe 10-20% accurate (accuracy is probably limited to the darker blue night scenes that have little colour).

The DVD/Bluray have brand new colour pallets, pulling the CAPS files directly and making whatever changes to the colour that they wanted as they went along. It’s not yet clear if this was a regrade or recolouring, and I’d love to be able to apply DrDre’s color tool to it to find out. Although I believe it was recoloured for the IMAX version and that the changes were made to the CAPS files themselves rather than simply applying a different colour grading. There are also substantial animation changes on DVD and Bluray that were made, again believed to be from the IMAX version.

SnooPac said:

Well, apparently the stupid mount edits the ISO file for some reason:
http://unix.stackexchange.com/questions/149459/mount-o-loop-changes-mounted-iso-image-file
If I had just run the command with a “ro,” in there before the “loop”, all would have been fine.

If that’s the case then the ISO definitely wasn’t corrupted - it’s just a slight different .iso file with all the same contents. So it did not cause the glitches Blackout had.

Williarob said:

I didn’t use x264 to encode it, but can certainly encode a compatible file for you. Just need to see if I can dig up the lossless version…

Aha, but, h264 is h264 isn’t it? It’s decoded in the same way no matter what writing library is used, so as long as the settings match they can be joined, no?

Williarob said:

Honestly, after all the hard work that went into making it a proper blu-ray, it pains me to see people just ripping out the video to make MKVs, but I get it.

Well they are much more versatile. I prefer to rip a BD I own and stream it from my PC than to watch it straight from the BD - and in some cases it’s necessary (there are some blurays I own that don’t have either English audio or subtitles for example). I don’t have to worry about glitches like if the disc skips, or if the HDCP loses sync.

As far as I can tell though there are no errors in the file.

Also Rob, could you please post the x264 settings (as well as the version you used and whether it was 32bit/64bit), that will allow me to encode the true 1981 craw to the correct settings. I’ve already got it ready (based on the 1.0 Extras file), I just need to know the settings. Also, if you have a better quality version than the one from the 1.0 extras please let me know (even if it’s just the same but at a higher resolution, given that I have to crop and resize the file from an already 1080p encode). Thanks.

Really great work on 1.6 by the way!

V

Here are the hashes on the one I downloaded from the Spleen:

File: Star_Wars_SSE_TN1_v1_6.iso
CRC-32: 8eda63f4
MD4: e960cb3bda80b4269393b3fdbbbfeaed
MD5: 70bda9ddfc4ccf4c6690d5028723a854
SHA-1: 9b5df42aa7a7d346711dd969f8650f18b6d5e99f
SHA256: 183e36fe4591e7f947b9d63d07f1bb920f1039a35e9853355d835e3a33df132f

RooBee said:

Guys… what do you think of this?

They just announced it. I mean, come on… that poster… Mike’s talkings with Disney… Could it really be true? With an announcement of the release of the whole trilogy next year to celebrate ANH?

It says “A New Hope”, not “Star Wars”. If they were screening the original movie it would say “Star Wars”.

Blackout said:

One problem with mine was the audio syncing: notably while the admiral is mocking Vader’s religion

I just checked that point and they were both in sync. So I’m not sure what caused your problem, but I am pleased that it doesn’t appear to be an inherent problem of mkvtoolnix. 😄 But certainly yes makemkv beta will also perform the job wonderfully, and there are certain types of video streams that it handles better than mkvtoolnix when joining together from branching. But as this was an audio problem and not a video problem, and as x264 would be one of the best streams supported for joining, that’s unlikely to be the problem.

Blackout said:

MKVToolNix and yes. My specs are probably just not up to par.

Btw, Kitt is the person I have been talking about.

I’ve just made two MKVs, one with mkvtoolnix and the other with makemkv (so I can see the results). Can you give me a specific example of a problem with yours? It is certainly possible that mkvtoolnix is having difficulty joining together certain types of tracks while maintaining interleave points perhaps that’s the source of the problem?

Blackout- which program are you using? And did you extract the .iso first?

joefavs said:

if the crawl and flyover is 0:02:27:058 long, do I want to start the audio for the second part at 0:02:27:058 or 0:02:27:059?

You cut it at the same point. If the flyover is 0:02:27:058 then you make two parts:

#1: 00:00:00:000 - 00:02:27:058
#2: 00:02:27:058 - End

towne32 said:

Unstarring it is an interesting choice, if they did that. That means that whomever is still leeching it while at work, etc, might have the 38 gigs of a broken download affect their ratio, though the torrent they clicked on was for a working download that does not affect their ratio.

The uploader self-reported it. It won’t affect ratios: 1. because it was freeleech, 2. because once deleted the tracker no longer collects the peer information. No one had completed it to 100%, but some got to 98% or something (I suspect this was part of the uploader’s problem was that his file had somehow become corrupted).

(duplicate)

I’m indifferent as to the whole Mike/technicolor thing at this point. Whatever happened, it seems like it was a complete mess on both sides and escalated into silliness. Hell, I agree with negative1 that it sucks that the torrent was removed from myspleen.

(edit to clarify …)
P2 and P3 are collectors that own (or owned) prints of SW. Both loaned them to Mike, and P2 paid for both to be scanned. This is where it gets complicated. P2 held raw copies of both scans, and agreed to give them both to N1. N1 either believed here that he had the knowledge and consent of P3, or believed he didn’t need it, and we don’t really know which it was. As it happened, P3 did not want his scan to be publicly released. Mike found out about this and was (probably justifiably) furious. What happened next is the important bit…

Mike convinced P2 not to give N1 a copy of either scan (including P2’s scan of P2’s print that P2 paid for). N1 would have viewed that as vindictive and unnecessary for the purposes of protecting P3 and P3’s wishes. Later Mike asked to rescan the Spanish LPP and N1 said no. So really it’s my opinion they both share culpability in that mess, especially if it is indeed true that N1 knew that P3 didn’t want people other than P2 and Mike accessing his scan.
(end edit - hopefully that’s accurate)

But other aspects are indeed “disasters”, and I don’t agree that his work is a free pass for his behavior (nor should anyone’s be).

Well, it’s his prerogative if he wants to go onto Myspleen and ask for the torrents to be removed, it’s up the Spleen admins to decide what to do. Rob didn’t exactly interject in the matter either. N1 came on here and asked for all these threads to be locked, and they weren’t. And again, none of the ex-members interjected (besides of course ignoring it and continuing to post). It’s odd behaviour yes, but it’s hardly worth bashing someone over. As I said ages back, if it’s not on the Spleen it’s not the end of the world, it’ll surely pop up on another tracker.

towne32 said:

Heh, always the biggest negative1 apologist on the site. And now literally demanding apologies from people posting on the myspleen comments. 😃

You don’t bite the hand that feeds you. Remember, N1 bought the print himself. He loaned it to poita and to MikeV so they could have their own scans of it done, and I believe (but do correct me if I’m wrong) that poita’s restoration is based on N1’s Spanish LPP as the primary print. Anyway, if you want to bash anyone and focus all that negative rage somewhere useful, aim it at Lucas, Disney, or JJ Abrams! 😄

(edit to clarify): Poita is in fact not using the Spanish print in his restoration.