neebis said:I'm not sure where the retail disc layer break is, but I ended up putting mine right after the scene where Luke is doing saber practice with the remote on the Falcon, where it cuts to a scene on the Death Star behind a conference table. On my set-top player, it cuts to the Death Star shot, then pauses, then the door opens and the scene continues. It was the best I could find given the size of my video, because there's no motion and no dialog or obvious audio to be clipped at the pause.
FWIW, the retail layer-break on SW is right after Tarkin says "Terminate her! Immediately!", on ESB I think it's during a shot of Falcon's cockpit window before the Mynock attaches itself, in ROTJ it's right as we see Wicket's feet appear.
Mielr said:FWIW, the retail layer-break on SW is right after Tarkin says "Terminate her! Immediately!", on ESB I think it's during a shot of Falcon's cockpit window before the Mynock attaches itself, in ROTJ it's right as we see Wicket's feet appear.
Thanks for the info. I think my video might be too big to match that spot.
neebis said:I ended up putting mine right after the scene where Luke is doing saber practice with the remote on the Falcon, where it cuts to a scene on the Death Star behind a conference table. On my set-top player, it cuts to the Death Star shot, then pauses, then the door opens and the scene continues. It was the best I could find given the size of my video, because there's no motion and no dialog or obvious audio to be clipped at the pause.
The scene you've chosen corresponds to a cinematic reel change.
It's possible theater audiences also experienced a pause at that spot, depending on the skill level of the projectionist.
However, in practice you must take into account the “fuckwit factor”. Just talk to Darth Mallwalker…
-Moth3r
Darth Mallwalker said:The scene you've chosen corresponds to a cinematic reel change.
It's possible theater audiences also experienced a pause at that spot, depending on the skill level of the projectionist.
That's a cool coincidence. I guess I picked a decent spot. (Or maybe I remembered it from years ago?) Thanks for pointing that out.
ESB fixes incorporated into the script (see page 1)
-G
g-force said:ESB fixes incorporated into the script (see page 1)
-G
Much better,it is gone.
But I am not going to make another lossless avi file from the new script until you know for sure it is "Final" LOL.
I just got an external WD 1 TB hard drive so I have plenty of room to store the lossless avi's from the "Take 2" JSC set and the "Take 2" 97 SE set along with the v4 GOUT script and whenever there is a final to this one.
Thanks G
Dark Jedi, forgive me if this has been mentioned before, but what kind of powerhouse machine do you have to be able to do encodes so quickly? Or did you only apply it to the clip in question?
A Goon in a Gaggle of 'em
bkev said:Dark Jedi, forgive me if this has been mentioned before, but what kind of powerhouse machine do you have to be able to do encodes so quickly? Or did you only apply it to the clip in question?
Personally, I use a Trim() command at the top of the script to test short clips from different sections of the film. You can also just load it up in Virtualdub and click different points along the timeline to see the output on various frames.
bkev said:Dark Jedi, forgive me if this has been mentioned before, but what kind of powerhouse machine do you have to be able to do encodes so quickly? Or did you only apply it to the clip in question?
I use the trim setting to test a particular area before doing the whole thing,but to do the whole movie it takes me about 18-20 hours(lossless avi)then about 2 hours after that to encode a 6 pass CCE SP @ 6500.
I have an INTEL Quad Core @ 3.0 GHZ,8 GB's RAM(I know WinXP Pro only see's 4 but 4 more are there LOL),Dual Nvidia GeForce 9800 GTX+ Video cards running in SLI mode,Dual 500 GB hard drives,and now an external 1 TB hard drive.
dark_jedi said:
But I am not going to make another lossless avi file from the new script until you know for sure it is "Final" LOL.
Like that's gonna happen... ;)
No promises, but I really do think I'm done for a while though.
-G
dark_jedi said:
I have an INTEL Quad Core @ 3.0 GHZ,8 GB's RAM(I know WinXP Pro only see's 4 but 4 more are there LOL),Dual Nvidia GeForce 9800 GTX+ Video cards running in SLI mode,Dual 500 GB hard drives,and now an external 1 TB hard drive.
Beautiful!
-G
My PC is actually geared for gaming,but I think it is going to be a loooong time before Diablo III comes out.
Hey Everyone,
In the last one day, I've read through this topic - all the 29 pages, yes. First of all, it's very nice to see someone who knows what he does, and can go near the limits. Thank you g-force for your commitment and hard work - and everyone else's work you based your script on.
Some suggestions for the script:
1. Levels() causes clipping in the video, burning out some details in the brights spots (perhaps even in the not-so-bright spots). It'd be better to draw an arbitrary curve and apply that to the video. It takes a while to draw the perfect curve, but you'll bring back some of the details you've lost before with the brightening. For instance, it'll very likely make that light grey smoke better.
You can draw the curve with BugsBunny's Gradation curves filter in "RGB weighted" mode, save it to an .AMP file, then feed it in AviSynth with GiCoCu like the following:
MergeChroma(GiCoCU("GOUT_levels_curve.amp",photoshop=true)
It's highly recommended to save the curve in .ACV format too to make further editing possible. Be careful with the curve though, make sure you don't brighten up anything more than it is neccessary. With a little rougher curve than the optimum you'll loose a lot of fine details. Check (and compare) the result on people's face in particular.
2. There's some halos around the edges; try YAHR() to reduce them. If you find YAHR's strength a bit too high (losing details), try increase sharpening first and stop right before the halos appear again. If you don't like to result, you may want to use supersampling. Anyway, you'll see what YAHR() does and if it improves your video or not. Probably yes.
3. From the screencaps I have seen there seem to be a very slight chroma problem. Try MergeChroma(aWarpSharp(depth=32, blurlevel=1, thresh=0.5, cm=1)). (Values are not set for this video.) Merge() instead of MergeChroma() can help too (perhaps combining too would be the best), although that causes small deformations. Still worth a try.
4. I don't have the GOUT discs at hand now, so I can't check the smearing. If it's a smearing similar to color bleeding, YAHR() and other aWarpSharp tricks above may help. If it's like fieldblending (that means the trails are coming from the neighbour frames - the smearing led to four-eyed startroopers) then there are a few filters to remove this. But I don't know how to use these, so it's really just a guess. I'm just hoping you'll find some solution for that. It'd be too bad if the smearing would hold back this being the best source for having the untampered trilogy.
zee944 said:If it's a smearing similar to color bleeding, YAHR() and other aWarpSharp tricks above may help. If it's like fieldblending (that means the trails are coming from the neighbour frames - the smearing led to four-eyed startroopers) then there are a few filters to remove this. But I don't know how to use these, so it's really just a guess. I'm just hoping you'll find some solution for that. It'd be too bad if the smearing would hold back this being the best source for having the untampered trilogy.
The smearing comes from a "dirt concealment" method that was used when the DefCol laserdisc masters were created back in '92. It had something to do with using surrounding frames to hide dirt specks, pinholes, etc. Unfortunately, it probably created more problems than it fixed.
Still, they ended up being the highest quality OOT laserdiscs available, and they used those same masters for the GOUT DVDs.
zee944 said:In the last one day, I've read through this topic - all the 29 pages, yes. First of all, it's very nice to see someone who knows what he does, and can go near the limits. Thank you g-force for your commitment and hard work - and everyone else's work you based your script on.
That's quite some feat. Thanks for the kind words.
zee944 said:1. Levels() causes clipping in the video, burning out some details in the brights spots (perhaps even in the not-so-bright spots).
Levels, if used incorrectly to increase brightness, will cause clipping of the whites. I am not using levels in this way.
zee944 said:2. There's some halos around the edges; try YAHR() to reduce them. If you find YAHR's strength a bit too high (losing details), try increase sharpening first and stop right before the halos appear again. If you don't like to result, you may want to use supersampling. Anyway, you'll see what YAHR() does and if it improves your video or not. Probably yes.
Already tried most of what you have proposed. The problem with YAHR (and to some extent even better de-haloers) is that they kill corners and small lines, and leave the video with a soapy, artificial look. One of Didee's poorest products IMHO. I'd be interested in looking at some specific examples if you want to actually try them on the GOUT and post what configuration works for you.
zee944 said:3. From the screencaps I have seen there seem to be a very slight chroma problem.
Could you be more specific about what chroma problems you see?
zee944 said:If it's a smearing similar to color bleeding
No, it's not. See Mielr's explanation above.
zee944 said:If it's like fieldblending (that means the trails are coming from the neighbour frames - the smearing led to four-eyed startroopers)
No, it's not even close to that either. And that is not what led to the four-eyed "STARtroopers". ;)
The smearing is from a form of noise reduction that is VERY similar to a weighted TemporalSoften(), except using only previous frames, and not having much if any scene change thresholding. This will NEVER (yes, you can quote me on this) be removed from an automatic filtering technique. The major problem is that the filter applied was extremely non-linear. It affected dark scenes more, and can have affects from between 0 and 5+ frames previous. Yes, you read that correcly, I have seen frames that have remnants from at least 5 of the previous frames.
-G
g-force said:
zee944 said:2. There's some halos around the edges; try YAHR() to reduce them. If you find YAHR's strength a bit too high (losing details), try increase sharpening first and stop right before the halos appear again. If you don't like to result, you may want to use supersampling. Anyway, you'll see what YAHR() does and if it improves your video or not. Probably yes.
Already tried most of what you have proposed. The problem with YAHR (and to some extent even better de-haloers) is that they kill corners and small lines, and leave the video with a soapy, artificial look. One of Didee's poorest products IMHO. I'd be interested in looking at some specific examples if you want to actually try them on the GOUT and post what configuration works for you.
Well, now I've tried EVERYTHING you suggested. Yahr by itself rounds all corners. If you run it on the opening scroll, it becomes almost unreadable. It also kills a lot of stars. Anything that ruins the image this much is not useable.
Tried (apprehensively) with pre-sharpening. Pre-sharpening is usually a bad idea, better to post contra sharpen. Tried both, but neither improved the result enough.
There was one method using YAHR that had some promise.
source = last
med = MT_luts(source,source,mode="med",pixels="0 1 0 0 0 -1",expr="y")
Median1(source,med,source.YAHR(),chroma="copy first")
Killed the halos, with much less artifacting. BUT, the crawl was still too aweful. If it kills the crawl, it'll kill something else that I care about too. No go.
Tried using supersampling. Doubled the image size and it still destroyed some detail, but it also didn't remove halos at all.
-G
I also tried YAHR and some other halo reducers before this latest discussion and I couldn't figure out any settings or combination of filters that successfully reduced halos without degrading the rest of the image (loss of detail, blurring, watercolor effect and re-sharpening or pre-sharpening didn't help). The source tape is just such poor quality it's tough to fix one thing without causing other problems.
OK I am going to do new versions of these movies so I am running some tests,here are some screens,and also the gold trim behind Solo is shaking again,it was fixed but it is back,so I was wondering is there any way to just add the "magic tree ESB" fix to the version 4 script?
here are the samples
video sample
http://www.sendspace.com/file/yygneg
and before anyone comments on the sub placement,yes I moved them,I want them in the video now like the HD files I have,so now when I do other GOUT projects I do not have to worry about adding them back in again,plus I like how it looked on the HD files,hell if I could get that font,I would use it instead of this one,but not sure what font was used.
looking at these samples again,to me it seems v4 looks better,these look blurry,like a little out of focus,so it would be nice to have the "magic tree ESB" fix for version 4,if you need version 4 I can PM it to you.(that is the version I used to make the first set)
g-force said:Levels, if used incorrectly to increase brightness, will cause clipping of the whites. I am not using levels in this way.
You're right. Doesn't it affect the chroma though? You're using it in YUV colorspace.
g-force said:One of Didee's poorest products IMHO. I'd be interested in looking at some specific examples if you want to actually try them on the GOUT and post what configuration works for you.
No need! I trust your observations, it was just a tip. YAHR() worked very well for me on a few blurry videos, never tried with good quality sources like this. I couldn't try it out as I didn't have the DVDs at hand.
g-force said:Could you be more specific about what chroma problems you see?
Forget it. After a closer look, it was probably only a sign of the edge ehancement.
Oh well. I thought my post will be useful, but not at all. You've passed my ideas long time ago.
And I'd like to ask something:
g-force said:I've also found out how to repair a TON of detail by collecting pixels from the same object in different frames. The results are a little preliminary (like as of before work this morning) but the results so far are breathtaking (I literally couldn't believe my eyes!).
What did you find out? Please share :)
G-force-
I watched ROTJ the other nite and it looks fantastic. I watched it all the way through and I didn't spot any problems (not saying there aren't any- I just didn't notice anything funky like I did in the ESB magic tree scene).
dark_jedi says that the 'gold trim' problem in SW has reappeared with the latest tweak (is this V. 5 or 6? I've totally lost track. LOL.) Is there a simple way to fix this without affecting other scenes (like you did on ESB)?
Thanks again for all the work! These versions are destined to be definitive (they are for me, already). :-)
Why the new versions Jedi? Is the script that much improved?
I don't think there is a new Jedi script- but I think DJ is planning on moving the subtitles up for the next version.
Yes all I want is the fix for the "magic tree" scene for the version 4 script,I do not like v6.64 so I will not make a new version to that,but the ESB problem needs to be fixed,and yes I am moving the subs up into the very bottom portion of the movie on SW and ROTJ,so to make a long story short,I will be doing all 3 over 1 more time.
V.4 just kills too much FINE detail. Yeah, it sharpens medium to low frequency detail, but just kills all the fine stuff. It also has too much artifacting on quickly moving objects.
Might want to wait until I have my de-ringing script ready. It's looking good so far, just want to be sure before I make a fool of myself (yet again)
But, that'll look more like V.6 anyway. If you want to fix V.4, just replace the entire "global motion" stage with the new one.
-G
zee944 said:
Oh well. I thought my post will be useful, but not at all.
On the contrary, You got me thinking about de-haloing and de-ringing, and how nobody (in general) has done this correctly yet. It turns out that it's really a deconvolution problem, and that is proving difficult but seems do-able.
-G
