RU.08

You_Too said:

@RU.08: I'm sorry if I misunderstood your intentions. Maybe we just have different preferences when it comes to upscaling.

Perhaps. I believe EEDI2 will give the best results for (at least) most scenes in ANH, and I can't believe I let it slip me by until you mentioned it. But I certainy don't agree with using it to scale horizontally, a straight splineresize would work better then using EEDI2 horizontally, and now that you've mentioned it the "pattern" it introduces when used in both directions is pretty much unmissable. Anyway, I'm going to see how EEDI2 goes on Robocop Criterion now; that DVD has a lot of the same problems as ANH (DVNR and alising) and the fact that I'm not working with an upscale will mean I should get good results (except for robovision scenes, obviously), I'll make a new thread once I have a script looking good, and I'll come back to SW (ANH & ESB) once I have the NTSC version. :) Yes, I'll make my own thread when I feel I have a compentent script. :)

Darth Editous said:

What's the idea behind using qtgmc after the first eedi2?

DE

Good question! EEDI2 doubles the vertical resolution and each new line is an interpolation, but the way that EEDI2 works it tries to join "lines" like diagonal lines together (in the new lines), and that in itself very effectivly reduces the "jaggies". It is not designed to work horizontally, but you can force it to by using turnleft, etc. EEDI2 does not modify the original lines which let's say are now the "odd lines" and the new ones being the "even lines", it only creates new lines. qtgmc is an advanced deinterlacer, but a very slow one, and will reduce the jaggies in the image without resizing; so essentially using it is hitting the image with a second "deinterlacer". You wouldn't want to use it before EEDI2 because then it's "damaging" the original picture before resize; plus the fact that EEDI2 already does a lot of the work for you means you really only need a much lighter touch with a deinterlacer post-eedi2. And as I've already mentioned I don't think EEDI2 should be used to stretch the picture horizontally because it then "joins more lines together" and your image comes out looking like it is made up of crisscross strokes. And sorry for missing your question the first time. :)

All I've seen of project blu is what's in the forum and stuff Harmy used in his despecialisation. I personally don't know how I feel about using EEDI2 in a way that it isn't designed; by using it to stretch the picture horizontally it intorduces far more side-effects, for instance have a look at the background in your R2 pic, You_Too, where the stone patterns have been somewhat reconfigured by EEDI2 joining what it thinks are "lines". Even the scuff marks on R2D2 have also been reconfigured into a pattern thanks to EEDI2. I was wrong about the grain removal, it's been obscured by EEDI2.

In any case, as I said to begin with I believe the results are beatable. I need to aquire the NTSC disc to go much further. But for now yes I've removed the jaggies without using qtgmc (thank goodness because it's frightfully slow) or loosing much if any quality, here's a screenshot you can compare it to my earlier one from the same scene. For reference, I'm using EEDI2 once for deinterlization and then nnedi3 for scaling (although it makes little difference at this point if I use nnedi3 at all):

Interesting. eedi2, qtgmc, turnleft, eedi2, turnright, spline36resize, LSFMod. I actually think I may have lost more detail this way, see what you think:

Moth3r said:

Best upscaling/antialiasing results I've seen so far is whatever dark_jedi used on his Blu V2 sample. I compared the sample against the output from a nnedi3 upscale, and it was much better.

GOUT NNEDI3

DJ Blu V2 Sample

I think his results are repetable, if not beatable in AVISynth. I only did a quick test where there was very little done to the image prior to upscaling. DJ's version has had substantial grain reduction and anti-alising applied before upscale and getting that right is definatly the key to improving the upscale result; however as I mentioned you still need to handle different types of scenes differently, espcially space scenes and scenes where the DVNR was very heavy.

Here is that scene with some grain removal and aa applied to the raw PAL gout (notice how the framing is slightly different, I would expect better  success for NTSC especially in regards to anti-alising). NNEDI3 is then applied.

laserdisc said:

Interesting. I didn't know that. I'm still curious about the potential increase in quality of running gforce/gout through an iscan. It doesn't just do upscaling but also does general image enhancement apparently. I've seen some people mention iscans on this forum but in relation to laserdiscs. So regardless of whether it's avi or vob i'd love to know how much quality is improved with an iscan.

You won't ever get the best results for the GOUT using a single upscale method. Here's an example of the improvement in overall picture quality that can be achieved using one method...

Top frame is a somewhat stabalized GOUT (no antialising has been applied before upscale), the next two show two takes at an almost idental upscale method. The first is using a filter in avisynth (nnedi3), and the second is using a commercial filter (Video Enhancer). Both are resized using the same method (Spline36Resize), both have the same simple filters applied to the upscale. As you will see there is almost no difference between nnedi3 and VE; however while these filters work well on some shots, they work terrible on others. Garbage-in-garbage-out, both display problems handling the ghosting trails left by the DVNR and both also have trouble in some space scenes. Furthermore VE cannot be used as a frameserver (although you can use avisyth filters within the program itself) which makes nnedi3 far more attractive.

Larger view (full resolution for 720p):

TServo2049 said:

RU.08 said:

They removed a lot more than just the "offensive" vaseline blob.

OK, this is something that has bugged me for years. I'm not buying the "vaseline" story. ...

Very interesting read, thanks for the info!

ChainsawAsh said:

See, all your methods use AviSynth.

First, I'm doing this on a Mac, which can't run AviSynth.  Second, I don't know a damn thing when it comes to command-line stuff - I need a GUI in order to do anything.

True AVISynth doesn't run on Mac. But it is still silly to do it manually when an automated process can do it for you near perfectly. Frankly, they should just get their original NTSC master tape out and IVTC it properly - paying close attention to the effects scenes.

Thanks for the amazon link. Bundled with two random classic episodes, no thanks. Especially not for £14.99 + postage :(

You can make it a normal AVI file.

I was placing an order on amazon.co.uk for some DVDs+BDs, but this title is OOP. :( I don't really know much about it, I have seen various Doctor Who's (the first one, Hartnell, and some of the others, and some of the new ones). But being as I am a bit of a Paul McGann fan it would be nice to see it.

Anyhow, yes the process is reversable. However, it's not *completely* reversable when your material has had effects added to a telecine.

You indicate that Speedup AND blended fields are present, in that case you need to do this in Avisynth:

Yadif(Order=0,Mode=1)
srestore(Frate=25,speed=1)
AssumeFPS(23.976)

50i > 25p > 4% slowdown.

If however, speedup is NOT present and the underlying material is in fact 24p, then it's:

Yadif(Order=0,Mode=1)
srestore(Frate=24,speed=1)
AssumeFPS(23.976)

50i > 24p

That's it, that should be all you need to do. If it comes out looking like shit, change the order in Yadiff. If it still looks like shit then try the second version above. The effects shots, you may need to go through manually since there's no perfect way to convert that back to 24p.

Srestore isn't human, so of course it won't be 100% perfect, but it does a very good job and was specifically made for this purpose.

You could also do it with the NTSC disc and compare the results, this code should work with the NTSC version:

Yadif(Order=1,Mode=1)
srestore(Frate=25,speed=1)
AssumeFPS(23.976)

60i > 25p > 4% slowdown.

Or, if that doesn't work, then:

Yadif(Order=1,Mode=1)
srestore(Frate=24,speed=1)
AssumeFPS(23.976)

60i > 24p.

All things being equal they should end up looking pretty much the same, although this will also depend on things like DVD bitrate, etc, so you might notice one looks better than the other in terms of underlying authoring quality.

Required plugins are yadif.dll and srestore.avs as such:

import("srestore.avs")
LoadCPlugin("yadif.dll")

Well since I accidently posted a 2nd post (see below) I may as well use this one to explain the theory behind the blended frames...

Your source is hard telecine 60i NTSC material (2:3 pulldown). Although very easy to completely remove now on any home PC, instead of being a perfect IVTC job, every so often you have a completely "random" blended field here and there because the process thought that was the progressive one.

Anyhoo, this is then converted/resized to PAL (field by field), and as such you shouldn't actually have blended frames, you should only have blended fields (unless it was authored as progressive and then you might have a problem!)

Then you have the effects shots. Because they were done at 30fps, you now have 24fps and 30fps material laying on top of each other (yuck).

But what if the material is actually not sped up? Okay first we have random blended fields resulting from a poor IVTC, and then we will have 1 blended frame per second.

Okay, there's a simple description of the problem and how it happened. :)

How to solve... you go through and yank out all the blended fields, and then put the frames back together, interpolating where the missing fields are. Simple. Well except for the effects shots because you only want to remove the blended 24fps fields, not the 30fps ones (in other words, if there's a blend in the effect but not the action you ignore it and accept it as being inevitable). And if it hasn't been sped up you simply remove the extra frame.

Srestore for AVISynth was made for this exact purpose. Well except it will prob identify blended effects as blended fields and you don't want that, so you'll need to do the effects shots - and only the effects shots manually.

Darth Editous said:

After a bit of Avisynth magic:



DE

While that still does look good, you don't want to overdo it. It should basically look like it was done on 16mm as opposed to 35mm, because of the comp'ing done to it (and because that's how it always looked until the layers were digitally cleaned of film grain before being digitally composited for the SE).

They removed a lot more than just the "offsensive" vaseline blob.

It's smarter to modify the script to output to the resolution you want. When I ran my modified version of G-Force's script I had it output to a 1:1 pixel ratio instead of anamorphic, without going through further steps of scaling.

Any way you look at it it's bad DVD authoring. If the source is 60i then it could at the very least have been "1:1" on the NTSC DVD. Dunno how or why they cocked it up, but it seems riddiculious when you think about it. Even if the effects shots were done at 60i it isn't that hard to bring it back to the original 24fps. If absolutly necessary, just re-do the effects. Anyhow, I stand by what I said, it isn't too hard to remove field-blending using automated, so long as you isolate the "effects" shots. :)

Isolate the effects shots, and start from there. By the sound of it, those are the only ones that won't be peftect at the end? I haven't seen the source, but I'm sure that - minus the effect shots - it should be easy to remove field blending. This happens all the time in music videos, exactly the same things: 24fps --> 60i (pulldown) and effects mastered in 60i.

It should be relatively easy to remove the blended fields using an automated process, I've done it before.

You_Too said:

I don't know if this can be useful here, but here's a recreation I made a while ago for Harmy to use in his Despecialized V2.0: http://img846.imageshack.us/img846/8849/alta1080fixed.png

My references for this recreation were some frames Mike posted on his Legacy forum.

That itself is a recreation, no?

How about thanks for join the forum? Really guys! TBH I know that the audio tracks (or a subset of them) are on theatrical 35mm reels in the hands of some collector's, so I would say "well we already have them". Still, it might be useful if your version had discrepancies in it - which is unlikely - and beyond that I am of no help.

But thank you very much for posting and I hope you'll stay :)

skyjedi2005 said:

In Lowry's defense the final color grading was done before it was sent to them and fucked up by Lucas.  They were only given a 1080P master to work with its hard to recover detail that is not there, especially when you dvnr the hell out of a 1080 scan.  That was not even state of the art in 2004, despite the bs claims of the Lucas folks during the hype of the BD release.

Try that would have been state of the art circa 2001.

And would have been excusable for the dvd format because the sins of the dvnr would not show as bad.  To use that same master for the blu rays and not do a new 4k or 8k scan is laughable to say the least.

I couldn't agree more. Not that I personally have much if any love for the SE anyway. You look at all the previous releases - Laserdisc after Laserdisc it was a new Telecine all the time! Then, because Lucas is an arse, he refused to release the 1997 SE on DVD. In 2000, I think, he released the Special Editions on VHS and Laserdisc! How pathetic you don't even release it on the current technological medium. The DVD didn't come until 2004. The bluray didn't come until 2011. That's SEVEN full years later. Lucas has managed to make more changes to the thing, and only he can explain why, and yet he didn't do a new transfer. As for 8k scan, I think 4k would be well enough for these films, after all 1080p does reproduce "nearly all" of the resolution.

Noise Reduction is fine when done properly. Lucas on the other hand always wants far more than should be applied, he basically doesn't want to see evidence of his movie being shot on 35mm, he wants to be able to pretend it was shot in digital.

And here it is compared with the '06 and '04 DVD's.

negative1 said:

here's a better comparison of the shots

gout on the top, frame below:

 

in our shots, the lighting might be off,

there's a lot of grain, and there is no

color correction...

---------------------------------------------------

 

These shots look so terrible on film because they were comp'd too many times resulting in an increased graininess and a much softer picture. Hence why they look super bad when Lucas wanted them smoothed over in the 90's to match surrounding shots better. For the special edition I *believe* they went back to the original film elements before being comp'd and did it digitally. A long story short; I don't think this shot needs to or even should look too much better, the fact that you've got it pretty much how it originally looked is great I think!

BTW if you wanted to post a clip or two some time I'd love to have a play with it in AVS and see what could be done with it. :)

Wow great news. I've never seen this version, but I'd love to see it once it's upped! Any chance on seeing this upped on Demonoid? Last year I watched the Coppla restorations in glorious HD on my telle and showed them to a friend of mine who had never seen them before! He loved the movies, I'm sured he'd love to see this version!! I was really impressed by the BD quality (my TV cust through the grain like butter so that was no issue).

WOW. Does he actually believe half the crap that he says?

If you look at Blade Runner, it’s been cut sixteen ways from Sunday and there are all kinds of different versions of it. Star Wars, there’s basically one version — it just keeps getting improved a little bit as we move forward.

Well if you look at Ridley Scott he hates making changes to his films. He only did Alien "Director's Cut" because they told him they were making an extended version for the Quadrilogy DVD and he didn't want that. Unlike Lucas he doesn't get to be the producer as well as the director; Lucas has no argument because he's the producer so he has the final say on the final cut of the thatrical version! He only made the Blade Runner "Final Cut" because it's what fans wanted; and all the changes were respectful. You didn't see him trying to replace all the repeated elements like the neon signs and those stupid pillars that give away to a vgetran audience that all they did to make a "new street" was move around the objects on it! You don't suddenly have Zora shooting first because Harrison Ford isn't a "cold blooded killer". Oh and by the way, they released all 5 versions including the Workprint that was never originally intended for public distribution and only existed to the public through illegal trading of bootlegs!!!!

Plus, Scott has made soooo many more good movies than Lucas did.

Maybe he can improve Indy 4 for us by inserting imperial stormtroopers riding dewbacks?