logo Sign In

Yet another preservation, Star Wars Trilogy: Throwback Edition (* unfinshed project *)

Author
Time
 (Edited)

Let me start out by saying I wish I had found this site and board sooner.  Since the late 90’s I have missed the OT.  In 2004 I missed it even more once I saw the retail DVDs.  The GOUT release got me close, but with its non-anamorphic video, it just looked too small on my large home theater setup.  I kept out a little hope for the blu-ray releases but as I suspected, nothing.

Here is where this board, and help form all of you have come in.  I have read through hundreds of your posts, made notes, and studied all of your screen grabs.  I am now ready to start my very own project, “The Throwback Edition”.  My main video source will be the GOUT DVDs.  My goal is to recreate the films I fell in love with as a kid, and bring them back to their theatrical presentation as best I remember.

I will be using Final Cut Pro, Color, DVD Studio Pro, and Photoshop in my project.  I am not a script guy so everything will be done “manually” if you will.  I will color scene by scene, fix aspect, create new subs, and author DVD with multiple audio tracks.  No timeline of completion has been decided, just want it done until I am satisfied with all three films.

One thing I should add, I love the look of film.  I like it when my blu-rays of 80’s movies have a hint of film grain.  I am not overly worried about cleaning up all the grain, gate, and age problems of the GOUT transfer.  Sure, I will filter some of it best I can, but these are movies that are over 30 years old.  I like the look of older films, they look really nostalgic on my large screen.

Thanks again everyone for everything posted on this board.  I will follow the tradition and keep updates coming.

Thus far I have ripped all three films.  Converted them to a Apple Pro Res HQ quicktime using mpegstreamclip and brought them into Final Cut.  Next will be to correct the aspect ratio, and then color.  Star Wars will be done first.

DVD’s will be 480p.  No need to create them in HD as my player does such a good job on the upconvert.

Author
Time

What can you do to the gout manually that will be so much better than what it already is when put through the g-force script? I don't mean to discourrage you, but the gout can only look so good, no ammount of manual labour will ever turn it into A-grade DVD quality, let alone anything else!

[ Scanning stuff since 2015 ]

Author
Time

I say welcome with your project.  Nobody ever succeeded by doing nothing.  Best of luck, I'm sure you will find many more problems than you've expected.  Some of the guys on here are very good at what they do and know these 3 films like the back of their hands so you've come to the right place. 

Luke threw twice…maybe.

Author
Time

RU.08 said:

What can you do to the gout manually that will be so much better than what it already is when put through the g-force script? I don't mean to discourrage you, but the gout can only look so good, no ammount of manual labour will ever turn it into A-grade DVD quality, let alone anything else!

I say I am doing it manually as I do not know how to use their script.  May not come out looking as good, but will be a fun project. Going through scene by scene give you the access to adjust the keyframes on filter and color as needed.  Butt also like I said a lot of the problems that bother people with the gout release do not bother me as much as I do mind the look of aged film.

 

 

Author
Time

From reading, it seems that what upsets most peeps here is the dvnr (digital video noise reduction) applied that ghosts the image, looses fine detail, and generally reduces the overall quality.

I would suggest learning how to use the g-force avisynth script as, if nothing else, it will teach you a bit about avisynth and get you started in the right direction. It stabalizes the horizontal image shake in the movies, something that would be near impossible to do manually. Plus you could potentially apply it to other DVD's. But it's up to you after all, and nothing you do will make it any worse than it was to begin with (rofl)!

[ Scanning stuff since 2015 ]

Author
Time

RU.08 said:

It stabalizes the horizontal image shake in the movies, something that would be near impossible to do manually.

No, that can be done in Final Cut.  I know 'cause I've done it before (not for the GOUT, but for other projects).

And removing the DVNR is impossible.  The only way to get a DVNR-free OOT is to use the JSC laserdisc.  Even the 97 SE television broadcasts have really bad DVNR smearing.

Author
Time

Removing the dvnr is only possible with replacement, yes I agree. Has anyone done a good transfer of the JSC yet (which could be used to drop-in replacements)?

[ Scanning stuff since 2015 ]

Author
Time
 (Edited)

As I am going through the gout DVDs a few question come to mind.  I am noticing a lot of pull down issues.  When ripping the DVDs the properties of the mpeg2 are 720x480 23.98fps.  

My question is this, where the DVDs created from the the same master that created the laserdisc, or were the DVDs actually created from the laserdisc?

Were laserdisc players able to handle 23.98 content, or was the laserdisc a 29.97 (25pal) framerate?

The gout DVDs may have already gone through two frame rate conversions, first going to the laserdisc, and second going to the DVD.  This surely will cause many of the motion problems we are seeing.  Not to mention the technology to do the conversion was no where near as good in 93 as it is now.


Author
Time

rmclain, the framerate on the dvd's is 29.97. Somewhere during the rip process you have applied an inverse telecine, reducing the framerate to 23.976.  The gout dvd's were created from the same 35mm film print which was used for the definitive collection laserdiscs (as far as i know).

The film was scanned via a telecine machine at 24fps because that is what film is captured at.  On the Definitive collection laserdiscs the frames are stored at 24 fps and a pulldown flag on the disc signals the player to apply a pulldown, making the output to the tv 29.97 fps. On the faces laserdiscs the pulldown is already encoded onto the disc and the player just assumes that it is 29.97 fps.

On the Gout DVD's the film was rescanned and digitally transferred to DVD at 29.97 fps with the pulldown applied already (as far as i know).

The motion problems are from the DVNR which was applied to the 35mm film print before it was ever transferred to laserdisc or DVD.

Luke threw twice…maybe.

Author
Time

althor1138 said:

rmclain, the framerate on the dvd's is 29.97. Somewhere during the rip process you have applied an inverse telecine, reducing the framerate to 23.976.  The gout dvd's were created from the same 35mm film print which was used for the definitive collection laserdiscs (as far as i know).

 

Strange, here is the stream info that is being read off on my 2008 DVD.  Mind you this is reading the DVD info only, no processing has yet to be applied.  You can apply any type of telecine inverse at this point in the program.

 

 

http://img859.imageshack.us/img859/8541/screenshot20110508at406.png[/IMG]

 

Here is another shot of a DVD I created for a client that I know is 29.97.  Again this is just the raw steam info opened in streamclip the exact same way.

 

 

http://img857.imageshack.us/img857/7627/screenshot20110508at412.png[/IMG]

Now I am not saying that what you are saying to be untrue, but strange that streamclip reads the raw stream as 23.98.  

 

Author
Time

Yes you are right.  The dvd contains the progressive frames and pulldown flags are applied to make the framerate appear on the television as 29.97 fps just like the CAV Laserdiscs.  At least this is how I interpret it.

The important thing to know is that the DVD's, Definitive collection laserdiscs, and the faces laserdiscs all originate from the same 35mm film which  had the DVNR applied before it was struck to film(as far as I know). This 35mm print of the gout films was blurred before it was ever transferred to laserdisc or DVD.

Luke threw twice…maybe.

Author
Time

That's incorrect, the DVNR is not on the film they scanned.  It was applied as a temporal filter during the mastering process, so there are presumably no digital copies without the smearing, but there's no way it would be on the film print; that doesn't make any sense.

Author
Time

The GOUT DVD uses the same master as the 1993 Definitive Collection laserdiscs, but it's not a laserdisc transfer.

And the DVNR smearing is not present on the 35mm film (that makes absolutely no sense), but it's inherent in the tape master, so there's absolutely no way to get rid of it, even by going back to the master tapes.

Author
Time

I stand corrected lol. I'm not an expert on the transfer of film to home video by any means.

Luke threw twice…maybe.

Author
Time

I actually do work in the field of post production and I handle and deal with downconversions, upconversions, pulldown/frame rate conversions all day long.  Most of what I master is for Broadcast, Blu Ray, and digital prints.  If workflow in the 90s is what it is today, then the film would be first converted to a digital tape format before laserdisc mastering would happen.  At the time I am guessing it would have been a D1, or a D2.  Seeing as how those are SD only formats I am almost positive that those tape did not support a 23.98 frame rate in 1993.  I would almost bet everything I had that it went to a digital tape at 29.97.   At some point this was converted back to 23.98, which is something that you never want to do.

At any rate we all agree the gout looks like crap.

Any feedback on my pics above on the color would be appreciated.

 

Author
Time

It's true that it was a 29.97 NTSC tape master (D1, I think), but you do want to IVTC back to 23.976 before you author it onto a DVD, so the DVD format can do the pulldown itself, while those with progressive displays won't have interlacing issues.

Problem is, when they authored the GOUT, they didn't IVTC quite properly.  So you'll need to go through and fix the interlacing errors caused by this.

And the color is a good start, but it looks a little green in a few shots, and a little yellow in others (the Ben/Vader duel shot stands out to me as being too yellow).

Author
Time

Can you give an example of the IVTC errors you are seeing on the GOUT disc? I noticed some flickering on R2D2 in a couple of places, which I think is an IVTC problem, but I've never seen any combing artefacts.

Guidelines for post content and general behaviour: read announcement here

Max. allowable image sizes in signatures: reminder here

Author
Time

Yeah that's right a D1 tape, which supports PAL or NTSC resolution component video (480i/567i), the telecine was most likely done directly when the film was scanned. In fact I would say this would be almost a sure and certain fact, since if the scanner hadn't scanned each field seperatly then those awful jaggies in the picture wouldn't be there. You're quite right you don't normally want to be in a hurry to put them back into frames.

[ Scanning stuff since 2015 ]

Author
Time

Moth3r said:


Can you give an example of the IVTC errors you are seeing on the GOUT disc? I noticed some flickering on R2D2 in a couple of places, which I think is an IVTC problem, but I've never seen any combing artefacts.

Post 46
141781 thru 141784 are combed in the NTSC ROTJ (also happens to be the beginning of a reel)

However, in practice you must take into account the “fuckwit factor”. Just talk to Darth Mallwalker…
-Moth3r

Author
Time
 (Edited)

Darth Mallwalker said:

 

Moth3r said:


Can you give an example of the IVTC errors you are seeing on the GOUT disc? I noticed some flickering on R2D2 in a couple of places, which I think is an IVTC problem, but I've never seen any combing artefacts.

Post 46
141781 thru 141784 are combed in the NTSC ROTJ (also happens to be the beginning of a reel)

 

These are perfect examples.  It also makes sense that the pal is not showing these as it has less information to add from a 23.98 source, and less to remove from a 29.97 when going back to 23.98.

Now a days when we use a machine called an Alchemist to do our 23.98 to 29.97 conversions there are several options on how to handle the fields.  One of which is blurring.  Some clients would rather have this than the jagged lines that tend to appear on scenes with very fast motion.  I notice it most in peoples hands as the move them around.  You will get almost a doubled blurred effect.  Its not as harsh as the jagged lines, but not any better.  

This can be seen a lot throughout the gout version.  I will try and post some pics of it later.  Seeing as how this has gone through two conversions (one up to 29.97 to make the D1, and then back down to 23.98 for the DVD) its no wonder its there.

Author
Time

I've seen dvd's mastered at 29.97 fps (interlaced) that were from a film > pal then field-blend to 23.976fps! The jaggies (in the gout) are there because the fields were scanned individually (i'm just about certain)... (this was perfectly normal in 1993), but the DVNR is totally inexcusable. The very few flaws in the film scan that it covered up was no where near the inexhaustiveproblems that it created... probably Lucas thought it looked better like that?

[ Scanning stuff since 2015 ]

Author
Time

I doubt he even looked at it or listened to it back then.

As for differences between the versions of the GOUT, I don't really understand them, because the PAL version did not use its own source; it's just converted from the NTSC.  This is particularly puzzling when we consider that the PAL RotJ is considerably more detailed than the NTSC, but the other two are not.

Author
Time

I'm no expert, prob rmclain73 can answer but in 1993 ivtc would have been difficuilt (and expensive), plus if the same settings were used for all three movies (likely) what it means is that they found the NTSC tapes for sw and esb, and probably couldn't find rotj and so used the PAL tape... OR... they found all three ntsc D1 tapes plus the PAL rotj tape. I doubt the reason for the difference is any more complicated than that... but I'm known to be wrong ;).

[ Scanning stuff since 2015 ]

Author
Time

RU.08, I was graduating high school in 93 so I am not 100% sure what the process or the cost was like then.

 

Getting back to my project, I posted some still above of some early color test that I would really like some more feedback on.  I am really torn on the sunset scene as to how to go.  I like it more on the warm side, but it seams a little red?  Having a tough time balancing the color of the sky and not turning Lukes clothes yellow.  I know I also want to take some of the yellow out on the end medal award scene. 

 I am rendering out as we speak a cut with some detail filtering added.  The early test looked really good.  I will be posting a video later this evening.