Harmy's THE EMPIRE STRIKES BACK Despecialized Edition HD - V2.0 - MKV & AVCHD (Released) — Page 111

JEDIT: Changed my mind.

The Lowry stuff definitely didn't happen until 2004.

The reason that shot in particular looks so bad in the official release is because it is one of those Harmy alluded to earlier: scanned into the computer using outdated 90's film scanning technology, composited into frame along with outdated 90's CGI, printed out to film, then rescanned into the computer again in 2004 and subjected to additional degraining for the Lowry process.  It has been through significantly more generation loss and destructive processing than a typical projection print, as exemplified by the -1 scan.

The whole movie doesn't look that bad, just the parts that have been through all the stages of crappy digital alteration.

It took awhile on TPB, but I finally finished the DVD9 DL. Great job Harmy and sources! It is mindblowingly awesome on 1080p upscale via a Sony BD player.

If infact the negatives of these films (and others for example Disney cartoons) are really all that worn out and faded, why wont the studios scan well preserved IB Technicolor prints? These would seem to be pretty much ideal- early generation and mostly true colors?

Of course this is known in general, but only recently while watching ESB V2.0 did I realized the snowspeeder harnesses have popped bubble-wrap as a material.  What an odd production choice.

Danfun128 said:

So how bad is the generation loss, DVNR, crushed blacks etc in the despecialized? Considering that the 2004/2011 release and the 1993 (GOUT) release are important sources...

 Well, the crushed blacks are a bit better, but no one has said this project was working from perfect sources, just the best available.  Really, what this comparison shows is how much more potential is out there if Disney/LFL would do a proper restoration of the OUT, rather than continuing with a master which we can now show specific examples of advanced generational losses from all the tinkering.

hairy_hen said:

I should also point out—and please treat this information seriously—that the 5.1 mix was designed for high performance sound systems and may not give ideal results on equipment that cannot cope with wide dynamic range material and high levels of bass.  It will also not sound its best if downmixed to stereo.  Obviously I have no control over how people listen to it, so it was necessary to do a bit of tweaking to the surround channels in order to make stereo playback at least sound acceptable (it would have been quite horrible if I hadn't done this, due to excessive comb filtering from out of phase signals).  But since the 35mm stereo mix is itself very good, I must recommend that anyone watching the movie with a 2-channel sound system choose that version instead, rather than downmixing the 5.1 mix.  Content-wise they are identical, and the greater midrange emphasis of its EQ makes it more suitable for the kinds of sound systems people are likely to have.

hairy_hen, any chance you have the 1980 5.1 DTS-HD-MA track without the high levels of bass?  I have a fairly serious home theater setup, and this audio is just too intense for me.  It would be great to have this audio track, but with bass levels that are similar to other films.  Thanks!

In all the time I've been working on these things, I think that's only the second time anyone has ever said I put too much bass into it!

I hate to ask, but . . . well, are you sure everything in your system is calibrated correctly?  It is not at all uncommon for the surround channel and subwoofer levels to be set too loud in home theatre setups.  I recently recalibrated my own system upon discovering I had not done it quite correctly the first time around, and hearing the proper balance between all the speakers is essential to presenting a mix the way it is intended to be heard.

The proper way to do this is to use an SPL meter to individually measure the output of each speaker at the listening position, and adjust the levels of each in the receiver until they all measure exactly the same.  Pink noise test tones, which have equal energy per octave, are usually used for this measurement.  The calibration is done so that the RMS level of the test tone should play back at 75 dB on the meter for each individual speaker.  (85 dB is the standard level for cinema calibration, but this tends to be unbearably loud in a smaller listening space.)  The SPL meter has to be in exactly the right place for this measurement to mean anything: pointed up at the ceiling, and positioned exactly where your head would be when watching a movie.  It isn't enough to sit at the listening position and hold the meter in front of you; it needs to be receiving the exact same signal as your ears.  Otherwise the levels will be off, and you will set the receiver incorrectly in compensation for the inaccurate measurement: consequently, you won't hear the mix correctly.  (This information comes from Roger Dressler of Dolby Labs, who absolutely knows he's talking about!)

Once all the speakers measure the same, it's time to set up the LFE channel.  Pink noise is again used, but for this it is band-limited at 120 Hz since a subwoofer is not expected to play any higher than this frequency.  Adjust the level until the SPL meter reads approximately 79 dB.  The band-limiting knocks 6 dB off the energy level of the pink noise, taking it from 75 dB down to 69, and the in-band gain of the LFE channel is then boosted until it measures 10 dB higher than the output of the main channels.  (One possible source of confusion may be that if the user was directed to calibrate their subwoofer 10 dB higher than the main channels without accounting for this reduction in level, and measured the band-limited signal at 85 dB rather than 79, they would then be hearing the mix with the bass playing back 6 dB too loud, which would sound bloated and just plain wrong.)

Note that the LFE doesn't necessarily have to be 79 dB exactly, since due to the Fletcher-Munson effect, our ears require greater bass levels to sound balanced with higher frequencies.  So if the SPL meter reads slightly higher than 79, it's probably okay.  But going much over 80 or so is pushing it too far.

Note also that even with these accurate measurements, what you hear is still greatly affected by the acoustics of your room.  Bass frequences in particular are notorious for creating phase additions and cancellations in various locations throughout the room, especially in small spaces where their wavelengths are longer than the dimensions of the room itself.  If there are standing waves at the listening position, acoustic treatment using bass traps and diffusion will be required to get the frequency response of the room under control, even if the subwoofer itself is ideally placed.  Adding a second sub can help in spreading the bass around the room more evenly.

With my system recently recalibrated using proper methods, I'm confident that what I've put into the 5.1 mix represents an accurate balance.  Certainly it is also a pleasing one, at least to my ear.  It is of course possible that not everyone will agree with this, and that any disagreement over bass level has nothing to do with system calibration at all, but rather aesthetic preference.  I'll admit to being curious as to which other films are being referenced, in terms of how much bass you think a movie ought to have in it.

If the LFE level just doesn't work for you at all, you could always try listening to the 1993 mix on its own.  The main channels will sound the same, but the amount of bass is more moderate.

Editing to add: I forgot to mention that when making SPL measurements, the meter should be set to 'C-weighted' and 'slow'; otherwise the numbers will be off.

I'll add also that I do have an idea for an alternate version of the 5.1 that wouldn't use the special edition as a source for the LFE channel at all, and that such a version might be more to the taste of those who would prefer a different approach to bass.  But I haven't started yet because I'm still figuring out the best way to do it, and have a bunch of other things to work on in the meantime, so it won't exist for quite a while yet.

Danfun128 said:

I know it's already been said, but, Wow! You can't even see Han Solos mouth in the first picture. So mouthless Solo is in your despecialized?

Looks like you underestimate just how terrible those official releases are.