... can we write a new petition? The current one is great, but I think it'd be more effective to address LFL (rather than GL himself) and argue that, among other things, a proper restoration of the OT would make good economic sense.
Does anyone agree with me?
I may have made some progress toward correcting the over-represented reds: http://originaltrilogy.com/forum/topic.cfm/Three-strip-Technicolor-Please-help/post/598910/#TopicPost598910
OK, so GOUT has over-represented reds. Challenge accepted!
Enter YellowMagic™ (a joke on Laserman's BlackMagic and _,,,^..^,,,_'s WhiteMagic).
I'm not claiming that these colors are correct, but I believe this may be right approach to de-emphasizing the reds in GOUT.
gout = gout.Tweak(cont=1.1, sat=1.3).ConvertToRGB24(matrix="Rec601")
gout.Overlay(gout.BlankClip(color=$D29628), mask=gout.ShowRed(), opacity=0.1, mode="lighten")
The $D29628 color was chosen because (1) it's yellowish, (2) it's close to -1's crawl text color, and (3) I didn't know what else to choose.
Pay close attention to skin-tones. Also, note how high I'm setting the saturation. This approach works for even higher saturation values.
Awesome post, msycamore!
I know that GOUT's master was prepared for '90s CRTs.
My problem with GOUT is that my primary TV is a 36" GE CRT TV (148lb.!) and GOUT still looks terrible on it while playing.
The first step is to calibrate the video monitor and the audio playback system. THX engineers use a Photo Research Spectra-Radiometer to carefully calibrate the black level, white level, greyscale, and color balance of the video monitor being used in the transfer. The Spectra-Radiometer measures the wavelengths of the light coming from the video screen and can display the actual spectrum on a computer. This gives the engineer valuable information on color and greyscale as well as on the actual purity of the phosphors being used in the monitor. The purpose of this calibration is to ensure that decisions made by the filmmaker on the color of the transfer are done under correct and repeatable conditions. The Photo Research device is calibrated regularly to National Bureau of Standards specifications.
It must have been broken that day :-D
Thanks for the information, You_Too! I wonder why it only works on that shot...
So the question remains: what the hell did they do?!
Why don't we try to look at this from a different perspective? What did LFL do to the colors? Are the colors the same on the '93 DC/'95 THX? If so, were the color changes carried out in the analog or digital realm? If analog, what hardware did they use? If digital, what software did they use?
Also, what source was used to create the D1 Master Tape?
Can someone with a bit more color experience than I try to take one of -1's images and GOUT-colorify it?
Wow, I don't know why the following AviSynth code works, but it seems to get rid of the over-represented reds in (at least, PAL) GOUT. I think it may have something to do with re-sampling. Can someone with more experience in color-correction chime in, please?
y = gout.MergeChroma(gout.BlankClip(), 1)
rgb = gout.MergeLuma(gout.BlankClip(), 1).ConvertToRGB24(matrix="Rec601").ConvertBackToYUY2(matrix="Rec601")
y.MergeChroma(rgb, 1)
It assumes that ``gout'' is in a YUV color-space; e.g., YUY2 or YV12.
(Note: I was trying to get Blackmagic Design's YRGB color-space in AviSynth when I discovered this.)
Edit: Of course, the colors will have to be Tweak()'d to get them right.
Edit 2: I may have jumped-the-gun here.
I don't know the proper terminology for this (black-level???), but PAL GOUT ANH seems to be more black...
Edit: Of course, both were cropped for this comparison. Furthermore, PAL was resized to NTSC (712x274), but I don't think that would drastically change how black the image is.
Musk.
frank678 said:
AntcuFaalb said:
Here's a GOUT-related question: Assuming that we don't make any color changes beforehand, how much do the reds need to be lowered to fit in with the rest of the colors? That is, what change would result in them no longer being over-represented? Is it just the reds?
This is pretty much THE question for me. I saw some posts where someone indicated they could repair the red channel out of the RGB channels. My understanding is that the red shift is an organic shift in the film - its not the red channel is damaged but the greens and blues have faded leaving the red over represented. Anyone know how correct that is? If so can the green and blue be repaired?
edit: had a go at a blind ajustment to blue and green curves to find a result = http://originaltrilogy.com/forum/topic.cfm/Calling-all-Color-Correctors-Can-this-source-yield-a-different-set-of-results-to-Gout/post/598240/#TopicPost598240
I asked You_Too to provide some examples of the over-represented reds (if possible) in my GOUT Bugs thread.
You_Too said:
Since I've been working a lot with the GOUT for mine and DJ's project, I can tell you at least some things for sure.
Well first, you should probably list ROTJ as PAL, because for some reason the NTSC versions of SW and ESB have the most resolution/detail intact while for ROTJ it's the PAL version which has the most, making it superior to the NTSC version.
Second, the reds are overrepresented in all three. That's why me and DJ could use the same color settings for all three and it worked. They all have the same overall color errors, except SW is more desaturated.
Third, the aliasing is there in all three. Just look closely on any half-diagonal line in scenes with a moving camera and you'll see it.
The chroma shift is unique to ESB though, and only happens at some parts of the second half.
There are other color errors here and there too though. Can't list them all since I haven't gone through all three movies completely yet, but some I know so far:
SW:
When Darth Vader appears for the first time, his belt glows red/pink instead of green.
In the scene in Luke's workshop where 3PO gets an oil bath, the lights on the wall behind him, a blinking light behind Luke while he's playing with his spaceship toy, and R2's blinking light all appears blue or purple, instead of green like they all are originally. The reflections on 3PO when he comes up from the oil bath are also purple instead of green.
When Luke uses his binoculars at night to look for R2, the stuff on the left side and the bottom appears orange instead of green.
When Luke gets hit by the practice orb in the Falcon, before Han laughs at him, that shot has a pink tone overall and everything that should be blue has been shifted to purple.
At one point in the death star where Vader appears in a corridor looking for Ben, his belt lights are red/pink again.
In ESB I know there are some desaturated or color shifted laser beams from star destroyers and ties, and some in ROTJ as well as the Falcon's laser being green instead of red at times.
This is excellent information. Thanks, You_Too!
Can you share some examples of the over-represented reds (alongside, possibly, examples of the reds reduced to levels that make more sense)?
CatBus said:
schorman13 said:
Yeah they are done, I forgot to mention that :)
Wait. My brain just woke up... do you mean the ducked commentaries are done and already in the updated existing archive, or done and waiting for the rest of the tracks to be done for a big update? If it's the former, I feel sooo sheepish.
Wait... should I re-download from MySpleen? Is it updated there?
Moth3r said:
What's the difference between a bug and a flaw?
Haha! Nothing. I didn't know about your thread, but to justify the existence of this one, let's say that a bug is something easily correctable by a non-professional (with a little help, in some circumstances, if needed).
Are there any specific issues with ROTJ?
I'm watching the PAL (Germany) disc now. The picture looks better to me, but I might just be crazy.
Is it possible that rather than upscaling from NTSC GOUT, they downscaled from D1 -> PAL GOUT?
why is using a new mix a bug?
I usually agree with you, but I'm going to have to go with CatBus here, especially now that we have a digital '77 mix available.
That's not going to fix the entire movie, though.
I agree, but I believe the infamous speeder shot is ruined enough to merit its own fix. Furthermore, it's relatively easy to drop in a replace that's similar in quality.
I don't notice things like the 4-eyed Stormtrooper, but I (and, I'm sure, others) do notice how destroyed those ~6 seconds are... even in motion.
none said:
When will Bugs be replaced with Deliberate Uncreative Decisions Supported by the Ease of Economic Expenditure? (DUDSbtEoEE)
Fixed!
Jay said:
We're experiencing the usual increase in site traffic that happens every September. That's probably part of it. We have more activity and content than ever before, which is weighing on the db and web server.
I also have a feeling that my hosting service views the hardware on which ot.com is hosted as legacy stuff that isn't getting the attention and resources it once did. It's probably outdated and has too many sites running on it.
We should see speed gains when the new software is running on the new hosting service (fingers crossed).
No, I don't have an estimated launch date :)
How many simultaneous connections can the DB handle?
CatBus said:
Actually I think it's just Star Wars that's wrong. I'm not sure it's easily provable if ESB and ROTJ are wrong, and I'd err on the side of caution for something trying to be a definitive list.
Agreed and fixed! My side-of-caution is the ?s :-)
CatBus said:
AntcuFaalb said:
CatBus said:
I thought h_h mentioned once that the GOUT version of Star Wars had the wrong soundtrack--i.e. something that was assembled to resemble the '77 mix in most respects, but wasn't the actual '77 mix. Anyone know the details, or am I maybe imagining things?
I thought they used the '93 mix?
Possibly--either way, the whole soundtrack can be called a bug.
Updated ;-p
Also, is the color-timing of ANH's crawl different from the rest of the video? That is, if you color-correct ANH perfectly (hypothetically speaking, of course) will you get the right or wrong colors for the crawl?
CatBus said:
I thought h_h mentioned once that the GOUT version of Star Wars had the wrong soundtrack--i.e. something that was assembled to resemble the '77 mix in most respects, but wasn't the actual '77 mix. Anyone know the details, or am I maybe imagining things?
I thought they used the '93 mix?
none said:
http://originaltrilogy.com/forum/topic.cfm/Whats-missing-from-GOUT/topic/6725/
Thanks! I'll have to look into this more closely to summarize.
First question: Is aliasing a problem for the entire GOUT or just ANH?