Does anyone still actually care for Star Wars after Last Jedi?
Edit: I meant care for the sequel trilogy. My bad.
There are places on the internet with a lot of different vinyl rips of Beatles albums, I would wager that the proper ones are among them. I could be wrong though. Would love to give this a listen in any case. Personally I’d release in 96kHz/24bit. Even if you can’t hear the difference, at least it allows some more editing for those who want to do it. It’s supposed to be a kind of archive after all.
This is just an idea, I’m not working on this, as I have enough other stuff to do, but thought I’d just throw it out there, if only to remember it.
The idea is to structure all Marvel MCU movies into one single movie with individual chapters, but everything extremely abbreviated. It would have a narrator and would be narrated in the form of a father reading a story to his children before sleep, with each chapter in the book being maybe 10-20 minutes, and all leading together into the finale.
Character’s voices would either be kept, or replaced by the narrator’s voice, or a combination of both.
Each movie would effectively turn into a short story, basically. Each little story would serve to introduce a hero to the eager listeners.
This would require a few things:
The entire thing would be a few hours max, meant to be consumer over the course of a few days without too much time investment - each night before sleep. 😉
I’m not really a big MCU fan at all, and I dread rewatching any of the movies, mainly because (in my opinion) there is a big lot of filler and generally not much depth in it, but I think such an abbreviated condensed retelling might be charming and could also be made a little more consistent in terms of plotholes and such. Some story arcs could even be changed, since the voice is actually telling what’s happening, so one doesn’t necessarily need actual footage of an actor saying or explaining or doing something, the narrator can do it.
Anyway, just an idea. If anyone should want to steal the idea, feel free.
Sounds great in theory. Any progress on this?
I have to admit I’m not a fan of the picture with the lady in red, the contrast seems too extreme to me, although high contrast in general ofc isn’t a problem, but here a lot of highlights are blown out, and, to be honest, the red + black and white look is kind of an overused cliche at this point.
Just my 2 cents.
I’ve been recommended by a smart dude something that may be even better: ESRGAN/BasicSR. It’s an AI-upscaling algorithm just like Gigapixel, but you can train it with your own datasets, so if you have a representative dataset of what you’re trying to upscale (you need high res and low res reference for training), you can get really incredible results.
I’m currently playing around with it, and while training is very slow (we’re talking many days), it’s impressive what it can do imo.
I’ve tried some pre-trained ESRGAN upscalers, and I’d say it is a more natural looking upscale than ai gigapixel, but I’d still say ai gigapixel wins in the detail, even if it looks a bit too sharp. And I’d still say in terms of noise reduction ai gigapixel and waifu2x are better at removing compression artifacts. Of course waifu2x doesn’t try and add or recover detail, but simply a high resolution upscale that retains all the original detail.
All three are really good tbh, and all have their purposes. I’d say a mix of them is ideal.
See, that’s where your own training data comes into play. Gigapixel is trained on a generic dataset you know nothing about. Of course they have probably put lots of effort into a good curated dataset for a good product, more work arguably than people working for free have done for ESRGAN. Gigapixel may be good at noise reduction, but it’s not perfect, because it may not apply perfectly to whatever dataset you yourself have. To give an example, I’m currently training ESRGAN to upscale old roq game videos (Quake 3 engine). For that I took a video I have in high quality (Prores) and made roq encodes at similar bitrates to the game videos. Then I put that into the training. I’m not far into the training yet (only 2 days or so), but it seems promising. I tried a screenshot from one of those videos in Gigapixel and while it was sharp, it misinterpreted a lot of the artifacts as detail, and I think that with enough training, ESRGAN will be much smarter about upscaling this particular kind of source.
I’ve been recommended by a smart dude something that may be even better: ESRGAN/BasicSR. It’s an AI-upscaling algorithm just like Gigapixel, but you can train it with your own datasets, so if you have a representative dataset of what you’re trying to upscale (you need high res and low res reference for training), you can get really incredible results.
I’m currently playing around with it, and while training is very slow (we’re talking many days), it’s impressive what it can do imo.
Amazing, thank you! When you release the raw file, can you release it in Open Matte as well? I worked a bit on a scan of the German trailer and noticed there’s lots of additional image information on the top and bottom on many frames. Would be worth preserving imo. 😃
[removed it bc I’m stupid]
Great news! I’m gonna assume your print is Open Matte? Could get some extra picture.
TomArrow said:
x264 is a DCT-based codec, so everything always gets converted into the frequency domain on a per-block basis, which of course can introduce rounding errors in theory, especially at 8 bit. It’s possible they have ways to avoid this, but it’s something to keep in mind.Yes. This is why I’m not convinced that x264 is truly lossless.
Yeah, me neither. But it’s certainly good enough for almost any scenario imaginable, except maybe some scientific uses where it really needs to be bit-perfect.
Just a few notes about the (hypothetical) sharing of lossless masters, if it’s of interest to anybody:
crf 0 is the wrong way to do lossless x264 afaik, the correct way is -qp 0 (as crf is conceived to achieve variable bitrate with varying qp (quantization parameter) across blocks/frames. QP explicitly sets it to 0 globally.
x264 only encodes YUV values. A conversion from RGB will automatically lead to rounding loss and thus not be 100% lossless anymore (but still good enough practically). ProRes also works in YUV.
x264 by default encodes in 4:2:0. Compare to RGB, which naturally is 4:4:4. So you automatically lose half the color information. The solution is to encode in the 4:4:4 mode of x264, which has to be explicitly stated.
Normal x264 only has 8 bits, whereas ProRes has 10 bits (or even more with a proper Apple ProRes encoder). For sharing a master for proper color correction work, a higher bit depth would be highly beneficial. This can be achieved with the 10-bit version of x264.
Lagarith as well as normal lossless AVIs only support 8 bits as well, therefore are not useful for sharing lossless masters, especially not if further color work is to be done.
Lagarith has 3 modes afaik: RGB, RGBA and YV12. YV12 is a 4:2:0 mode and tends to produce significantly smaller files than RGB. RGBA includes alpha channel, should be irrelevant for most cases. RGB is default. The mode can be changed in the codec settings (for example in VirtualDub when selecting Lagarith).
x264 is a DCT-based codec, so everything always gets converted into the frequency domain on a per-block basis, which of course can introduce rounding errors in theory, especially at 8 bit. It’s possible they have ways to avoid this, but it’s something to keep in mind. Same applies to ProRes by the way, also DCT-based; but ProRes is not meant to be lossless to begin with, so that’s okay.
ProRes, unless you use the 4:4:4 mode, also is 4:2:2 and thus not ideal for anything that will require further color work imo.
Currently the only way (known to me) to have a truly lossless master with high bit depth is an RGB 16-bit TIFF/PNG/(some other formats) sequence or some exotic AVI codecs that support higher bit depths, but aren’t widely supported or maintained.
Of course it’s still possible that x264 it’s not 100% lossless, because it was never really made for it and who knows what it does to the signal behind the scenes, but some people have evidently achieved losslessness in x264, see here: https://stackoverflow.com/questions/6701805/h264-lossless-coding
Either way, it’s certainly good enough, if you use the 10-bit version, in addition to qp 0 and 4:4:4 color setting. But at that point the bitrates will quite likely even exceed those of ProRes (as I believe ProRes in High profile has the equivalent of qp=4 if I’m not mistaken, at least that’s what ffmpeg shows), and come at the cost of being several times more inefficient than ProRes when it comes to editing. Most editing software doesn’t even support importing x264 10 bit, so you would have to convert it back to ProRes or the likes anyway in order to edit it.
With that said, I think that if you drop the strict “lossless” requirement, we could share very high quality masters (comparable to ProRes) suitable for color grading & VFX if we just set the qp to values above 0. Again, with 10 bit x264 in 4:4:4 mode. As a starting value, I would probably try qp 4, but maybe it’s possible to go even higher. To test the quality of the result, you’ll need to do 16 bit screenshots (as PNG/TIFF don’t support 10bit), for example using AviSynth+ & VirtualDub, and then I typically load that stuff into Photoshop and play with the Levels to blow up dark areas to see if any artifacts become noticeable.
I think this way there could be a notable size improvement over ProRes, which would help with transferring. The recipient could convert back to ProRes then for his own work.
Probably wanted it to look like the shadow was showing some color from the little lights above lol.
I can confirm the iTunes copy doesn’t have shifting aspect ratio (I have a copy). Can’t be shifting, because it’s cropped to the OAR.
What is so interesting about this version? Is it the new master, like the iTunes version?
That’s awesome man! Wish you the best for this project!
Awesome. I like these raw looking scans. It’s basically like the Grindhouse of the ROTJ and ESB scans, correct?
Is it available for sharing? I am not familiar with the etiquette here, pardon my ignorance.
Small project of mine.
I realize the Dreamastered version is already GOUT-synced, but I’m a fan of the untouched colors.
Naturally, syncing the already encoded video to the GOUT and reencoding it would degrade the quality unnecessarily, so I used this project as my first “proof of concept” for syncing it almost losslessly using TMPGEnc Smart Renderer 5. With the help of this tool, I was able to reencode the video only at the split points/cuts (around 20 across the movie), and end up with a resulting video stream that is virtually identical to the original release.
For syncing I used Chewtobacca’s script from the GOUT-sync thread and “adapted” it to the interface of TMPGEnc Smart Renderer 5.
Final video is 100% in sync with the Despecialized Edition, so I was able to mux in all the audio tracks from there into this. DTS-HD MA tracks were losslessly transcoded to FLAC, to reduce bloat.
The original optical 35mm PCM track from the scan was automatically synced as well with missing frames filled with silence. Was also included as FLAC.
As a little gimmick, I included a Dolby Headphone track of the 5.1 mix by Hairy_Hen.
All credit goes to the people who have created the content originally. Thank you, guys!
This little project and the full description is available at Blutopia.
Just found this a while ago, just wanted to share this if anyone wants to make an edit such as this. It shows an alternate ending of the film, wherein Luke doesn’t die.
Ok but what happens to EpIX and the highly expected Luke’s Force ghost ?
Elevator scene is very good. I’ll add that to my “TLJ: wishcutlist to help me find the best fanedit available” 😃
“Luke, why are you translucent and blue?”
“Oh that. That … umm … okay, I admit it. I’m not really here. I’m sitting in my hut drinking walrus tiddy milk and I’m basically just here for moral support.”
Okay, I’ve been having this idea in my head forever, but ultimately I figure I probably just don’t like the two movies enough to actually give it a try, so here’s the idea, just to preserve it: Combine the two movies into a single story, rearranging all elements that are available. It may not be 100% coherent, but as long as it has enough “flow” I think it might work.
The following is a rough outline of how I imagine it. Depending on how well that works out, some parts may (will) need to be rearranged to make it better.
In detail, here’s how I imagine the flow to go:
Lots of scenes still missing that can hopefully be spliced in here and there. The main aim of this whole thing is A) to get rid of the ridiculous and boring chase plotline in TLJ. and B) to explain Rey’s proficiency with the force through her training with Luke. and C) to try something insane.
Not sure if this would remotely work, but it just might be a cool experiment.
Based on the YouTube clips, I will not be using any of the deleted scenes footage for my edit. Most of them are superfluous or silly. The only clip that I could see being salvaged is Luke mourning Han’s death. However, that would mean going from a shot with Rey and Chewie in it to a shot of him mourning. Either that, or I would need to find a scene later on where I can cut to Luke alone in his hut.
On a side note, I never thought I’d see Rose biting Hux’s finger or Tom Hardy as a stormtrooper spanking Finn.
The funny thing is, in the commentary Rian says that they were able to take away the shot of Rose biting the finger “because she is already strong enough of a character to not need this scene”. Rian thinks this scene makes her look strong. Way to go lol!
Replace all women with Nicolas Cage via that app?
Lmao! That stuff is hilarious. Have you seen the Titanic one?
Wacko idea combining both movies:
Why not let Rey go to training before the whole Starkiller thingy is even introduced. Make it a mellow entry to the story where the villain (Kylo) still stays mysterious. She arrives and starts her training after retrieving the lightsaber and having been confronted with Kylo once.
Fast forward 5 years. First order has built starkiller base. Rey is in full training mode. Now, somehow mix together the ship chase and the starkiller thing and the planet being destroyed and rey and kylo having their first lightsaber fight and things suddenly make much more sense. She now has had the training to justify holding up against Kylo a little.
Not sure about the details (would like to include Snoke etc. as well) but that can still be worked out. Lots of stuff would obviously fall away.
Exactly, Home Alone 3 has little to do with the first two parts. Different kid actor etc. and yeah it’s not very good.
BOOK OF ELI RE-COLORIZED
DESCRIPTION
The desaturated look of Book of Eli is absolutely not my thing, so I decided to push the colour out a little more.
Played around for a while to get a good compromise with a blanket 3D LUT until I was satisfied.
Here’s a screenshot comparison (also included in subfolder of the files): http://screenshotcomparison.com/comparison/128273
Since the saturation is so low in the original, there’s lots of obvious chroma blocking when raising the saturation, which is why I applied a grainplate for each RGB channel separately. This results in some extreme grain/noise in some areas where there is strong chroma contrast, which I kinda like. Looks better than the blocking, anyway.
Audio is the lossless 5.1 DTS-HD MA audio from the Blu Ray transcoded to FLAC 5.1. On top, you get a FLAC 2.0 Dolby Headphone track rendered from the lossless 5.1 soundtrack, with amplification set so that there is no clipping. This track basically simulates a 5.1 surround setup on your headphones.
ENCODING & FILES FOR YOUR OWN RE-ENCODE
Final encode is done by applying the 3D LUT through ffmpeg in 16-bit, then piping into and encoding with 10-bit x264. If someone wants a blu-ray compatible encode, I have included the .avs-file, grainplate, 3D LUT and a .bat script using ffmpeg for the LUT application and piping into x264 (that took some tinkering to get it to work), so you can just make your own version with your own encoding settings. I personally don’t have a Blu Ray player, so I don’t care about compatibility much.
I know that not all the choices I made with the 3D LUT may be to the liking of everybody, so feel free to adjust the 3D LUT any way you like for your own encode to make it more to your liking. An example of this may be the intense red tones, which I chose because I liked the look and also because there are some red elements in less saturated scenes which otherwise don’t pop nicely. Another may be the dark yellow patches on the skies, which were originally green and brighter. I like it this way, but maybe you don’t. Partly the choices were also dictated by scenes that would have looked bad if I had done it differently, but you can figure that out for yourself.
TRAILER
Also included is a trailer which I created to give a sneak peak at the colors and picture quality. For this I recreated one of the movie’s trailers (I think it’s called Trailer #2) with scenes from the Blu Ray, then applied the same processing and LUT as for the movie. Shots that were trailer-exclusive were first regraded to match the Blu Ray as best as possible, for example by creating a conversion LUT for other shots of the same scene that do exist on the Blu Ray, using DrDre’s ColorMatch software, then also had the same final LUT applied.
Since the best quality trailer source I could find had an Apple logo, that one was patched out with a Youtube trailer. That one had slightly different color, but I was able to fix that with two passes of the ColourLike plugin (Planar & RGB).
The best audio I could find was a 318kbps 2.0 English DTS track from the German Blu Ray (despite the video only being in letterboxed SD there). I retimed it from 25 fps to 23.976 fps by interpreting sample rate as 48000*23.976/25, then resampling the result to 48000 with a good SRC. Normalizing was applied before resampling. Result was saved as 24-bit 2-channel FLAC. That one I ran through the Dolby SDU4 Hardware decoder using a good audio interface, from which I created a 5.1 FLAC track (BC minus 3dB, split to SL and SR, LFE created from low frequencies of all resulting channels, without removing them from those). From that 5.1 I also created a Dolby Headphone track, which is a FLAC stereo.
So the trailer comes with 3 audio tracks:
Video encoding process is the same as for the main video.
This was a funny undertaking in general. Many of the scenes in the trailer were unfinished, missing matte paintings and VFX. For example the original trailer was missing an error arrow piercing someone’s throat and a head being chopped off. The remade one has all these … beautiful things.
I did not recreate the title cards (is that what they’re called?), so their quality is as good as the original trailer allows.
RELEASE
This project was released on Blutopia. As far as I am concerned, feel free to share it anywhere you like.
What does the restoration work consist of? How much work is it? Is it possible without PixelFarm products?