logo Sign In

RETURN OF THE JEDI 1983 THEATRICAL VERSION RECONSTRUCTION DVD by Harmy (MKV, NTSC DVD5 AND PAL DVD9 AVAILABLE) — Page 9

Author
Time
 (Edited)

I PMed Chewtobacca to see whether it would be possible for him to change that shot without reencoding twice if I uploaded just that one corrected shot, we'll see what he says. If it's not possible, I'll just leave it as it is because as I've already said, I'm not reencoding and reuploading the whole thing again because of one vowel...

Author
Time
 (Edited)

If you upload the corrected sequence, I can do it with Avisynth and a re-encode, which I am doing anyway for the NTSC version.

I know it's picky, but I confess that I noticed Wookie too and things like this do irritate me -- no doubt more than they should.   Could people check for other subtitle errors, because if there are any we might as well fix them in one go? Also "captain Solo" needs to be "Captain Solo".

Author
Time
 (Edited)

OK, does that mean you'll have to re-encode twice? Once in Avisynth and then again in DVD Lab? But then you're probably exporting uncompressed AVI from Avisynth, aren't you? So that should cause very little quality loss.

Author
Time

The Avisynth work will be lossless.  Trust me.  There will still be just one re-encode.  :-)

Author
Time

Sounds good!  Leave a few frames on either end of the clips just to be on the safe side.  I'll cut them out when I splice the clips in.

Author
Time

Also, when you're adding Belbucus' audio, it isn't gout synced, so it needs exactly one second delay to fit the video for the PAL version. I think this delay can probably be applied to the NTSC as well but you may need to test it out.

Author
Time

Okay!  I didn't realize that.  The NTSC delay will probably be ever so slightly different then.  If someone can post the exact delay value, that would be good.

Author
Time

Well, actually the precise delay for the PAL version is 24 frames ;-)

Author
Time

OK, the clips are encoded and sent. Thanks, Chewtobacca, for your effort.

Author
Time

Harmy said:

OK, the clips are encoded and sent. Thanks, Chewtobacca, for your effort.

 A big thanks to Harmy and Chewtobacca for both their hard work.

Author
Time

Murry Sparkles said:

Harmy said:

OK, the clips are encoded and sent. Thanks, Chewtobacca, for your effort.

 A big thanks to Harmy and Chewtobacca for both their hard work.

I second Murry's thanks. I appreciate all your hard work, Harmy and Chewtobacca. It's good that there's finally a new reconstruction of the theatrical version of ROTJ. While ESB has been reconstructed by PaulisDead and Adywan, ROTJ really has only been reconstructed by OCPMovie and he didn't have access to the GOUT at the time. Pls. understand that I think OCPMovie's work was impressive but it is really nice to have this new version.

Author
Time
 (Edited)

 

Chewtobacca said:

If someone can post the exact delay value, that would be good.

 

Darth Mallwalker said:

Yes he did. satanika, I reckon
It's on usenet 544 days. 549 as I'm typing this
Search 'audio for GOUT projects' belbucus_delay_values.png
If that doesn't work, send me PM

Use that value if-and-only-if Harmy's STAR WARS card appear at frame 689, like NTSC GOUT.
(690th frame, since first frame is called 0 -- I'm going by memory here, so double-check)

If Harmy's title card comes something other than 689, I'm confident you'll get it sorted...

 

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

Author
Time

As I recall, the precise delay values for Belbucus' audio tracks were 1.017s for Star Wars, 0.948s for Empire, and 1.015s for Jedi.  This refers to the pcm tracks running at correct speed, and provided a complete synch with the ntsc gout audio.  To synch with the pal version, you'd probably want to add the proper amount of silence at the beginning of the track first, and then speed it up after that.

Assuming of course, as Darth Mallwalker points out, that the frame count and timing of this version is identical to the gout.

 

Now, about the 5.1 audio.  I made the edited version mainly for my own use because I like the way it sounds, but clearly an unaltered version would be ideal for projects such as this.  I'd have provided one already if I could, but my editing has always been done in stereo and then sent to Satanika for processing, since upmixers do not seem to work on my computer.  Accordingly, I've only ever had the final mixes as AC3 files, not the multichannel wavs, so unfortunately it's not a simple matter of editing the changes out again.  You could conceivably decode the AC3 to its component channels, put the unchanged sound back in, and then re-encode, but I don't recommend that at all because it will further degrade the sound quality by being twice lossy.  Really the only way to do it would be to make a new upmix of the unchanged RotJ 1993 mix and add in my LFE channel, then make a new 5.1 AC3 that way.  (The LFE is critical to it being 5.1, otherwise there's no difference from simply using Prologic II in the receiver on the stereo track.)

I reckon I'll have to see if Satanika would be able to lend his services again.  dark_jedi also expressed interest in having a non-edited version for his v3 gout project, so clearly there is some demand for it . . .

Author
Time
 (Edited)

Thanks for the information, hairy_hen!  I'll try my best with the PAL audio.  According to Harmy, this reconstruction sychnronizes with the PAL GOUT, which when slowed down apparently synchronizes with the NTSC GOUT, so the NTSC shouldn't be too much of a problem.

EDIT: Oh and thanks to Darth Mallwalker too!  I should have noticed that .png.

Author
Time
 (Edited)

Darth Mallwalker said:  Use that value if-and-only-if Harmy's STAR WARS card appear at frame 689, like NTSC GOUT.
(690th frame, since first frame is called 0 -- I'm going by memory here, so double-check)

The Star Wars card does appear at frame 689 as when previewing the Avisynth script in Virtualdubmod, so that should indeed be the 690th frame.  That's fortunate because it means that Belbecus' mix should synchronize perfectly.  The delay values that hairy_hen posted above match the newsgroup .png.  Well remembered, hairy_hen!  It looks like it should be plain sailing.

So far I've spliced in the footage with fixed subtitles and added new mattes.  I'm rendering out my AVI now.  Harmy, I'm uploading a small preliminary sample of the NTSC encoding so you can have a look.

EDIT:  Harmy, your Inbox is full.

Author
Time

I'm happy to hear the encoding has started :-)

My inbox is now cleaned.

Author
Time

Yeah,  I used 005's list, so the reconstruction should really be quite complete.

Author
Time

So let me get this right: the NTSC DVD5 has the corrected subtitle spelling, but the 6.4GB PAL MPEG-2 video file doesn't?

Guidelines for post content and general behaviour: read announcement here

Max. allowable image sizes in signatures: reminder here

Author
Time
 (Edited)

Yes, but I don't provide the links to the faulty MPEG-2 anymore, as there will be a PAL DVD9 with the corrected subtitles and the same quality. And possibly an MKV.

Also, I wanted to ask, would it be ok to create an x264 MKV with the correct speed (23,97 fps) and PAL resolution, or rather would such a file be playable in standalone media players that support MKV? It sure as hell would be playable on a computer and it seems the best solution for computer playback - highest SD resolution but correct speed.

Author
Time

Yeah, I think that could work with resolution 720x576 encoded at x264 and at 23.98fps. But then I am no expert.

For me personally, I will be going for the .mkv version as I would be playing it on my media player. 8)

Author
Time

720x576 MKV at the proper speed?!  I'd be all over that!