RAIDERS OF THE LOST ARK LaserDisc audio gap(s)

[pipefan413]

I watched a restoration of RAIDERS OF THE LOST ARK last night which included audio that I believe was taken from a 1992 LaserDisc PCM rip. For the most part it sounded good, however there are at least 2 film frames' worth of audio missing, and I believe it might actually be as much as 4 frames' worth in total.

The most obvious issue is at a point a few minutes before the 1 hour mark where the audio just drops out completely for 4190 samples of silence (88 milliseconds, just over 2 video frames at this 24000/1001 frame rate). This is after Indy says "that's it" and it cuts to a different shot:




The waveform looks like this in the 48 kHz upsampled LaserDisc track:




I don't know where the reel changes are in this film but it wouldn't surprise me if this was one of them and 2 frames were missing from whatever audio source they had as a result, but they recovered those 2 frames for the LaserDisc video and there was a mismatch. Pure conjecture though, of course. Having said that, I don't think this is a reel change since there aren't any cue marks on nearby frames at least on the litemakr scan I have.

Based on what limited info I have about it, I think the audio shown above is sourced from the 1992 US release LV 1376-WS. Does anybody happen to have access to any other LaserDisc audio sources for Raiders of the Lost Ark that may not have this section missing, either to use for patching, or as a complete replacement? I can patch it with a 35mm optical source if not, but it would probably sound smoother if I used a more similar source.

[pipefan413]

I realise I'm talking to myself by double posting here but probably worth updating: I forgot to say that I found schorman's preservation also had the gap but seemed to do a very short fade out to make it less abrupt, which is at least preferable. However, it seems the optical Dolby Stereo does actually have continuous audio here. What I assume I'd need to patch this without using the optical is some other LaserDisc rip from one that isn't a CLV (perhaps a CAV?) therefore has the side changes in different spots.

Anybody happen to know if that's available?

[spaceboy710]

I don't know much about this topic besides what is posted on LDDB but it looks like the only CAV release was LV 1376-2 in 1984.

https://www.lddb.com/laserdisc/11030/LV-...e-Lost-Ark

I don't own it (or any Raiders LD yet) though so I can't confirm anything. According to LDDB, this and the CLV released around that time have the original theatrical mix. They are both analog-only releases though.

[PDB]

I believe that was my raw capture of the LD which I'd have to see if I have anymore. I think I only have the synced version now.

Gaps in LD tracks are not uncommon so the fact that it is slightly different is not a surprise. For example, I had to plug my recent Aliens TC project as the Pan and Scan LD was missing information at a reel change.

Also of note the WS LD of Raiders might be a slight remix as there is one or two sounds know to be different.

There was a cap of the JPN Pan and Scan LD which had PCM 2.0 and the original sound but the sounds on the LD was weak. Most likely a digitization of the analog tracks.

[pipefan413]

I'm honestly not arsed if the overall mix or quality is different / wrong on any of these alternative sources, all I want it for is patching over a roughly 2-video-frame gap in the audio which seems to coincide with a side-change (as in, the LaserDisc side runs out, not the reel).

I could use the optical but it would irritate me and might be noticeable, especially since I've now pointed it out so you'll probably be listening for it.

[Chewtobacca]

Yeah, you're likely to find these "tucks" (like the fade you noticed in schorman's track) in most fan-made syncs. For sizable gaps, patching is a must, but a two-frame gap doesn't bother me, unless it's at a really noticeable point in a film. Usually, I make do with a short crossfade at an opportune moment.

[pipefan413]

Yeah, you're likely to find these "tucks" (like the fade you noticed in schorman's track) in most fan-made syncs.  For sizable gaps, patching is a must, but a two-frame gap doesn't bother me, unless it's at a really noticeable point in a film.  Usually, I make do with a short crossfade at an opportune moment.

If it had been schorman's version I'd heard when I watched the film, I probably wouldn't have noticed an issue... but it wasn't. I believe the track I heard was a slightly patched resync of the LaserDisc option included with the litemakr 35 mm scan, which evidently just left the gap as-is and just put it in sync with the video. As a result, it was very noticeable indeed, since the audio stopped dead for about 2 frames before resuming again. The fade in schorman's version makes a big difference, but I'd ideally rather patch over it so there's no gap at all.

I'm probably going to try it with an optical source first to see if it sounds decent enough or just sticks out like a sore thumb, but if anybody has a different LaserDisc capture that's probably going to be more seamless.

[schorman]

The different Laserdisc versions of Raiders all have side changes at the exact same places, if I recall.

Truthfully, a VHS would probably be the best place to get audio for patching, assuming there aren't any missing frames.

Watch out though, because there are two audio mixes of Raiders on home video.

[Chewtobacca]

As a result, it was very noticeable indeed, since the audio stopped dead for about 2 frames before resuming again.

In that case, I would definitely have crossfaded the audio or taken some other measure to bridge the gap.  I prefer not to combine mixes unless it is really necessary, which is why I don't mind the odd crossfade; however, if you can't track down the VHS, you might try a DPL1 downmix of the BD's 5.1.

Code:

flr = GetChannel(a, 1, 2)
fcc = GetChannel(a, 3, 3)
lrc = MixAudio(flr, fcc, 0.3205, 0.2265)
bl  = GetChannel(a, 5)
br  = GetChannel(a, 6)
sl  = MixAudio(bl, br, 0.2265, 0.2265)
sr  = MixAudio(bl, br, -0.2265, -0.2265)
blr = MergeChannels(sl, sr)
return MixAudio(lrc, blr, 1.0, 1.0)

Follow with the usual volume-matching and a bit of EQ if you are keen.

[pipefan413]

As a result, it was very noticeable indeed, since the audio stopped dead for about 2 frames before resuming again.

In that case, I would definitely have crossfaded the audio or taken some other measure to bridge the gap.  I prefer not to combine mixes unless it is really necessary, which is why I don't mind the odd crossfade; however, if you can't track down the VHS, you might try a DPL1 downmix of the BD's 5.1.

Code:

flr = GetChannel(a, 1, 2)
fcc = GetChannel(a, 3, 3)
lrc = MixAudio(flr, fcc, 0.3205, 0.2265)
bl  = GetChannel(a, 5)
br  = GetChannel(a, 6)
sl  = MixAudio(bl, br, 0.2265, 0.2265)
sr  = MixAudio(bl, br, -0.2265, -0.2265)
blr = MergeChannels(sl, sr)
return MixAudio(lrc, blr, 1.0, 1.0)

Follow with the usual volume-matching and a bit of EQ if you are keen.

This is a good idea and that code is extremely useful except that I don't know what it's intended to be fed to! What would this be done in?

Currently, the only way I know how to downmix 5.1 to DPL (as I did for Near Dark's custom encodes of the newer StudioCanal logo jingle) is doing it with eac3to, but that is almost certainly DPLII or perhaps DPLIIx (I don't actually recall right now).