Would need to see more moving footage but it appears to still have less issues than any print would've had by week 2 of it's theatrical run. And given what you're starting with - this is pretty amazing.
Agree - it looks very nice. Some of the ghosting is made slightly worse, but the improved detail looks worth it.
"Close the blast doors!"
Puggo’s website | Rescuing Star Wars
Danfun128, yes the 1993 LD masters are 480i telecine's with the exception of ROTJ which is a 576i telecine (from memory I can't remember if the NTSC version of ROTJ was converted from the PAL tape or if there was a separate 480i telecine used for it). Back in 1993 it was more-economical to scan the film for each video format than to convert between the two. It's also what produced the horrible aliasing (happens when each filed isn't precisely aligned).
There are several ways to upscale video. To answer your other question, in order to do a clean upscale you need to fully degrain the source. You can put the grain back in when you're done, or generate film grain that is a close approximation to what was in the original source. As the source is DVD, there's no accurate way for the grain to be represented without being severely affected by the MPEG2 compression - so only a small part of a DVD image can ever hold the grain, and you can see this effect often in DVDs where grain seems to appear in static areas of the image and disappear over the more complicated and moving areas. Degraining is a delicate process - you want to remove as much as possible while removing as little fine detail as possible. It doesn't matter if you lose a little bit of fine detail, it's all about the balance.
[ Scanning stuff since 2015 ]
I stand corrected. I assumed there was a high definition master of the GOUT. So, I guess this is approximately what a high definition master would have looked like back in 1993.
Here is the first video sample of Star Wars SRV3:
This is the low res source material:
https://drive.google.com/file/d/0B8_LYKyZDiajVG1UWGhfcTMyTkU/view?usp=sharing
This is the high res reconstruction:
https://drive.google.com/file/d/0B8_LYKyZDiajRENkTjdKTGFMcG8/view?usp=sharing
For full quality just download the samples.
It's not Harmy's Despecialized Edition, but considering it's based on a single low resolution source of rather poor quality, I think it looks pretty good.
Here's a quick example of a "proper" upscale. I did this in 5 minutes by modifying my TPM upscale script. It is not fully optimised for the source. It is a somewhat complicated script, but it basically uses eedi2 and QTGMC for deinterlacing, then upscales by a factor of 4 using spline36resize, then downscales to the output resolution via ResampleHQ. There's some noise in this frame for some reason, it may be a bad DVD rip I don't know I can't be bothered re-ripping the DVD right now, and it's based on the NTSC disc.
http://screenshotcomparison.com/comparison/122056
Yours has more micro detail, but also a lot more noise overall.
[ Scanning stuff since 2015 ]
@ RU.08
Although your upscale is much smoother and certainly removes many of the artifacts in present in the source, it introduces an unrealistic smoothness to the frame. In that sense it literally is a smoothed and cleaned up low resolution image on steriods. There is none of the detail or structure present that is native to high resolution images. In a sense you eliminate both the gain and the cost of my methodology, but essentially introduce a different cost. I guess it's a question of preference. Personally, I prefer mine (big surprise), but I can understand why you would prefer yours. That's why I would indeed describe yours as a "proper" upscale, but mine as a "proper" high res reconstruction.
However to get a better sense of the quality perception of the upscale/reconstruction, it would be interesting to see yours in motion and compare it to mine. I would say take the low res sample I put up and upscale it with your method for a proper comparison.
@ RU.08
A closer inspection of our methods also reveals a large amount of distortion in your upscale, compared to the Avisynth Spline64Resize. Just look at Vader's buttons. Apparently, the algorithm's desire to simultaneously sharpen edges, while smoothing noise, causes it to take liberties with the shapes in the image.
Avisynth Spline64Resize vs "proper' upscale (zoomed 3x)
http://screenshotcomparison.com/comparison/122072
Avisynth Spline64Resize vs super resolution v3 (zoomed 3x)
That's the problem with deinterlacing using eedi2. Here's a better version (well I think it's better):
http://www.screenshotcomparison.com/comparison/122076
It's still degrained, but this time I left the residual grain in. No distortion on Vader's chest-plate this time.
And by proper upscale I mean a more complete upscale. You wouldn't use just spline64resize to upscale video to HD.
[ Scanning stuff since 2015 ]
@ RU.08
The last one looks much better indeed. Yours certainly looks cleaner with less artifacts. It certainly is a lot better than a simple Spline64Resize. Compared to my method, the downside is that there's much less detail. Personally I'm not a big fan of sacrificing detail to gain a smooth image, but that's just a matter of taste.
My goal is to obtain a high resolution version of the GOUT, warts and all, so I've accepted that the final product will have excessive grain, and a number of other flaws inherent to the GOUT.
I'm aware that for many of you who are expecting a completely restored original trilogy, my grain filled high definition GOUT isn't good enough, but it's what I set out to do. As far as GOUT based upscales go, at least there's now another choice on the menu. Mine is a little rough around the edges, but I do believe it's the most detailed GOUT upscale to date, and I guess that may be of interest to some.
If you're using QTGMC, always use it after the upscaling itself and use this command:
QTGMC(Preset="Placebo", Edimode="EEDI2").SelectEven()
If you use anything lower than placebo, it'll be a blurry mess. Placebo is extremely slow with high res material but the setting with the sharpest results.
Star Wars SRV3 (crappy name I know, any suggestions?) will probably be finished by next week. I will make it available to anyone who's interested, but I've got little experience posting stuff on newsgroups and such, so some guidance would be much appreciated.
Well as I mentioned before, grain is not retained by DVD to begin with. Take the vader and leia image we've been looking at as an example. On Leia's top there's quite a bit of grain, but over on Vader and the ceiling lights and the walls there isn't. DVD simply can't retain it, it's just not possible for MPEG2 video to do it. In any image you will only ever have the grain of the source partially represented. And remember the way that telecines scan mean that what you get is not necessarily a true representation of the finer grain structure on the film.
But there's nothing wrong with grain, it's the noise that's been introduced with your system that I dislike - all that haloing. And that isn't in the source, it's introduced. Vader's chest plate in your version looks great, if not for all that extra noise that comes with it.
[ Scanning stuff since 2015 ]
Are you saying that despite the higher resolution, the defcol/faces handle grain better?
Nobody sang The Bunny Song in years…
Not really no, they introduce analogue noise. Just that the grain structure on the digital tape is pretty much gone when MPEG2 compression kicks in. Have a look at ESB -1 grindhouse and compare it to the GOUT" DVD. You'll see that the grain covers the whole of the image, and it dances, whereas the DVD has some of it only in part of the picture because the rest is smoothed out with MPEG2 compression, and much of it doesn't dance like it should. DVD tends to present this unnaturally static grain.
[ Scanning stuff since 2015 ]
@ RU.08
I understand, the problem is I don't find the loss of detail to be acceptable. You don't like the noise enhancement, that's fine. The price of reconstructing detail is that the "detail" in the noise is also enhanced (the haloing is actually part of the source, but is enhanced). However, the cost of your method is equally large, but you accept it, because it's part of the source. To me that's not good enough. The reference should be the high resolution source (in this case the print the GOUT is based on). The focus is put on undesired noise enhancement. However, to me what was lost in creating the GOUT dvd's is just as important. I've provided a way of regaining to a large extend what was lost, at what I think is an acceptable price. As I said, it's a matter of taste. Your upscale has less noise, but lacks detail. Mine has detail, but has noise.
Inspired by the discussion with RU.08 I decided try to reduce the noise. It appears that due to the high resolution of the upscale, much of the noise inherent to the upscaling method can relatively easily be removed using an advanced filter, without a significant detail reduction. Here are some comparisons between the noisy and denoised SRV3:
http://screenshotcomparison.com/comparison/122096
http://screenshotcomparison.com/comparison/122097
http://screenshotcomparison.com/comparison/122098
Here are the comparisons with the Avisynth Spline64Resize:
http://screenshotcomparison.com/comparison/122116
http://screenshotcomparison.com/comparison/122117
http://screenshotcomparison.com/comparison/122118
The noise is now generally less than for the Avisynth Spline64Resize and less than for the source material.
Here's a denoised video sample:
https://drive.google.com/file/d/0B8_LYKyZDiajY1YwNGFHZmdEZUU/view?usp=sharing
The detail is retained, but the cost has been reduced. Thanks to RU.08 for keeping on the pressure. ;-)
@ Laserdisc Man
Here's the NTSC sample for MagicUp:
https://drive.google.com/file/d/0B8_LYKyZDiajQzZtN1BTbHFOMjQ/view?usp=sharing
DrDre said:
Inspired by the discussion with RU.08 I decided try to reduce the noise. It appears that due to the high resolution of the upscale, much of the noise inherent to the upscaling method can relatively easily be removed using an advanced filter, without a significant detail reduction. Here are some comparisons between the noisy and denoised SRV3:
http://screenshotcomparison.com/comparison/122096
http://screenshotcomparison.com/comparison/122097
http://screenshotcomparison.com/comparison/122098
Here are the comparisons with the Avisynth Spline64Resize:
http://screenshotcomparison.com/comparison/122116
http://screenshotcomparison.com/comparison/122117
http://screenshotcomparison.com/comparison/122118
The noise is now generally less than for the Avisynth Spline64Resize and less than for the source material.
Here's a denoised video sample:
https://drive.google.com/file/d/0B8_LYKyZDiajY1YwNGFHZmdEZUU/view?usp=sharing
The detail is retained, but the cost has been reduced. Thanks to RU.08 for keeping on the pressure. ;-)
These are some great comparisons. I was concerned about the 'double haloing' effect that the noisy SRV3 had on the source, but the denoised version does seem to minimize or eliminate that. Could you post similar comparisons for the shot of Leia on the Tantive for example, I'm really interested in how this changes the distortion in the eyes of characters, since the previous one tended to add artifacts in the iris especially.
You probably don’t recognize me because of the red arm.
Episode 9 Rewrite, The Starlight Project (Released!) and ANH Technicolor Project (Released!)
@ NeverarGreat
The complete movie is being processed as we speak. I decided to go for slightly less denoising. The Tantive IV scenes had very little noise compared to the source, but in close ups details like facial structure were somewhat smoothed away. It did look very clean though, so maybe some of you will prefer that version. The good thing about all the added detail, is that you still have a good amount of detail compared to a standard upscale even if you denoise rather strongly. So it doesn't have that cartoony look that many of the denoised GOUT upscales have. I will post comparisons for both versions later today.
This project looks very interesting. Are you looking into doing anything about the anti-aliasing? The jaggies are one of the biggest flaws of the GOUT for me.
You know of the rebellion against the Empire?
I suspect not much can be done with the AA. That is inherent to the DVD, and probably no amount of work can circumvent it.
The transfer to the LaserDisc is where almost all the problems lie. Many of them could have been addressed when the transfer to DVD was made.
This will be by far, the best the GOUT will ever look. Only so much you can do with this material. The results are quite amazing, I am stunned this much detail is actually there.
Got the NTSC DVD of EP4 as source; lot better than PAL DVD (and also a bit more image on the sides); for my latest test, I cleaned up the source, upscaled it with two different upscale filtes - one called MagicUpPlus, the other called MagicUpSR, based on SuperResolution technique - and merged both 50/50; left the grain plate out this time.
SuperResolution V3 Noise Reduction Vs BasiClean + MagicUp merge (Plus & SR):
http://screenshotcomparison.com/comparison/122268
Can't say which is the best, because one has something better than the other and vice versa, so I decided to put each against the HD version - taken from HDTV, cropped and resized - note that the images don't match, but this is beyond the scope of the comparison...
BasiClean + MagicUp merge (Plus & SR) Vs HDTV:
http://screenshotcomparison.com/comparison/122269
SuperResolution V3 Noise Reduction Vs HDTV:
http://screenshotcomparison.com/comparison/122270
Now, it's you turn... pick your poison! (^^,)
Sadly my projects are lost due to an HDD crash… 😦 | [Fundamental Collection] thread | blog.spoRv.com | fan preservation forum: fanres.com
Wow, that's quite impressive.
A small video sample would be the best comparison between the two.
Is the HDTV capture of the 2004 DVD by chance? Because I heard there are also broadcasts of the 1997 versions as well.
Thanks DrDre! As I wrote before, this thread gave me several ideas to improve the quality of my upscale filter; test clip will be ready soon - which shot would you like to see? Frame numbers will help a lot!
About HDTV: yes, it's the 2004 version, as I'm pretty sure there was no HDTV broadcast of 1997 SE, but only DTV...
Sadly my projects are lost due to an HDD crash… 😦 | [Fundamental Collection] thread | blog.spoRv.com | fan preservation forum: fanres.com