Jared's Pile

My summer in Michigan is almost over, and there are some things that I need to wrap up before returning to Florida. Since I’m unlikely to fit either of my VCRs in my luggage, much less both, I’m scrambling to get as much use out of them both so I won’t have to wait until next summer.

All of yesterday, I’d used the Domesday-modded VCR to get several RF captures of three tapes – Winds of Change, Pryde of the X-Men, and Panda’s Adventures (Panda no Daiboken). I’d already had a few captures of Winds, but a couple more might might help to further reduce its visible noise. Multiple captures will likely have the same benefit for the other two films.

Panda’s Adventures is an interesting one, because it appears to have been transferred from film to video as-is, meaning its original 2.35:1 aspect ratio is squeezed into a 1.33:1 frame. I can only imagine that the distributor was in such a rush to get tapes on shelves when VHS was a new format, they just decided that kids wouldn’t notice or care. I sure !

Originally I’d only planned to capture the audio, to combine with the Japanese DVD, but I don’t know how well it’ll line up. Audio not lining up is a surprisingly common issue I’ve run into…

---

My “final” edit of Metamorphoses – the French version of Winds of Change – has been rendered out and made available to my patrons, though it’s far from a perfect edition. Even after having its frame rate changed from 25 fps to 24, it doesn’t quite sync up with the English version. The weirdest part though is that the French audio is even more out of sync by the end of the film, but not because of anything I’d done – it appears to be the same on the original tape. Since I plan on discarding the French audio for the next version, it’s not something I’m going to worry too much about.

What’s fascinated me about Winds/Metamorphoses, is the original version’s status as lost media. After a disastrous premiere, seven minutes were cut from the original film, its segments were rearranged, the soundtrack was replaced, and narration was added. The Japanese version is closer to the original, being only two minutes shorter, and having its segments in the original order, but still uses the replacement soundtrack. Something I’d discovered over a year ago, though, is that the English/French & Japanese versions of the film both have bits that the other lacks. When combined, these bits bring the film closer to its original length. Combining all three versions isn’t something I’m in any rush to do at this time, though.

---

One last thing I’d thought I’d do with my multi-format VCR, was modify it to also work with the Domesday Duplicator, so that I could RF-capture my Metamorphoses tape. After adding the connector, the VCR doesn’t work. I guess I shouldn’t be surprised – all summer long, practically any time I’ve wanted to use this VCR, it puts up a fight.

At first, the problem was that the motor that drives the tape tray didn’t seem to work. After attempting to address that problem, the unit makes a high-pitched noise when it’s turned on, which can’t be good. I don’t know if the wire I’d added is the problem, or if it might be something else. Maybe one of the components just died and needs to be replaced, which…

I try not to think about how much money I’ve spend on this project, so far. The two French tapes cost me a combined $300, and one of those broke the first time it was rewound. The VCR to play them cost another $300+, and has had one problem after another… but at least I was able to get enough traditional captures to assemble a full, fairly clean copy of the film. If it’s dead now, before I can use it with the Domesday… well at least it served its original purpose.

I’m not even sure if a Domesday capture would be much of an improvement over what I already have. Looking closely at the video, it looks like its resolution was cut in half before it was transferred to tape. Out of 576 vertical lines, the letterboxed portion only uses around 400, but the effective resolution of that area may only be 200 lines. The Domesday Duplicator isn’t going to change that.

---

All of this film & VHS nonsense aside, I’m looking forward to my return to Florida as a bit of a reset. All summer long, I haven’t really been able to relax. Whether it’s work, or personal favors, I’ve been doing what someone else wanted, not what I’ve wanted. Now I’m started to feel the need to draw or even animate again, but I need to get back home to do so if only so that maybe – MAYBE – I won’t be interrupted.

My VCR’s motor didn’t take much effort to fix. A little bit of super glue, some solder to bridge the traces, and by the next day it was working again! Better still, the dab of grease that’d gotten on the one copy of Metamorphoses didn’t have any impact on it, which was a huge relief.

After capturing that one Metamorphoses tape, I ended up with more questions than I’d expected. It seems like every time I look at it, it has more surprises in store for me. A known problem may be due to the fact that I’m capturing it on a VCR that doesn’t support SECAM – periodically, the brightness will shift either up or down, then back again. What was surprising, this time, was that the brightness shift happens roughly every 3.5 seconds. It’s so precise that it has to mean something, but I don’t see much point in figuring it out. The important thing is that it’s not the tape – I’d captured it twice, and in both captures the shifts occur at different times.

The biggest surprise was when I tried to de-interlace the video. That it’s interlaced at all is somewhat infuriating – I thought the standard method of transferring film to video in Europe was to speed it up from 24 frames per second to 25. In this case, though, it appears that the first 24 frames of every second are interwoven with the next frame, while the 25 frame is a whole repeated image. This is a somewhat easy fix – for the first 24 frames of every second, half the image just needs to be time-shifted.

…except that doesn’t actually improve the quality. Instead, what it reveals is that the film was apparently processed digitally, and its vertical resolution was cut in half before it was transferred to tape.

While the halved resolution is disappointing, at the very least it’s still much clearer than the copy I’d found on Youtube. And since my VCR can’t play back the tape in color, that Youtube video could still be useful – I could extract its color, and combine that with either or both of my captures… and that’s where I ran into another surprise – my copy and the one on Youtube are cropped slightly different.

I’d bought two copies of the film, just in case one of the tapes ended up being bad in any way. Now I wonder what I’ll find when I’m able to capture the other tape. Maybe it’ll be cropped like the Youtube version, or different still? Maybe it’ll even be full resolution!

For now, the captures I have may just be something to tinker with when I’ve nothing better to do. The raw signal from my VCR is either too weak or otherwise wrong for capturing with a Domesday Duplicator, and since this film’s resolution was halved, I’m not sure if there would be any benefit to using the Domesday. And even if I do use it, it may be quite a while before VHS-decode can process SECAM. The only other option I can think of is to find another VCR that does support SECAM, but the ones I’ve seen are expensive. And whether they work is another issue…