althor1138

I've got a pretty good idea of which star wars ntsc ld's are notable/desirable to own but no idea which pal versions are worth grabbing. Can somebody give me a list of notable pal releases and what is good about them?

BTW, the software decoding is working in Linux and I've done an RF tap on my newly acquired s315 and it works perfectly. The software is still in testing phase but it is already very functional. The guy doing it is updating it quite a lot now. He's working on the comb filter now and digital audio next. Once the comb filter is updated i will probably resume capturing. Also, I just picked up a couple of Japanese p&s versions of the trilogy. I will add them to the first post when they arrive.

I'm just saying that even if you had a healthy stephen hawking and a young albert einstein in place of the 2 ewoks that were chopping with the axes, the chances of the at-st wandering into a spot close to the middle point of these 2 suspended logs would be pretty unlikely and then being able to synchronize the chopping so that both logs released  at the same time would be almost impossible.

The timing required to make sure that the trees hit on both sides simultaneously probably would require some fancy calculating and I doubt it could be pulled off manually in the heat of battle(with an axe no less). I would say plausible but not likely.

AntcuFaalb said:

althor1138 said:

This is the exact reason I'm going to try the software decoding method. It will hopefully alleviate most of this craziness.  From what I understand, a lot of it gets introduced in the players before it ever hits the tbc.

Thankfully (for me) the Runco LJR-II only has a tiny amount of it; it's barely visible.

The image I provided came from happycube and I think it's from his Pioneer LD-V8000.

 Is it decoded using his software or just capped from the output of the player?

UPDATE: It's still doing the same crap.  I've asked over at lddb what I should do about it.  Maybe a new pickup will fix it. I have a feeling that either the gear on the motor is messed up or the motor itself is going bad.

I would tentatively say that the R7G is working again.  I got it all disassembled and was staring at the motor holder and noticed it was missing a screw!  The crazy bastard that serviced it last forgot to put a screw back in it! Anyway, I cannibalized a dvd burner and found a screw with the same thread size,pitch, and length and put it in. You could wiggle the motor around with your finger before inserting the screw so I can imagine that was the problem.  I'm not positive that the little shafts that hold the gears aren't broken at this point but I thought I'd test it first to see what happens. It seems like a newer motor holder because apparently the old ones are grey and this one looks black just like the ones I've got in the mail.

It's played successfully for an hour and is on side 2 now. I'll probably burn it in all night and see if it skips again.

AntcuFaalb said:

Spaced Ranger said:

I had looked around for a plug-in to handle "ringing" or, more accurately, "edge-enhancement haloing" (that's sharpening done the wrong way).

Try anything you find on this monster. :-D

 This is the exact reason I'm going to try the software decoding method. It will hopefully alleviate most of this craziness.  From what I understand, a lot of it gets introduced in the players before it ever hits the tbc.

AntcuFaalb said:

nafroe said:

One hell of a collection you've got going there!!

Thanks. I can post images of the rest of it if althor1138 doesn't mind.

 Do it!  This thread is about laserdiscs after all!

Awesome!

Well, I bid on the unopened swe on tradera hoping it was the technidisc.  It wasn't. It was the pioneer pressing from '92.  I already have the Mitsubishi pressings so I will get around to them both eventually. I don't really feel like trying to find the technidisc at this point.  I've tried 3 times now to acquire it and the first time mysycamore lost his copy in the mail on the way to me so for the time being I guess I will not try to purchase it anymore unless I find one that has clear enough pictures for me to be absolutely certain before purchasing it.

I am in the process of acquiring an PAL player so anybody that wants to donate PAL Star Wars discs to the cause please feel free to drop me a PM.

It looks like at this point that the software decoding method is going to work.  Once the PAL player arrives I'm going to mod it and use it for all future NTSC and PAL captures.  This has the potential to surpass any laserdisc player circuitry or capture card. It is about the best way to currently preserve a laserdisc too because it captures the full bandwidth of the RF stream before it ever hits any circuitry in the player. 

I almost feel like if new frames are found that they should be added and then that would become the gold standard until new frames are found again. The yavin hangar frame for example does exist in the ntsc gout it is just as half a field instead of a full frame. Shouldn't we consider that field/frame to be part of the gout? Does adding this frame back in throw off the audio sync or is the audio sync already off because of the missing field? Anyway, I support whatever is decided upon because it allows the community to come together on one page when doing projects.

I actually found an old pm where I asked him this a long time ago.  Here's his Response:

Jacket should *PROBABLY* have "Fox Video" logo at at the bottom.
(Older ones said "CBS/FOX".) Auction photo looks correct.

Inner sleeves should *PROBABLY* be of 2-ply construction:
paper on the outside with polythene lining.

To be certain, you must ask the seller to check the mint markings.
See examples there:
http://www.blamld.com/LaserDisc/Technidisc.htm
Look for the name Technidisc with an icon resembling a CRT screen
etched into the 'dead wax'

I guess there is not a good way to know since they are unopened.

Well now if we can only confirm that the one for sale on tradera is a technidisc maybe I can get ahold of it.  He wants a high price for it but I guess I'd bite the bullet since it's local and unopened.

Sweet. So really the only thing worth preserving from the Technidisc pressings is SW then?

@Happycube.  Thanks for the advice. I did find a lddb'er who'd posted a video on how to fix the d505 and it looks like the same setup so that helps a lot to have a visual reference to look at.

@Mysycamore. The R7G chops everything off above 255.  Side 1&2 of SW-JSC have some sort of weird mastering problem on my pressing where the black level starts at 16 even though it has the full range of a Japanese disc.  This means that there is white information all the way up to 260-265.  On the HF9G I could just extend the luma white point in virtualdub and recover everything.  On the R7G I have to turn on d-ext and bring most of it back down under 255 so it doesn't get chopped off by the player.  I guess there might be a tiny bit that gets clipped off but it is only in a couple of scenes on the tantive and maybe 1 or 2 elsewhere I think.  I've been able to write a script that masks that information and copies it from the HF9G to the R7G but honestly I can't see a difference at all other than I have set the white levels higher on the R7G capture. I've only ever seen this on swjsc side 1 & 2 so I don't think this will ever be an issue on any other disc. BTW, could you do me a favor and let me know if you think the SW&ESB LD's on tradera right now are the Technidisc pressings? Is there a Technidisc ROTJ and is it any different than the 2 previous pressings? I'll probably throw a bid down on it if it's the Technidisc.

I usually try to get it right during capture. I usually do a 5 minute capture and see if the black level appears to be right.  Usually  on a Japanese disc the black level will be between 0-16 so there's no need to  mess with anything because that can be adjusted in post no problem.  It's more important imo to pay attention to the white level.  The white level on the JSC for example extends actually a little bit above 255 and the only way to capture this without it getting chopped off is to extend the luma white point in virtualdub during capture. One of my players chops off everything at 255 but luckily it has d-ext so I can turn that on and recover that information if I use that player for capturing.

Btw, the black level on the JSC is around 0(for the black bars) but the active picture area's black level is a little above 16. I believe this was a goof-up on their part and is the reason why the white levels extend past 255. Here is a little example.  I had the wrong capture settings here and that's why the black level for the bars appears over 16 instead of down around 0 like it should be but you get the point.

I would probably just capture it 5 times, average it, and then put it on absw.  If there are other raw riles available to merge it with I'd like to download those.

msycamore said:

Great to see you're doing everything you can to improve on your previous work, Althor! Too late for the sample, care to reupload or have you all come to a decision already? Also nice to hear you are planning a THX-cap. Man, you're a machine! Keep up the good work! :)

 Nothing definitive yet.  I've not gotten any feedback yet on the reference disc captures.  Here's some screenshots and on absw i've posted the raw reference disc video files which are about 10GB.  I'm currently uploading R7G-SWJSC-5xRAW.  Side1 is finished now I believe.  The R7G captures are not perfect but right now it seems to me that they are a step up from the HF9G. I figured now that the HF9G-SWJSC and R7G-SWJSC is available it can be evaluated by better eyes than mine.  The reference disc tests are interesting too, for sure.

I am also currently trying to implement software decoding by capturing the full RF signal with a makeshift software defined radio in linux. There is a whole thread devoted to it over at LDDB so I won't delve into what that's all about here.

I'm not so great at compiling custom drivers and using the commandline for everything. I've got all of the hardware to do it now I just have to figure out how to use it lol. It's a slow process but if I succeed I'll definitely let everybody know.

UPDATE: Well the R7G is not tracking right. It begins looping and you can't fast forward past it. It started on side B but is now doing it on side A as well.  I believe the motor holder is maybe broken and/or the slider/rail mechanism is gummed up.  The player is in excellent condition but if it hasn't been used for ages stuff like this happens. The m-holder on these is notoriously brittle apparently.

Anyway, I've ordered the m-holder and new belts. The new m-holder is made from much more durable material from what I hear.

It's kind of a delicate operation though to remove the slider/rail mechanism and then pull out the optical pickup assembly to get to the tracking motor.  I'm not going to spend 500 dollars to send it to the US to get fixed though and I'm certainly not sending it to a local shop so some 20 year old kid can try to fix it, break it even worse, and then tell me it was way beyond repair from the beginning anyway. So it seems if anybody is going to send this thing to the grave it'll be me.

Thanks so much for taking the time to set up this information and put it together! As you said it's around but it's a bit hard to find sometimes. I plan on Gout Syncing any further releases so that schorman's laserdisc audio archive should be compatible with my projects.

When you say the title card appears on frame 689, I'm assuming that means starting from frame 0?

It's a shame that NTSC Gout is missing a few frames in spots here and there.  Perhaps we could add those frames in and then start calling it official gout sync or something.  I guess that is no man's land though.

@ Catbus.  If you look at Chewtobacca's SW Pal script in post 2 where it says deleteframe(144053), that just means that you delete the missing Yavin hangar frame to sync up to the NTSC Gout which has only half of a field for that frame if I remember correctly. Assumefps(24000/1001) sets the frame rate from 25 down to 23.976.

If I can find one where the shipping isn't 4 times more than the cost of the discs I'll order one.

Color Bars Frame #1

Color Bars Frame #2

Magenta Frame #1

Blue Frame #1

Green Frame #1

Video Sweep Frame #1

Video Sweep Frame #2

Video Sweep Frame #3

Video Sweep Frame #4

HF9G vs R7G using the ati theater750 PCI-e card. The raw video files are available on a.b.sw. As always, view the full sized picture. Left click to view full sized.
Test Pattern Frame #1

Test Pattern Frame #2

Test Pattern Frame #3

Test Pattern Frame #4

Just because there is slightly higher chroma resolution with 4:2:2 doesn't mean that it is perceivable. I'd recommend putting out only blu-ray compliant files. Even muxing into a file container such as mkv can have unwanted effects.

The best would be to release the .264 and audio files together and let everybody mux them into whatever they want. I guess then there would be twice as many people asking, hey what do I do with these files? I can't watch them! LOL.

EDIT: BTW, it looks fantastic. Thanks a lot for sharing it with us!

Here's what I always do.  I hold all 6 films in my hand. The prequels in the left hand and the original trilogy in the right hand.  Now, continuing to hold all of the films in this manner, I shove my left hand into a pot of boiling lead. I then proceed to watch them in this order: SW,ESB, and ROTJ.

A small book could be written on the strangeness of the JSC. It would be fascinating to talk to those guys who made it and just ask them what was going on when they made it lol.

Now that I've been shown how to do it, I pretty much always manually IVTC because everything else fails too much. The JSC was too much for me but Darth Mallwalker IVTC'ed the whole trilogy for me. As long as there is only a cadence change once in awhile, manual ivtc is quite easy and reliable imo.

If the VBI was visible on the captured frames I'm pretty sure it could also be useful in avisynth. I'm sure a function could be written using averageluma() to automatically "read" the white flag line and determine if fields should be deleted or not, for example.