Info: Laserdisc has still something to show up...

…despite the fact that it’s an analog, low definition, obsolete format… I love that old big shihy disc! (^^,)

As I wrote in this thread, I managed to write a script which mix and upscale two versions of the same material, PAL and NTSC. I called “PaNup - PAL and NTSC upscale”, plain and simple (and easy to remember, I hope).

Because I’m lazy, I searched “in the maze” of my old hard disk, and found two captures, PAL and NTSC, of the same scene. They were made more than two years ago, and I was at the beginning of my capture skills, and they were only test captures, so quality lacks, and there are missing/repeated frames all over… but I found some frames that (almost) matches, so I said myseld “let’s give them a try!”…

I took those two captures, PaNup’ed them, denoised, ColourMatch’ed to Harmy’s DE, and here you are the result:

(if you want to download the small clip, to analyze better each frame, this is the link)

Not high definition, but stand up at par (IMHO) to the DE. What do you think?

...forgot to wrote, one is the *enhanced* laserdisc capture, the other is Harmy's DE...

I have been doing this sort of thing using a registration program, I would love to see your script.

Registration program? What is it?

And, did you guess which one is the DE? (^^,)

By the way, many compliments for your job, I can't wait to see a 10K scan of a 35mm (or, better, a 70mm) done with your scanner... if it's Star Wars, here we will be all happy, but I could "live" with another movie, too...

I did another test using an old NTSC SE capture I made more than two years ago - rough, no optimal capture settings, no high quality cables, mid-low laserdisc player for NTSC standard... -> (click for bigger version):

(you can download a few frames clip here - just 1.33MB)

Top spanish HDTV ("TodoTV") - ColourMatch'ed to be close to the LD (if you see strange colored artifacts, it's the script fault...), bottom LD capture, denoised, upscaled and regrained a bit.

I can't think what could be achieved with proper capture settings, high quality cables, and the Pioneer HLD-X9...

Now, I don't want to tell that a denoised and upscaled LD video is on par with HDTV, but if no other HD version exists, at least we could obtain a decent version using laserdisc as source (some examples that came into my mind: Ransom Extended Cut, The Arrival with alternate ending...)

I just compressed it using virtualdub...

Chewtobacca said:

I guess DE is on the bottom.

Ehr... actually not... (^^/)

OK, here you are my latest test - download (6.5MB)

I captured two brief scenes, from the PAL UK and NTSC US 1997 SE; denoised, upscaled, and overlaid.

Comments are welcome!

I've done these tests only because...

IT!. . . COULD! . . . WORK!!"

Actually, it is possible to use both NTSC and PAL trilogies to build up a better version... I thought to do it some months ago, but as it seems the interest is near zero, I don't know if I'll do it... a lot of work, captures, alignment...

First of all, I apologize, as I have uncorrectly resized the picture at wrong aspect ratio... as the 1997 SE AR is about 2.25:1, and I resized them at 2.35:1... it's THE FORCE... of habit!

I used tfm/tdecimate to IVTC it, using default settings, so maybe it is why the quality lacking a bit... but hey, was just a test, after all! (°°u)

I took the old PAL capture, the new PAL capture, mix them and... surprise! History repeats itself... it's the same technique I used with my OUT and SET projects... at least I have reached some advantages:

1. I was sure this technique (overlay multiple copies of the same material) work, and I have a confimation
2. during this time my knowledge and skills advanced, and it could be seen in the mix Vs ruLes
3. for ruLes, something went wrong with the captures I did at the time - player, settings, cables, who knows? - or I simply applied a wrong color correction setting - albeit I'm sure this scene received no color correction at all...

And I discovered that if, for mistake, you feed a non-mod4 video file to virtualdub, something terrible could happen...

Probably I'll use it into one of my next project... keep an eye on the "other preservation..." forum!

OK, I resurrect this thread after years because I think I have achieved quite good results - if you think that I used a clip taken at the time, and with same hardware and software setup, of the ones used for the previous comparison...

This implement PaNup & SuperResolution, and... well, without much further ado, go here for more info - can't copy & paste again those links... (^^,)

Great news!

Have you thought about the phase inversion trick to cancel out any chroma errors?

happycube said:

On my end, I think I'll relatively soon have a complimentary technique to merge two captures of the same master (preferably from different disks) - should be particularly good for rotters (since most of the time the rot will be in different locations) and disks that are pressed a bit noisily (points at Technidisc ;) )

We currently have TooT (two-out-of-three: average the 2 most similar pixels of the 3) and median-of-three for doing this with three captures (three different copies of the same master). I'm doing this with THX 1138 right now.

Requiring one less LD is most certainly welcome!