The 12voltvids conversion method on laserdisc via Cloner Alliance Box Pro

[aramkolt]

I started getting into laserdisc (which is an inherently composite format) and thought I'd post an initial capture of the 12voltvids method produces without any fine tuning.

His setup is per this video:https://www.youtube.com/watch?v=UwRAxVlblZQ

It is player via composite to DVD recorder (DR430KU - acts as a comb filter and deinterlaces/upscales to 1080p), out through HDMI to Cloner Alliance Box Pro capturing in 4:3 aspect ratio at ~16Mb/s H264 progressive. Doing it this way doesn't rely on the Cloner Alliance to do the deintierlacing/upscaling. I also noticed with the Cloner via composite capture that it does a sort of zoom/crop that can't be disabled which results in a loss of 10% of the area the sides/top/bottom, probably to hide the padding side bars as well as head switching noise. I don't believe any zoom/cropping occurs when capturing via HDMI.

Some quirks I noticed are as follows:

1. Some digital compression artifacts are pretty visible, hard to say if that is due to the internal digital stage within the laserdisc player (it's a CLD-D704 which has a forced analog to digital and then back to analog conversion inside) or if it is due to the Cloner Alliance.

2. Framerate is 60 as opposed to 59.94 - haven't tested yet to see if that's the DVD player doing that or the Cloner Alliance, but I'm guessing it is the Cloner Alliance.

3. Proc amp settings within the Cloner Alliance affect HDMI capture as well, so that'll take some tweaking in terms of luma and chroma levels. I kind of figured that HDMI wouldn't have the proc amp apply since the signal there is already digital coming into the unit.

4. I perceive a bit of a fixed video lag over the audio of a few frames. Usually DVD recorders will compensate their processing time with audio delay to match, but not sure if that's happening here or not.

5. There's an interesting frame duplication thing going on here also - the original film I'm sure was ~24fps, so it seems the DVD recorder knows this and detelecines (probably appropriately?) with duplicates to get up to the ~60fps. This means that single frames often will have two following duplicates before there's a new progressive frame.

Also of note - many of the higher end laserdisc players from Pioneer (CLD-D703/4, CLD-79, CLD-99) have "chroma smear," so you'll notice that there's a sort of visible blurring to the right of luma/chroma at high contrast edges - I don't think that is due to this particular capture chain other than from the player itself.


A better test would be to hook it up to a pattern generator as that'd weed out some of the above issues, but figured I'd post anyway since I had everything hooked up. It should give a general idea of the quality that 12VoltVids setup would likely produce for laserdisc (assuming his player also has that chroma smear issue).

I am aware that the best practice is to capture laserdiscs via DomesdayDuplicator for rare content, but I'm mainly curious how close you can get to decode via traditional capture means or for live playback/viewing.

Curious what others think or if there are tests that others would like to see done with the Cloner Alliance Box Pro.

Here's the same sample video uploaded to YouTube if it's an easier watch there: https://www.youtube.com/watch?v=IfV8M80LaDg

Just as a reminder, you can advance single frames on Youtube with the ">" key and you can see the duplicates pretty easily.

[dellsam34]

Business profit a side, I would capture composite straight with no consumer devices in the chain with a pro composite capture device, some of the BrightEye's can be had for cheap in a composite configuration, I would also avoid compression and/or encoding during capture, But this is the hobby way, If you intend to do this on a business scale for profit like him, Gotmemories and the likes then none of these quality time consuming methods matter.

[aramkolt]

I take it you mean the Brighteye BE-25? That'll take composite and output SDI which from there you can capture with whatever will accept SDI. I do have one of those for testing also.

I would say the particular DVD recorder here does a fair job with deinterlacing for those that just want something ready to upload to YouTube without additional steps at probably one of the highest bitrates for a consumer device. I would have to guess that capturing in 8Mbit MPEG2 (still interlaced) and then deinterlacing/upscaling in post would probably give a better result, but that's on my list of things to test as well.

I'm pretty sure the 12voltvids method is quite a bit better than the elgato at least. The audio is obviously better I think which is an often overlooked negative aspect of the elgato video capture.