TonyWDA

From one Fantasia diehard to another, welcome to the site.👍

So, for this, you’ll need a 35mm magnetic stereo print of the '82/'85 releases and a facility capable of capturing the sound from those prints to access any of those recordings; the interstitial commentaries from those dubs have, unfortunately, never been released to the public in any other form.

I do know someone who has a print of one of those digital stereo reissues and, color fading aside, claims it’s in pristine physical condition. They’ve offered to loan it to me before for preservation, so once I cross off a few other to-dos from my list, I may have that arranged for another project I still have brewing. I’ll keep you and anybody else interested posted.

In the Thunderbean release? Theoretically, yes, you can try, but for shots like this…

It’s a little more complicated because you’d need to feather mask the area of the faded image and then color correct it to match the surrounding information, which itself needs to be corrected to remove the excessive green tint. The problem is that the mask won’t stay over the faded portion unless you stabilize the footage first, as it has a noticeable gate-weave baked into it, and since the severity of that weave varies from scene to scene, one pass is unlikely to cut it. And this is all without a better print to keep pulling up for reference to ensure color accuracy, so what exactly would you (general “you”) be correcting the footage to?

So yeah, not impossible, per se, but it’s a lot of effort that, thankfully, won’t be necessary— at least not for this project. I can tell you this, though: if I had no other choice, I’d exhaust everything I suggested in the last paragraph.

My guess is that whatever print they had for the mag audio was too physically compromised to pull a stable sound capture. I can tell you from personal experience that Fantasia’s a tough find in good condition w/ the mag stereo mix; more than half the time, the colors turned almost entirely to red, the print has vinegar syndrome, and/or each reel has warped to some degree. The funny thing is that Thunderbean’s release does use the mag mix for the interstitial segments, but that’s it, and those portions also spread the LCR channels across the six-channel field.

CMGF said:

Wonderful day! I also want to thank you Tony for your help with tsMuxer.

Anytime. Happy I could help. ✌️

CMGF said:

And do you know maybe where is the surround mix in Thunderbean’s release comes from?

That’s the 2000 DVD mix, but the front left, center, and right channels of that mix were spread across all 6 channels in Thunderbean’s release, which doesn’t reflect the mixing Stokowski intended at all. In his music sheets, Stokowski marked precisely where, in the auditorium, the score would pan to during very specific passages. The music was never supposed to fire from every speaker, all at once, for the entirety of the program.

Oh, yeah, I wrote about this in a previous comment. The mono soundtrack is its own dedicated mix, not simply a downmix of the stereo soundtrack. It sourced the original score stems and, what sounds like, outtakes for that portion of The Nutcracker Suite. If you try lining up the stereo and mono mixes, both versions refuse to lock together without extensive editing.

Color-wise, it’s a lot more consistent. I don’t say that to slight Thunderbean, but because their release comprised more than one source, it looks stunning half the time but badly faded and marked with visible matting the other half— likely from going through the same projector aperture plate so many times. Thankfully, both the prints I’m working with for this project have none of these anomalies, and they’re brimming with strong color density. Things are still moving at a glacial pace over here, but I promise that this project is anything but shelved.

Depending on what scanning unit your print goes through, a digitized print seldom looks exactly as it does over a light table (which is how the pictures were taken) or when it’s projected. At least, in my experience, it doesn’t. So grading is necessary to get the image closer to how the print looked before scanning. Sometimes the adjustments to color, brightness, and contrast are minor, other times they’re dramatic, but in any case, grading is still needed to some degree, especially if there are inconsistencies in hue/saturation, gamma, and shadow detail across individual shots. For instance, the sample video posted here a year or so back absolutely needed correction, as there was an inordinate green veil over the image that didn’t belong there.

Sure does. 👍🏻

Also, be sure that ”Field order” for both the video file and project settings are set to “None (progressive scan)”. Don’t touch the pixel aspect ratio or else you will distort the image.

You can slow the PAL video frame rate down to 24fps or 23.976fps without re-encoding the source or downgrading to 480p by using tsmuxer. Done this hundreds of times myself.

Assuming your PAL DVD rip is contained in one .vob file, simply drop that file into the program window and check the box next to “Change fps,” which enables a drop-down menu next to it. Select the frame rate you wish to convert the video stream to and save the corrected video to either a .ts or .m2ts container.

I can’t speak for compatibility in other NLE programs, but I’ve never had any issues importing video I corrected with tsmuxer into Vegas Pro. Works like a charm.

You may want to hold off for a bit. Rugrats Movie has very recently been regraded/remastered for Paramount+, and that’s likely going to be the same transfer used for the upcoming Blu-ray that’s only a couple of months away from release as of this writing. All three theatrical installments are included, and I know for sure that Paris has also received a new grade. I wish I could provide a screenshot comparison between the first film’s Amazon HD stream and the new one, but when I went to play it on Paramount+, it wasn’t available. Perhaps to up preorders for the new physical release? Not sure, but I would wait until you get ahold of it before doing any major color regrading— it may not even be necessary. I did sample a few minutes of the first film’s remaster, and lemme tell ya, it does look great.

1. That can certainly be arranged. I have both on my drive(s) somewhere. Once I dig them up, I can send them over.

2. You don’t need an actual, physical surround sound speaker setup to do the comparing and contrasting. Media Player Classic and the a/v ffdshow filters run on Windows. If you’re running that operating system and have that software installed, you can load up the 1994/2003 audio tracks in separate player windows then do the channel soloing stuff from your computer while wearing headphones. But if you’re willing to take my word for it, then none of this is really necessary. The 2003 DVD audio (the one that’s NOT the “Enhanced Home Theater Mix”) is the original mix. Save for one minor hiccup, or burp in this case: if memory serves, the belch that Simba lets out in the “under the stars” scene after Hakuna Matata has reverb thrown into the front channels to sweeten the effect. In the LaserDisc/theatrical mix, it’s completely dry— no reverb in the fronts. You’ll have to patch that once you have the files.

Quality of content is sometimes more important than the quality of presentation itself. While the 2003 DVD audio is indeed compressed at 448kbps, it’s still the highest quality version of the original theatrical sound mix, with minimal tinkering, currently available. Take it from someone who’s studied the LaserDisc Dolby Digital soundtrack and DVD audio extensively over the years— repeatedly jumping from one mix to the other to compare and contrast. Aside from some noticeable tweaks in the dynamic range during the more intense sequences, they’re virtually the same mix, and absolutely the original one. Elements you can only make out in the centre and rear channels play out identically in both mixes.

If you want proof, the best I can recommend at the moment is for you to get ahold of the discrete LaserDisc Dolby track and 2003 DVD audio to hear the similarities for yourself and put all doubt to rest. Play them both in separate video player windows (my go-to has always been “K-Lite Codec Pack” version of Media Player Classic) and if you have ffdshow installed, isolate the centre channels in both player windows using the ffdshow audio decoder. Jump back and forth between both audio mixes, playing a few seconds at a time from the same scene(s), and see that they are, indeed, the same (original) mix. Do the same for the front channels, then the rears. You’ll find a lot of tells.

If your commitment to including lossless audio only is resolute, then go for the Dolby Surround version of the original audio— also on the LaserDisc. Other than that, the Blu-ray tracks are your best bet, but you won’t be getting the original sound design. So it’s either original but compressed (with exception to that aforementioned Dolby Surround mix), or lossless but remixed with questionable choices— a topic which goes far beyond the scope of this reply. Pick your poison.

I have a second print on loan and am currently in the queue at the scanning facility. At the moment, they’re swamped with an incredibly time-consuming commission so it may be several weeks (minimum) until I get the invoice— barring unforseen events. It should be well worth the wait; without giving away too much too soon, I will say that this new print is perhaps the only one of its kind left in near-mint condition, complete without a single disruptive splice in it whatsoever, in gobsmackingly-good IB Technicolor. It has something of a reputation in collector/projectionist circles, so you can imagine my surprise when the owner offered to loan it for the project.

I’ll update here as soon as I have the files and start any preliminary color grading. Things are moving, just very slowly.

But they are moving right along.

The differences between the 2003 DVD and the 2015 Signature Edition are most evident during scenes that originally looked much cooler. The latter release is dimmer, particularly and especially during those night scenes, and has a noticeably warmer (albeit very rich) palette throughout.

Top: 2003 DVD
Bottom: 2015 Regrade

What’s important to keep in mind is that neither version is “right” or “wrong.” The 2003 DVD is exactly what the movie looked like on 35mm, and the go-to for purists. However, Brad Bird himself supervised and approved the Signature Edition transfer, so in the end, it all comes down to which grading you most prefer, and I tend to be split between them myself. While I mostly prefer the original color timing, tons of individual shots— sometimes entire scenes in the Signature Edition look like they were sourced from an (of course, nonexistent) IB Technicolor print.

CourtlyHades296 said:

Are there plans to make an Uncensored Ultimate 1990 Edition?

That is the project’s objective, yes: an uncut “general release” version of the film using the best visual and auditory elements currently available— and I do mean the very best. I may or may not have some big news to share before the year wraps. For now, that’s all I can say. Hoping to update here very soon. 🙂

4throck said:

Those are the 3 original channels that were manually panned towards the speakers.

Yes and no. The audio on subsequent mag prints is a copy of the nitrate source material, yes, but it’s a premixed version of those optical tracks, which, by themselves, had no panning information hard baked into them whatsoever. So, what you’re hearing on stereo mag prints is not a 1:1 transfer of how the audio was printed on the nitrate optical tracks, but a faithful reproduction of how they were presented in roadshow runnings of Fantasia, sans “back house” surround effects.

4throck said:
Just to be clear, those 3 channels are not L, C and R. They are 3 discrete arbitrary sound channels.
The mix would direct them to whatever speaker combination was desired.

Again, this was only true of the first generation nitrate optical tracks, which had no specific house “horn” assigned to them and could be manually (and later automatically) panned to any given speaker over the course of the program. But on the magnetic 35mm prints of the film, you can safely refer to the individual channels of audio as L, C, and R because, when Disney and RCA conducted the wide-band phone line transfer to save the nitrate print in 1955, they also used the only remaining Fantasound reproducer to do it, which allowed them to preserve the original dynamic range and effectively steer each optical audio track to the left, center, or right of the front stereo field as they were intended to be presented. Thus, successfully generating a final left, centre, and right channel for the 35mm stereo reissues. This is the very same mix you can hear today on the stereophonic LPs, reel-to-reel tapes and cassettes, CDs, VHSs, LaserDiscs, and the 2000 DVD release.

So, yes, they were just unassigned “arbitrary” optical tracks on the original nitrate release prints, but discrete and correctly labeled left, centre, and right channels on mag prints because of how that surviving nitrate audio was mixed, transferred, and preserved on magnetic tape in later years. Therefore, there is no need to recreate the panning effects heard in the front of the sound stage; Disney did that before the 1956 reissue. The only effects missing now would be the aforementioned “back house” surround effects which, as you’ve also mentioned, would essentially mirror “front left” to “surround right”, and vice versa, as the film plays.

leogarcia said:

However, the DVD release of this short includes a 5.1 track, so I’m planning to mux that into the final! I’ll also include the print’s original track for completeness’ sake.

Whatever you do, make sure both video and audio are running at the same frame rate, or one will drift out of synchronization with the other over time.

If you plan to release your scan running at the original 24fps, then convert the DVD audio— which runs at 23.976fps— to run at exactly 24fps. You can convert the runtime easily by using eac3to. Assuming the short was properly scanned at 24fps, you should have no issues marrying the video and corrected DVD audio together. If the opposite is true— video running at 23.976fps out the gate— then just rip the DVD audio and sync it to the film scan. Again, shouldn’t be a problem.

In the event that you need help correcting either the video or the DVD audio using the aforementioned software, lemme know. I’d be more than happy to help.

FWIW, I attended the same 35mm screening at the SVA Theatre that @bslatky brought up some time ago. Before I move forward, I want you to strongly consider the source: I’ve watched and rewatched this movie on DVD with reckless abandon when I was a kid— lines of dialogue, the character’s inflections, incidental music and, yes, even the color timing are practically burned into my psyche. Subsequent viewings as I got older did little to help. So I went into that screening well prepared to point out any (IF any) differences between an original 2004 35mm print and the digitally-sourced home video releases I’d involuntarily committed to memory.

I remember my first and final impressions as if the screening took place merely yesterday, and it can’t be stressed enough: the movie looks no different on 35mm than it does on home video. At least, on this print it didn’t. Frankly, it felt like watching a blown-up version of the DVD/Blu-ray transfer with occasional scratches and dirt marks peppered throughout. Color wise, I couldn’t think back to one instance after the screening where the print didn’t match what was already on home video. I had an identical reaction after a screening of an original 35mm “Iron Giant” print at the MoMA in 2009: point blank, what’s on 35mm is what’s on the pre-Signature-Edition DVDs. So, while I generally admire and fully appreciate the effort to restore/preserve the 35mm film aesthetic where it’s sorely missing, I question exactly what this particular project aims to “restore” to The SpongeBob SquarePants Movie that wasn’t already there in the first place. Trailers aren’t always, if ever, a good reference source because the presented color timing was either not final or adjusted during editing to achieve a desired, albeit temporary look. A theatrical camrip isn’t dependable, either. In the theater, where you sit, how bright the projected image is, the condition and color of the screen that the image is projected onto, and the bulb used in the projector all factor into how you see the displayed image. Add to this equation a consumer grade video camera that likely didn’t faithfully capture just what the film looked like on the screen, and you have the least reliable visual reference for a full-fledged restoration project.

Please don’t misunderstand: I’m not poo-pooing your project. I think it’s pretty cool, actually. I simply believe that a different use of terminology would better suit what you’re aiming for here, which is more of a regrade of the approved color timing than a “restoration” in the traditional sense. Strictly in the interest of regrading, any trailers and publicity photos that are to your liking are perfect references for a regrade, and I can’t recommend the color matching tool created by OT’s very own Dr. Dre enough for a project like this. If you use it well, your regrade will look almost dead accurate to whatever reference(s) you choose; provided the colors in both the source and target images aren’t too different from each other, and even then the accuracy of the tool can surprise you. I’ve used it several times for personal restoration/regrade tests and projects. It works like a charm.

Godspeed.

Starbond9 said:
The center channel (if a true center not just a mix of L+R) seems to have been lost

The “centre” music channel, as it was originally presented, does exist on the magnetic stereo releases. I was given three discrete audio tracks— left, centre, and right— when my print was captured. It’s why the 2000 and 2010 home video releases also have discrete centre channels. Disney didn’t artificially generate that with some plug-in or standalone software, it’s always existed, and it does contain information unique to that channel; it’s not just a cheap sum of the left and right signals.

Starbond9 said:
For example, apparently the church bells at the end of the Bald Mountain came from the back of the theater, for example.

Check out John Culhane’s book on Fantasia from 1982. He was fortunate enough to attend the original roadshow release at the Carthay Circle Theatre in Los Angeles, CA, which employed the use of an automatically-steered Mark X unit. He confirmed somewhere in the book that when the bells began to chime towards the end of Bald Mountain, the sound did, in fact, come from the back of the room. It startled everyone.

That bridge will certainly be crossed once there’s a clearer picture of how this preservation project will be cut together. There’s nothing I can just slap the LaserDisc tracks over at the moment, and odds are the video will not be in perfect sync with the audio to the 1990 release anyway, so I’ll have to do all of the patch work myself if I decide to include those audio tracks. All in due time.

There’s been a misunderstanding. The extended audio I was referring to was the humming still heard even after the fade to black on the DTS 7.1 mix.

In regards to the chorus, the only conclusion I can draw from the volume differences in the mono to stereo mix is that, again, when Disney prepared the downmix for wider distribution in 1941, the opportunity to play with the original elements to prioritize different elements of the score, once again presented itself, and they did just that. What’s heard in the stereo mix is essentially what was heard on opening night-- the mixing on the mono track was prepared after the fact, and its mixing is completely unique from the final stereo. Therefore, not originally what was heard on November 13th, 1940. Ave Maria isn’t the only segment in mono with striking mixing differences from its stereo counterpart. In The Sorcerer’s Apprentice, a cymbal clash is placed right over Mickey’s second dive into the water after he wakes up from his dream. In A Night on Bald Mountain, an almost muted but sharp, sinister brassy note can be heard just as Chernabog begins to play with hellfire. Dance of the Sugar Plum Fairies from Nutcracker? Just try lining up the mono mix to the original stereo track. Two completely different takes. Those are the few that come to mind. There’s plenty of other mixing minutiae that point to the mono mix being a total reworking of the original stems and not just a simple fold-down of the final theatrical stereo track, and I’ll try to point them out in some way in the (hopefully near) future, but trust me, they’re all over the place. The finale’s chorus mixing is only one.

Anytime. Glad I could help. =)

Probably should’ve brought this up sooner, but that extended humming at the end of the Ave Maria also exists on the Blu-ray’s DTS 7.1 mix. If you have access to the audio, skip to the very end and listen closely.

With all due resect to the OP, I’d take that post with a grain of salt; I’ve worked extensively with the same vinyl recordings for a very long time, and once you adjust the timing (vinyls always tend to run either infinitesimally slower or faster than the master source depending on what player is used) they lock to the picture perfectly because they are, in fact, not outtakes. On both the 35mm mag tracks and vinyl/reel tapes releases, one particular incidental string noise heard on the left channel in the Toccata segment, as well as panning and discrete channel dropout across all segments, are identical to one another in both sources. Take my word for it, these are not outtakes. The surviving recordings on film and commercial audio releases are one and the same. That is all that has survived. Any actual outtakes would’ve been on the raw, unedited recordings sessions on the original nitrate film sources which, sadly, no longer exist.

The monaural downmix prepared for the wider 1941 RKO reissue was the basis for the subsequent mono pressings on commercial audio and the 1990 VHS/LaserDisc releases. Having personally studied the mixes just as carefully as I have the stereophonic recordings, I noticed identical instrumental prioritization across all releases (which itself can likely be attributed to an entirely new mix that sourced the original stems when they still existed).

Short of having access to a magnetic stereo copy of the 1956, 1963, or 1969 releases, the stereophonic vinyl and reel tape releases are technically the closest you can get to the “original” Fantasound, but not without its own warts. The original stereo tracks had zero reverberations baked into the final mix-- dry as a bone all the way through, and that’s exactly how it sounds on the surviving recordings used in the mag stereo prints. So the reverberations weren’t “removed” in the 1990 reissue-- they were never there in the first place. So, while the dynamic effect of hearing the score play in a concert hall does make the commercial pressings a worthwhile listen (I can’t count how many times I’ve enjoyed them), it’s ultimately an artificial effect, and the release still has the issue of a faint phantom centre channel that needs to be adjusted for proper listening, as well as occasional hard pans that aren’t nearly are strong as they are on the mag tracks.

TL:DR: the stereo and mono reel/vinyl pressings are totally listenable, but the optical mono and, especially, the mag stereo tracks will get you the closest to the original sound elements.

I discovered this very thing when I first tried cutting the Blu-ray footage around the LaserDisc audio, and the only conclusion I can draw is that for that particular portion of the entr’acte, Disney either used an alternate take for the general release cut, or they had Taylor come back and record a more streamlined version of the “Soundtrack” introduction to slap right over the orchestra b-roll footage. The former seems more plausible, but the funny is that the introduction audio to the Pastoral segment just moments later lines up with the Blu-ray footage perfectly.

A similar sync issue came up during the Bald Mountain/Ave Maria introduction. Not a single… word… synchronized properly. So I had to keep cutting (and in one instance even stretch) words around to achieve a somewhat decent result, and I still wasn’t completely satisfied with it. In addition, when Taylor brings up Ave Maria, he refers to it as “world famous” in the roadshow footage but “immortal” in the general release. That and the aforementioned syncing issues left me to assume that the finale’s interstitial audio used in the general release version was also an alternate take.

Got the 1956 trailer. 35mm IB Tech and in relatively great shape. A little splicey towards the beginning, missing the first 15 seconds as well as the last 20, but certainly nothing that can’t be patched using frames from the IB Tech feature print. As promised, this and all other theatrical reissue trailers should be sent in for scanning very soon. Might be difficult to make speedier progress during the holidays, but rest assured I have not forgotten about this project. I’ll keep updating as frequently and as quickly as I’m able.