hairy_hen

Argh, I'm really sorry to keep delaying and making people wonder what's going on.  A combination of being kept busy with schoolwork, internet connection problems, and going through something that could soon turn into a nasty breakup (and would leave me without a place to live) are contributing factors.

Personally, I blame Zoidberg.

The audio tracks (yes, they are all 100% lossless) are ready to go as soon as they can be uploaded successfully.  For some reason I haven't been having much luck with that, but I'll get those blasted things up there if it kills me.  The 70mm has been optimized to the point where I'm probably not going to touch it again unless a true discrete channel copy became available somehow; I also have the mono mix with the film version of the Fox fanfare spliced in, and the stereo mix from the digital source with all the dropouts and other errors completely eliminated by patching in parts of the analogue version.  Both are level-matched as precisely as possible to the 70 in order to maintain consistency of dialogue when switching between tracks—rather than relying on DialNorm to do this, I rendered new files myself after careful listening tests, to maintain the highest quality.  I'll also add my isolated score, which in addition to synching the music to the movie has also been heavily re-equalized to sound like it does in the movie, and not shrill and nasty as it does on the CD's.  Maybe I should start a new thread for all these things so they can be found in one place.

So, that's where I am, anyway.  Sorry for making everyone wonder . . .

I totally got the hint, but thought someone else would have said something about it before.  ;)

Yes, it was.  What I'm talking about now is essentially the same thing, but since I recently gained the ability to upmix using real Dolby Prologic II, the channel separation and imaging are now somewhat improved compared to how it was previously.  I'll also eventually do the other two movies the same way, as well.

Yup, I absolutely will be making my stuff available for this project.  I tend to post in Harmy's thread simply because it is the most active, but that does not by any means imply that what I'm doing is exclusive to his projects or anything like that.  The goal here is to present the movies in the best possible quality, and anybody who shares that intention is welcome to make use my stuff.

Speaking of that, d_j, what program do you use for Bluray authoring?  The reason I ask is that I'd prefer to use Dolby TrueHD for reasons of playback compatibility, but it is more complicated to use correctly on the authoring side, so it would help to know what your capabilities are in this regard.

Okay, seems like I always get stuck without much time to use the computer right when I went to get things done, but I'm about to encode some test files on the lossless tracks to see what works best for Harmy's upcoming efforts in Bluray authoring.

The audio tracks are basically done.  I want to test them some more to make sure there aren't any problems, but there isn't really any actual work left to be done on them at this point, assuming I did everything correctly.

Harmy, question: when you tested importing TrueHD into your Bluray authoring software, how did it want this to be done?  Did it ask for both the MLP and AC3 files separately, or were they still interleaved when you took them off the source disc?  As far as I can make out, when making an MKV file it's okay if the two are on different tracks, but in Bluray format they have to be muxed together, so hopefully the program is capable of combining them, since the Dolby encoder will only provide them as separate files.

The reason for them being combined is for the sake of backwards compatibility: with an HDMI or multichannel analogue connection the lossless audio will be used, while bitstreaming over S/PDIF uses the 'hidden' Dolby Digital track automatically, ensuring that no one will be left without the ability to hear the program.  In this way it somewhat resembles the DTS 'core' scheme; the difference being that in Dolby's version the two tracks are unrelated and not dependent on each other.  With only one output file instead of two, HD MA is no doubt generally easier to use from an authoring perspective, but I do think that TrueHD will be worth it since there is more Dolby-capable equipment out there.  (I'm able to encode in both formats, so if I'm wrong and it doesn't work for some reason, DTS will be provided.)

Darth Real Life requires that I go out of town for the next few days, but once I get back I should be all set to encode and upload everything.  This isn't a frivolous journey—it's a conference of audio professionals which is being hosted by one of my teachers, who owns a recording studio and in every way is an extremely knowledgeable and awesome guy.  I learned so much stuff from him this past year, and now to get school credit for meeting with his friends in the industry and hearing their views on audio . . . well, for someone like me that's a very exciting prospect.  ;)

For 720p, a BD50 would be a waste of space, resulting only in an unnecessarily huge download.  As I understand it, with highly efficient video codecs like h264 you eventually get to the point of diminishing returns as the bitrate goes higher, yielding ever-larger file sizes with minimal visible improvement.

Along a similar vein, I could certainly put out my audio tracks in 24-bit resolution if I really felt like it: after all, Pro Tools does calculate all its results in 32-bit floating point for the sake of accuracy.  However, since I'm dealing with laserdisc sources that were 16-bit from the start, there isn't actually any more detail in them than this to begin with.  As long dither is properly applied, any quality difference when converting them back will be quite miniscule.  And since lossless compression schemes seem to work more efficiently on 16-bit material anyway, the choice is clear.

Now, if there were true 24-bit copies of the soundtracks to begin with, or if the video sources were sharp enough to justify 1080p, then releasing them at these quality levels would be worthwhile.  But given the sources that are actually available, playing the numbers game isn't going to create any additional benefit.

Depending how many tracks are included and how much disc space is available, using redundant copies in Dolby and DTS together may be impractical.  I've done a bit of testing and have found that TrueHD gives a significantly smaller file size on its own, since it contains lossless audio only, where as DTS-HD MA includes a lossy core as part of the same file; but since the Bluray format requires TrueHD to have an accompanying AC3 track as well, the total ends up being about the same.  That is to say, the total file size of TrueHD + AC3 ends up being approximately equal to DTS core + extension, depending what lossy bitrates are used.

Therefore, using both DTS-HD MA and AC3 together would require significantly more disc space.  Whether this would be a problem or not remains to be seen.

Since not as many people are able to hear DTS properly, due to their equipment or software not fully supporting it, my general inclination is to say that Dolby is the way to go.  The only thing that could potentially be an issue is that not all Bluray authoring programs support TrueHD in the same way.  It isn't a problem with the audio itself, but rather with how the authoring software handles it: the TrueHD and AC3 streams are supposed to be interleaved with each other for compatibility with the Bluray format, but the Dolby encoder creates them as separate files, so results may vary.  It looks like tsMuxer used to have a problem with this, but it may have been fixed.

I will continue to investigate these matters, and report back on my findings.

It will only take me a couple days to completely finish with all the new audio tracks, if that factors into it . . .

As a complete list, this would include the 70mm version newly upmixed with Prologic II, the digital copy of the stereo mix with the dropout errors removed, the mono mix with the proper fanfare at the beginning rather than the CD copy it curently has, and as a bonus my isolated score.

I haven't decided for sure, but at the moment I'm leaning towards encoding them as DTS-HD MA rather than TrueHD, since while Dolby files seem to have greater playback compatibility, they are apparently more complicated to use correctly in Bluray authoring.  But I think this should be investigated more thoroughly before I make a final decision on the matter.

Regarding the issues with The Lord of the Rings—anybody who is upset about this could always try reading the books.  Not only are they 100% faithful to Tolkien's artistic vision, I hear the colour timing on them is fantastic.  ;)

I haven't been following this thread before, and only just read it through in its entirety.  While I don't wish to rain on the parade, it should perhaps be tempered with some healthy scepticism.  I'm not convinced that it's entirely worthwhile to go to the trouble of working with cinema DTS discs unless it is absolutely necessary.

To use The Matrix as an example: there has been talk of the Bluray soundtrack being inferior to previous versions, and I have to say I'm baffled as to where this claim comes from.  After having heard it numerous times on the old DVD, I found not even the slightest hint that anything on the Bluray had been altered.  In every way it was exactly the same mix—same balance, same dynamics, same everything.

Now admittedly I don't yet have an HDMI receiver and therefore can only listen to the 640 kbps AC3 track, rather than the Dolby TrueHD.  But it would astonish me greatly if the two actually turned out to be any different from each other.  If there is a difference, it may just as easily turn out to be something happening during playback, such as an unwanted dynamic range compression mode being activated, or a difference in DialNorm values causing the overall volume to come out lower.

When comparing soundtracks in this manner, you must make absolutely sure that they are completely level-matched, or any comparison of their relative dynamics will be rendered meaningless.  Do not ever believe that you are immune to being fooled by differences in playback volume, for even experienced professionals can be tricked into believing they hear superior sound quality when in fact all that has happened is that something was made louder.  It is shockingly easy to fall into this trap, and I believe that a lot of what people think they hear is actually only this kind of placebo effect.  This particularly applies to bass content because of the Fletcher-Munson curves, whereby our hearing is particularly insensitive to low frequencies, and consequently is very susceptible to a false perception of level differences if playback volumes are not equal.

To give a textbook example of this, consider The Phantom Menace.  I have heard repeatedly about how much better the movie sounds on laserdisc than on DVD, but this is actually false.  This web page measures the output levels of both, and determines that the two versions are virtually identical to each other in dynamic range.  The only difference is that the laserdisc version is mastered 8 dB louder than the DVD.  That's a huge discrepancy: of course one is going to sound more powerful than the other!  But if you turn up the volume by 8 dB to compensate when playing the DVD version, this will be entirely cancelled out.

The recent Batman films have also been mentioned here, but again I find no difference whatsoever in how they sound.  Last summer I attended a triple feature screening of all three at the Cine Capri, the best movie theatre in Phoenix, and what I heard there was identical to what is on the DVD and Bluray releases.  The total effect was considerably more powerful, of course, due to the huge auditorium and the absolutely killer Dolby setup they've got in there (even more so now that they've installed the new Atmos system), but the mixes themselves haven't been changed at all.

So, while I absolutely applaud the goal of preserving theatrical mixes and making them available, it is important first to verify that there is actually a need to do so in the first place.  There are plenty of films whose soundtracks have indeed been badly mangled on home video (the site I linked to also has another page that conclusively demonstrates how much better the original mix of Jurassic Park is, for one), and it would be best to concentrate such efforts on theatrical mixes that truly are unavailble through any other means.  And don't be tricked into thinking that the delivery format somehow results in better sound, either—all that really matters is the mix itself, and whether it is packaged as Dolby or DTS is largely irrelevant providing everything else is the same.  The AC3 codec is more efficient than DTS, able to achieve similar sound quality at only half the bandwidth, so comparing the numbers doesn't really mean much on its own.

I haven't posted much recently, due to being busy with various things, but I haven't forgotten.  ;)

Actually I've had a rather exciting new development, in that I now have access to official Dolby software, which means I can upmix with real Prologic II and record the result directly into Pro Tools.  (This is something I've been wanting to do for a long time, but only now have finally been able to make it happen.)  I'm currently running tests to determine how much of an improvement this would be over the existing 5.1 mix; if it's to my liking, then a new version will soon be on its way.

The main potential benefit of this would be improved separation—less leakage of sound from one channel to another.  There have been occasional comments about it being an issue, so this should help on that account.  I may edit the LFE track ever so slightly, but on the whole it will still be virtually identical to the previous mix.

I'm also able to make use of the official AC3 encoder.  When I tested it in comparison to the freeware version I'd been using before, the improvement in quality was immediately apparent.  I've learned that this is because there are various coding tricks Dolby came up with to improve efficiency, resulting in superior sound at the same bitrates, which unofficial encoders do not duplicate even though they produce output compatible with the AC3 spec.  The improvement over the previous version is most obvious for 5.1 at 448 kbps (which is most significant for the DVD), but every track will certainly benefit from this, so I consider it highly necessary.

On the lossless front, I can encode in both Dolby TrueHD and DTS-HD MA, so I suppose it comes down to whichever one has greater compatibility.  I know that VLC can play any Dolby file but not DTS, due to licensing issues or some such, but I don't know what other people use to watch HD files on their computers, or what works best with mkv in general.

My isolated score for SW is now completely finished and has been uploaded.  I'm not sure what the forum rules have to say about publicly posting download links for such a thing, given that the music comes from a retail CD release that can be easily obtained, so I'll err on the side of caution until I've heard that it's okay to do so.  If it matters, in addition to the synch and edit work, the extensive EQ applied to the tracks has significantly altered the tonal balance of the music from how it is on the official release, making it sound much more like it does in the actual movie.  So the final result has relatively little resemblance to what is commercially available.

In addition to the isolated score, I also made unedited versions of the equalized tracks so that the music can be listened to this way on its own without the video.  These were processed at 24-bit, 88.2 khz before converting back to 16/44.1, which theoretically gives even greater accuracy in rendering (especially in the high frequencies, which have undergone the most alteration).  Since I was dealing with previously mastered material, I decided to use a standard triangular dither instead of any noise-shaped variety, to keep things simpler.  The iZotope resampler was used in all cases.

I have to say I'm rather pleased with how this has turned out, for though the flaws of the source cannot be fully overcome, the EQ change has extracted a much improved sound quality from these CD's.  When listening to them, it is as though a veil has been lifted and something closer to what Eric Tomlinson actually recorded is allowed to come through.  On the isolated score, the level has been lowered by 6 dB to match what is actually heard in the film, so there will be no significant volume difference when switching tracks on a custom DVD or Bluray; while the other version was lowered only slightly to prevent possible clipping.  On both, the cantina band tracks are further reduced to a level appropriate for background music, rather than blasting away at high volume as they do on the CD; and I also gave them some reverb in order to present a sense of ambience, avoiding the overly dry sound of the direct recordings.

Since there are other isolated scores out there, I should say that my version is not meant to be in competition with any of them—it's just my take on what the music of Star Wars is supposed to sound like.  It is as much for my sake as anybody else, and was made in part to practice my audio editing skills with material I enjoy listening to.  But if other people obtain my work and like the way it sounds, then I'm very glad to have made that possible.  ;)

Moth3r said:

Excellent feature Harmy, really good to highlight all the problems with the official releases.

Would be nice to see Hairy Hen do something about the restoration of the 70mm mix.

Also, does everyone pronounce AVISynth as "avvysynth" or is it just you? ;-)

It is indeed an excellent video, and a great way to show the uninitiated what some of the problems are, and what a gargantuan task it was fix them by combining all those different sources together.

If there is interest in something similar for the audio side of things, I'll see what I can come up with.

Mike Verta has talked about this kind of thing before.  According to him, scanning Technicolor prints is possible, but getting any kind of usable result requires the latest and greatest in scanning equipment, and that sure as hell ain't cheap.  The high contrast of the prints is such that duplicating them using traditional methods is impossible.

As far as colour is concerned, provided they are stored properly they will never fade, and therefore make excellent references.  The dye matrix master was made directly from the negative and thus the Tech prints are one generation closer to the source than regular 35mm, potentially yielding higher detail (though they're still fairly soft—the high contrast makes them look somewhat sharper than they actually are).  Because of the way they were made, no two have exactly the same colour balance, although they are pretty close to each other on the whole.  Extreme deviations may occasionally appear on a few of the prints, but these are due to human error in the lab and should not be considered representative of the film's true look.  There is a possibility that the colour timing isn't exactly the same as it was for standard 35mm anyway, since going from the negative and not the existing interpositive means it would have had to be done over again, but presumably they got it close enough not to worry overmuch about this.

The different colour balance inherent in 70's projection bulbs means that putting it through a neutral light source will yield inaccuracy.  This must be compensated for to get a meaningful result, or else the image won't show enough of the warmth it was intended to have.

And yes, unfortunately they all seem to have had their first reel hacked off and replaced with Eastman versions containing the Episode IV crawl, so there just isn't as good a version out there of the film's beginning as there ought to be.

My settings are specific to the Pro Tools EQ3 7-band parametric equalizer plugin, and different equalizers each seem to measure their adjustments somewhat differently, so I'm not sure how well it would translate, particularly since a parametric works in quite a different way than a graphic equalizer.  Much is dependent on the 'Q' setting, which is the width of the frequency range being affected, and how the bands may overlap and interact with each other.  I'm sure it's possible to come fairly close to what I did, though, so I'll look into that.  But it might turn out to be easier just to put out a 44.1 khz version in addition to the isolated score.

Last night I found that my EQ didn't quite work for the sunset cue as well as it does for other tracks, for in comparison to the film version the strings seem a bit too muted and not sweet enough, while the bass (which unlike the higher frequencies has been left untouched) takes on a bit too much prominence.  So I'm going to come up with another setting specifically to match that cue to the movie, and then I'll have both to choose from for the rest, depending which sounds best on each track.  The main title was so shrill on the CD that I had to use fairly extreme settings for it to sound right, but other cues may benefit from a more balanced approach.

For the distortion that sometimes shows up, I don't think this is caused by clipping.  The dynamic compression they used is a form of peak-limiting which ensures that the waveform rarely if ever reaches digital maximum, usually stopping a few tenths of a dB under that.  Also, the distortion can sometimes be heard even when the level is somewhat lower, so I suspect that it either has to do with flaws of the source tape or the process used to transfer it.  The EQ change seems to make it slightly less noticeable, but it's definitely still there.  I may investigate possible methods of reducing it further, but I doubt it can really be fixed completely.

For anybody who's interested, I'm making an isolated score for the first movie using the 2-CD set.  I'd been thinking of doing this for a while, but now that I'm more comfortable with using Pro Tools for editing and mixing purposes, I decided to go ahead with it.

The 2-CD release is the only complete presentation of the score and therefore must be used for any project of this type, but unfortunately it sounds shrill and harsh, due to the broadly excessive elevation of the high frequencies compared to how the music actually sounds in the movie.  To combat this, I spent some time dialing in an EQ curve that would more closely match it to the 1993 mix, which comes from the original 70mm version and should be considered the most accurate tonal reference.  The result is hardly exact, of course, considering how different the two sources are to begin with, but I was surprised by how close the final result actually comes.  (The CD is also somewhat peak-limited and shows occasional distortion, but there isn't anything I can do about that.)

Before starting I read a lot about different equalizers in an attempt to find that mythical perfect tool that would give the best sound and turn all my work to gold, but eventually realized there was nothing any of them could do that the default parametric EQ in Pro Tools couldn't match, and with considerably greater flexibility and ease of use.  After all it isn't the tools that matter so much as the technique, and what is appropriate for the task at hand.  Now obviously if I had a $10,000 analogue unit like Steve Hoffman, I'd be using that instead of any digital imitation, but all that really matters is that the process is transparent and doesn't colour the sound for the worse.  At any rate I'm following the advice of he and other respected audio engineers in using equalization only for cutting frequencies away from a sound—think of it as 'sculpting', if you like—and not for boosting.

Conversion of the audio from 16/44.1 to 24/48 was performed with the iZotope SRC before importing, and Pro Tools handles its internal calculations at 32-bit floating point resolution.  The final result will be converted back to 16 bits with iZotope's MBit+ dither.  Since the source is a CD, there isn't actually any more detail than this to begin with; processing at higher resolution simply gives greater precision in rendering, and dithering allows for high quality to be retained when reducing bit depth.

The broad nature of the treble reduction means that the tape hiss is unavoidably lowered along with it, so to compensate I used a signal generator (along with another EQ) to add low level white noise back into the music, in order to maintain the proper analogue vibe.  (This can be considered the aural equivalent of the simulated film grain added to the effects shots in the Despecialized Edition to replicate the look of the optical compositing.)  A noise gate allows the hiss to trigger with the start of each track and fade away at the end.  Since the entire set shows equal shrillness, the same EQ settings should work well for every cue.

In synching up the tracks, I was surprised by how many edits there are in the music that I'd never noticed while watching the movie.  Matching them requires patience and attention to detail, but I haven't attempted to replicate every cut exactly.  A film soundtrack can sometimes get away with clunky editing, since the dialogue and sound effects can be used to obscure the transitions, but in a music-only presentation I think it important that the edits make sense from a listening perspective.  Therefore I am attempting to allow each track to sound continuous and not 'cut up', where possible.  Occasionally the synch may be off by a small amount, but hopefully this will be unnoticeable.

Now that I've got the music sounding just the way it should (within the limitations of the admittedly flawed source, that is), finishing is just a matter of synching and editing the remaining tracks.  Since I've got my finals coming, this will probably take me longer than it otherwise would, but I'll have it ready as soon as I can.  For now, here's a preview in the form of a comparison between three versions of the main title, as heard in the 70mm mix, the harsh and unpleasant CD, and my EQ'd version.  All three have been level-matched to remove any bias based on their perceived loudness; the file is an mp3 so that the download size isn't too large, but encoded with maximum bitrate/quality settings.  The actual release will of course be lossless.

http://www.sendspace.com/file/ong3z9

While information about the '85 mix is hard to come by, certain qualities of its sound can lend themselves to informed speculation about its origins.

1) The '85 mix has full stereo imaging compared to the '77, which has been panned inwards.

2) The '85 mix shows somewhat improved fidelity compared to the '77, most noticeably in its lesser amount of tape hiss.

3) The '85 mix lacks the pronounced music reverb heard in the '77 during the quad laser battle and the end credits.

4) The '85 mix has been dynamically compressed to a greater degree than the '77.

5) With the exception of C3PO's tractor beam line, the '85 is identical in content to the '77.  Nonetheless, certain sound effects occasionally 'feel' slightly different, perhaps due to the use of different compression and/or EQ during mastering.

Now let us consider a few other facts:

1) The 70mm version had a few additional sound effects not heard in the 35mm stereo mix, which also do not appear in the '85 version.

2) The '97 SE mix, which is said to have been derived mainly from an original 4-track master, does not contain these 70mm additions either.

3) The stereo imaging of the '85 mix seems to line up exactly with what is heard in the 70mm and '97 mixes (apart from the changes, obviously).

4) The 70mm and '97 versions do not contain the additional music reverb.

What may we deduce from this evidence?

1) Despite their overall similarity, the '85 mix could not possibly have been derived from the original 35mm stereo.  The missing reverb alone would tell us this much; and the imaging discrepancy lends it further credence, for while there are ways of widening a stereo field, it is unlikely that artificial processing after the fact would yield a result that aligns so closely with sources that had not been narrowed to begin with.

2) The '85 mix must therefore have been derived from an earlier generation source that did not contain the reverb or any added sound effects, and had the full panning width.

3) The similarity of the '85 and '77 stereo versions, in spite of the differences they do display, is such that they must have a common ancestor in their background.

When we add up the facts, it seems to me that the most logical conclusion would be to say that the 1985 stereo mix is derived from the original 4-track master.  It is a fresh downmix of an early, discrete channel source, with the tractor beam line added on top and the dynamics compressed down to a level that was deemed acceptable for home video release.  This explanation accounts both for the similarities and differences between versions and makes the most sense given the audible properties of each, as well as fitting with the information available to us about these matters.

The 70mm mix hasn't been officially released in its original form, but the 5.1 version included here is a close recreation of it, which I assembled using various laserdisc sources and a lot of very careful listening.  While it doesn't sound exactly the same as the real thing, it is very close, and it would be pretty difficult to make a more accurate version without access to the actual master.

So in effect, the 5.1 mix can be considered to be the audio equivalent of the Despecialized Edition's video work.  ;)

Further tweaking to any of the 5.1 tracks will be minimal, and would be done as much for the sake of my learning to work with professional software as any other reason.  ESB and RotJ may benefit somewhat from the inclusion of lossless LFE from the 2011 Blurays, but even that won't make a huge difference in sound quality: much more of an improvement could be heard if the user upgrades to a hifi subwoofer and plays the current lossy version over that.  ;)

Mavimao said:

he didn't use any special voodoo, he simply just spliced in the 85 mix in those places

Ah, that makes sense.  I thought that must have been it, but couldn't remember exactly.  The only thing that would stop it from being a seamless transition is that the '77 version has the channels panned inwards by a fair amount (presumably this was to reduce crosstalk in the surround, since upmixing processors were less advanced in those days), while the '85 has the full stereo width, but presumably he was able to match this by ear using the pan controls on each channel.

I just had a big test today, but now that it's done with (I already know I aced it!), I'll be getting started on this very soon.  ;)

As far as lossless tracks for the Bluray are concerned, if there only ends up being enough disc space for the three theatricals, that will be fine, since the '85 and '93 sound very similar to the stereo and 70mm versions anyway for the most part.  High bitrate AC3 tracks should be sufficient for them in that case.

I haven't actually got Pro Tools up and running yet, because I only just got it in the mail and my student-discounted ID still needs to be verified.  I've been starting to learn it this semester and I agree that it is kind of an ungainly and cumbersome program—my folks got it for me so I could do assignments at home and so that any future career work won't be out of my reach due to not being familiar with the standard software used by many professionals.  All my Star Wars stuff so far has been done using a Mac program called Sound Studio, which doesn't have as many features or controls, yet has a more understandable interface and far greater overall ease of use.  But if I make myself use Pro Tools for SW stuff from now on, it'll be a good way to learn.

In addition to these little fixes on 35mm stereo track for the first film, I'm still planning on revisiting the 5.1 mixes for ESB and RotJ, so I'll see I fare on those as well.  ;)

@Harmy: It shouldn't really be that much work, but it is a bit tedious and of course I still have studying to do also.  I'll try to be as quick about it as I can.

I downloaded (and am currently seeding) the 2.1 AVCHD at the spleen and it looks awesome.  Practically every tiny little thing you could think of that wasn't quite right before has been taken care of, and the movie has never come this close to visual perfection, especially in terms of colour timing.  I salute you, Harmy, for doing such amazing work, because it most definitely paid off in the best possible way.  :)

It's just as well if there's going to be yet another release, though, because there's something I want to do to for the audio as well.  For the sake of my academic career in sound I recently obtained a new Macbook Pro in order to be able to run Pro Tools, the professional recording software, and this means I'm now able to make use of the official DTS-HD MA encoder suite as well.  The AC3 copy of the original '77 stereo mix which I provided for the Despecialized Edition was made from the analogue capture, because the digital version has numerous defects such as audio dropouts and pitch distortion which have gone uncorrected.  The source for the analogue capture reportedly contains these same defects, but since Belbucus had already corrected them all, the simplest solution was to use his version instead, since to my ear the errors make for an unacceptable listening experience.

However, when I went to encode the various mixes of the movie in DTS-HD MA for the Bluray release, I decided to listen again to the digital version of the stereo mix and compare it in fine detail to the analogue capture.  As expected, the digital does show a slight increase in overall fidelity, but my main discovery was far more significant and startling: the analogue capture has actually been dynamically compressed to a great extent in certain scenes.

What the reasoning for this is I don't know, though perhaps it has something to do with available headroom on analogue laserdisc audio compared to 35mm Dolby Stereo on film.  Other parts of the tracks seem virtually identical, but there are numerous places in the film where the levels have been brought down significantly, and this leaves them sounding underpowered in comparison.  Without even closer inspection, I can't say for sure whether the track has been put through a compressor to reduce the peaks or whether someone simply sat there with a fader and manually reduced the gain of the whole mix when they felt the levels were getting too high; but I'm starting to suspect it was the latter, since in one scene I can actually hear the level go back up by a large amount after first being much quieter—this happens in an obvious and obtrusive way and definitely isn't supposed to sound like that, so either way someone wasn't paying attention at the board!

What this all boils down to is that the dynamics of the digital version have clearly not been tampered with, and that it represents the 35mm stereo mix more accurately than the analogue version.  Because of that, I'm taking it upon myself to remove all the defects in order to provide the best possible copy.  Had I been aware of this discrepancy sooner, I would have done this a long time ago, but I had previously believed the two tracks to be identical.

Thus far in the digital version I have identified 13 audio dropouts, two instances of pitch distortion due to tape damage, and one spot in which the entire stereo image shifts radically to the left due to the right channel dropping in level before returning to what it should be.  I'm not nearly as experienced or capable as Belbucus when it comes to audio work, so left to my own devices I'd probably have been at a loss to correct these errors, but since he took care of them all in the analogue capture already, all I really have to do is splice in short sections of that version to cover them up (making sure, of course, to level-match the two for a seamless transition).

As soon as I have more time I'll be taking care of this, and then uploading DTS-HD MA encodes of the stereo, mono, and '93 mixes for use on the upcoming Bluray.  The '85 and 70mm mixes have been provided already, I believe.  There is a slight problem in that I'm not actually able to listen to the resulting files once they're encoded, but since I have the official DTS software and its verification capability, there shouldn't be any problems.

Hey guys: I haven't been around much lately, due to excessive real life business, but I'm definitely looking forward to the release of version 2.1.  I plan on providing optimized lossless audio tracks for all the original mixes to be included here, if Harmy wants them, so stay tuned.  ;)

In this case, 'not exactly the same' is just semantics.  Sure, there are a couple changes in sound effects, a missing music cue added back, some reversed stuff is now unswapped, and the third rear channel is discrete instead of matrixed; but these things don't really amount to much in the way of actual significance.  Like zombie said, the entire sound and 'feel' of the mix is vastly different than anything that was made in 1977, and even though the Bluray corrects some of the most egregious technical oversights, this is comparable (if you'll forgive the overdramatic example) to putting a bandaid on a major stab wound.  To have a version that sounds right, the only choices are either to do it over again, maintaining similarity to the original in all aspects; or better yet to simply use the real thing and be done with it.

In the course of my audio studies, it has become increasingly clear to me that changes in sound have to be made carefully and with great attention to detail, because the only way to undo them is to go back to the source and start over.  Things like equalization and dynamic range compression are impossible to reverse once they've been applied, because they fundamentally alter the tonal qualities of the recording.  My professors have emphatically emphasized on many occasions that it is far better to get the sources right from the start and have good quality material to work with, rather than using messed up tracks and trying to 'fix it in the mix' afterwards.  The official SW mix has some of the worst-sounding EQ that I've ever heard—together with the screwed up balance between elements, it has the effect of emphasizing distorted and boxy sounds to the exclusion of all else.  Even the most talented audio engineer in the world would have a tough time salvaging it into anything remotely listenable.

Fortunately, the original mixes have all been archived at extremely high quality, so the only thing stopping them from being heard is the resistance from those in charge.

pat man said:

Yes most of the changes was in 2004/2011, but can't they use the 2004/2011 files as a reference for the changes (Apart form the color, crush blacks) and audio form the 6.1 blu-ray (not from the 2004 DVD and the No! NOOOOOOOOOOOOO!!!!)?

The audio on the Bluray is exactly the same mix as the DVD version, aside from a few tweaks here and there.  If any part of that godawful, abominable mess was used for a release of the original version of the movie, it would be such a travesty that the term 'completely unacceptable' doesn't even begin to describe it.