|
|
|
#527
03-07-2021, 02:43 AM
|
||||
|
||||
|
Quote:
The first image has nicely saturated red and green colors. https://visions4netjournal.com/wp-co...E546CABA0.jpeg The second image appears “warm” white balanced on my display. https://visions4netjournal.com/wp-co...8E38482719.png I took the liberty to lower the temperature only and nothing more in this third image as appears on my display. https://visions4netjournal.com/wp-co...4702B849D.tiff Now these images may look completely off to you, but this is what I see with P3 color gamut. Wayne warned me that my RAW Tiff files should be viewed on a WCG color managed monitor with matching software. I use P3 color gamut to match the color profile on my iPad Pro. Visions4 Magazine https://visions4netjournal.com/
__________________
|
|
#528
03-07-2021, 08:29 AM
|
||||
|
||||
|
Quote:
non-linear processing algorithm will "do the right thing". I post-blur in Photoshop so there will never be significant moire. Careful comparisons of my computer screen and the TV on the same image says that I'm not losing real detail even in the posted jpegs. This shows up clearly in swept frequency B&W calibration tests. I have of course tried in-focus shots and careful processing and that works, but you really have to be careful to avoid overload ... unprocessed pictures look very very dark.
|
|
#529
03-07-2021, 08:47 AM
|
||||
|
||||
|
Quote:
CT100 gray scale. Its quite warm looking. I tried P3 gamut and could not tell the difference.
|
|
#530
03-07-2021, 05:33 PM
|
||||
|
||||
|
You really have to consider each step of the process and what it does if technically correct, plus where it could go wrong.
The DVD or BluRay was supposedly transferred according to Rec.709/sRGB matrixing, and adjusted by a colorist looking at a calibrated Rec.709/sRGB monitor. This means mostly that the gain of R-Y is going to be increased non-linearly compared to NTSC to make the displayed colors correct. It also means that a display using the intended primaries will limit the colors to the Rec. 709 space. Now you play it back on an NTSC display that hopefully has been white balanced correctly and has the color difference gains set per NTSC specs. This means that when the signal says pure green or pure red or pure blue, you get pure NTSC green or pure NTSC red or pure NTSC blue, each of the same intensity needed to make white, and zero of the other primaries. BUT - when the signal on the disk is for pure green, it is expecting a display that produces pure Rec. 709 green, not NTSC green. Colors that are not fully saturated are affected too. Since your NTSC CRT has a wider gamut, it expands the saturation of the colors compared to what the colorist saw or what you should see on a modern HDTV. So - calibrating the color bars on the NTSC tube will NOT produce correct saturation on picture material unless that material was mastered in NTSC color space. Hence the need to turn down the saturation on the pictures to do an approximate conversion back to the expected Rec. 709 colors. Unanswered questions: Is Dorothy's dress supposed to be white? Was it actually white on the set? Was it gray tinted, or even slighly blue to prevent an overexposed fluorescent look due to the high contrast of the Technicolor process? Was it actually white on the original prints? I don't think we know. If you can capture a video frame directly from the BluRay and view it in Photoshop, you could measure the color represented on the disk so you know what you are aiming for. An acceptable substitute is to use your fixed manual camera white balance, turn the color all the way down on the color bars, and make sure the gray scale is neutral in your photos - tweaking the white balance manuallly if needed to make it so. But, even when the white balance is perfect, that is only the first requirement - it still remains that the disk is not mastered in the correct color space for the NTSC display.
|
| Audiokarma |
|
#531
03-07-2021, 08:34 PM
|
||||
|
||||
|
Now I know why you told me to “give up” which is not in my DNA. We quite possibly will never see NTSC color because it was never adopted (?), unless under carefully controlled conditions. There is no impetus for creating NTSC color because BT.2020 already exists and exceeds NTSC color gamut. P3 is about 85% of NTSC from what I’ve read.
You raise an interesting question about the dress. We recently purchased the new 4K version of the Wizard. This BluRay is more than just 4K. It is certified as 4K Ultra HD Premium. To achieve this certification it needs to have 4K resolution, minimum of 10 bit color depth, minimum of 90% P3 color space and it must use BT 2020 color representation. UHD Premium certification also requires the display to display high dynamic range (HDR) with a minimum range of 0.05 to 1000 nits and finally immersive audio object based sound. To me HDR is is more important than 4K resolution, and visually more dramatic when viewing. The new UHD Premium BluRay of Oz goes beyond and supports Dolby Vision with 12 bit color depth, up to 10,000 nits and most importantly its dynamic, meaning it adjust each frames color, contrast and brightness on a continuous basis. My LG E8 OLED is UHD Premium certified and supports Dolby Vision and Dolby Audio. I have photos of her dress from this disc prior to using Tiff. I would say the dress material is white linen and slightly transparent. I can see the texture of the material and the transparencies. The skin color behind the dress may give a look of gray on monitors/televisions. There is one scene in particular where the shot is wide and I can “see “ ambiance of the over head set lighting I will post a few photos using Tiff on another thread.
__________________
Last edited by etype2; 03-07-2021 at 08:45 PM.
|
|
#532
03-07-2021, 11:29 PM
|
||||
|
||||
|
The TK-40 and TK-41 cameras were designed to produce NTSC correct color and the TK-41 was in use well into the mid-60s (and probably beyond in some stations). Supposedly some stations started fudging color adjustments (I'd imagine many didn't especially stations that went color before 1957 since they already owned NTSC phosphor studio monitors and probably didn't have a newer monitor to compare against) to a compromise between NTSC and what later rare earth phosphor needed. Some color tapes from the mid-60's and older may contain correct NTSC signals... Finding them and finding copies that haven't had the color messed with could be tricky.
One likely example I can think of is the Eisenhower tape where he became the first president to be shown on color TV. Any modern source is not going to be set up to produce correct 1954 NTSC colors... unless some special signal source with the right software to convert color spaces is built specifically for the purpose. Such a source would probably have to use a digital input to tell it which color space it's getting for input so it could process it properly.
__________________
Tom C. Zenith: The quality stays in EVEN after the name falls off! What I want. --> http://www.videokarma.org/showpost.p...62&postcount=4 Last edited by Electronic M; 03-07-2021 at 11:36 PM.
|
|
#533
03-08-2021, 05:20 AM
|
||||
|
||||
|
I left out a key word in my above post, meant to say NTSC color gamut.
It is my understanding from internet articles and this forum that 1953 NTSC primary colors were too saturated and couldn’t be made bright enough for use in the consumer (CRT) TVs of that era with the exception of the 15GP22 and the 21CT55, so the NTSC Color Gamut was never used for volume commercial production of color TVs. As a result, the NTSC Gamut was never really an actual Standard Color Gamut, and there is essentially no consumer content based on the true NTSC Color Gamut? So instead of the official NTSC gamut colors, the phosphor colors that were actually used in early color TVs were developed by the Conrac Corporation, which eventually became the SMPTE-C Color Gamut Standard. TV production studios used Conrac color monitors to produce their broadcast TV content, so it was the Conrac Color Gamut rather than the NTSC Gamut that was the real early color television Standard Gamut. Correct me if in error. EDIT: I’ve seen photos of the CT-100 and 21CT55 used as monitors in 1955 at NBC studios Los Angles and New York (perhaps monitoring 1953 NTSC color gamut from TK-40 cameras?) If so, what programs and only kinescopes?
__________________
Last edited by etype2; 03-08-2021 at 05:40 AM.
|
|
#534
03-08-2021, 09:54 AM
|
||||
|
||||
|
In CRTs, the 15GP22 and 21AXP22 (early ones, at least) had NTSC phosphors. There is some question as to when the blue phosphor was changed in 21 inch tubes. The early 21 inch sets with sulfide (non-NTSC) green had NTSC decoding, resulting in some color errors. If you study RCA's schematics year by year, you can find when the demodulator components were changed to improve reproduction with the non-NTSC phosphors.
In cameras, the TK-40 and TK-41's (image orthicons) were strictly NTSC. This was determined by the optical filtering in the three color channels. There was no matrixing of signals between the red, green, and blue before gamma correction. When Plumbicon cameras came out, they were much less noisy, and had all solid state circuits, which meant that stable linear matrixing before gamma correction was possible. At this point, the dichroic prisms could be designed for optimum light efficiency and minimum signal to noise, with a reasonable approximation to proper response that could be matrixed before gamma correction to best match either NTSC phosphors or PAL (modern) phosphors. With the optics not tuned to a particular phosphor set, and the capability to adjust the matrix freely, a camera maker might start with a theoretically correct NTSC matrix and tweak it for appearance on the monitors of the day. When we made the HDTV format test tapes, we went through a process of measuring the cameras and adjusting the matrices to be theoretically correct for HDTV phosphors. We found that the BTS cameras were already correct, but the Sony had a matrix that had been tweaked for a certain "Sony look." Material that started with TK-41s has NTSC colorimetry on the original tapes, but there is still a question of how much the color was tweaked when the material was restored and transferred to modern media. The Eisenhower tape as well as the last-day RCA tape from the New York World's Fair on YouTube look surprisingly saturated to me. Maybe the color is as original, and maybe not. One confusing factor is that the TK-41's gamma correction was less than 2.2, resulting in an increase in overall system gamma (contrast) and an increase in saturation, a similar trick to that used by Technicolor and color transparency film. If you want to see how this works, take an image into Adobe Lightroom or Camera Raw and play with the Contrast slider. This apparently works on the contrast of the individual R, G, and B, so when you increase contrast you will also see the saturation increase. (The Highlight and Shadow sliders apparently work in Lab space, and do not affect saturation.) Last edited by old_tv_nut; 03-08-2021 at 09:58 AM.
|
|
#535
03-09-2021, 04:21 PM
|
||||
|
||||
|
To expand on the question of Dorothy’s dress in The Oz movie, the first link is a low illuminated scene with her face lit by the fire pyros. You can clearly see her skin tone behind the blouse. This is rendered as a darker white or grey on roundies with lower resolution and luminance.
https://visions4netjournal.com/wp-co...A4EE25B10.tiff This link shows a brightly lit scene to simulate being outdoors and her blouse looks “whiter”, but the transparency still shows. https://visions4netjournal.com/wp-co...53E696D01.tiff So color tones change with illumination and the colors of the environment. Both files are unedited Tiff, 16 Bit, P3 color space. Visions4 Magazine https://visions4netjournal.com/
__________________
|
| Audiokarma |
|
#536
03-09-2021, 05:30 PM
|
||||
|
||||
|
One other thing to note: if its from a very recent BluRay one cannot assume
that the colors are directly from the original three negatives. They, and everything else, could be reconstructed by what amounts to deep fakery by a computer.
|
|
#537
03-09-2021, 06:33 PM
|
||||
|
||||
|
“Deep fakery”
No such thing happened. During the two year restoration process among many other procedures, each of the original three-strip Technicolor nitrate film negatives (kept at a Warner Brother climate controlled 35F vault) were scanned at 8K 16-bit, composited them together and then applied a new color grain. The film was rescanned with the Lasergraphics Director 10K scanner. The restoration process peeled away decades of dust and dirt and used HDR to bring out colors and details that were in the film but not seen before. Examples, the rivet in Tin Man’s nose or freckles on Dorthy’s face. The images posted are from a STREAMED version of an UHD Premium BluRay disk as appeared on an UHD Premium certified OLED television with infinite black level, meaning the blacks are so low they can’t be measured. https://youtu.be/x-6MYP0RxTI https://postperspective.com/the-wiza...-hdr-remaster/
__________________
Last edited by etype2; 03-09-2021 at 06:42 PM. Reason: Typo
|
|
#538
03-09-2021, 07:29 PM
|
||||
|
||||
|
Quote:
The techniques are similar.
|
|
#539
03-09-2021, 08:42 PM
|
||||
|
||||
|
“deep fakery”
... and when I had my wife’s 49 year old engagement ring cleaned and polished, returned it to her finger and we saw renewed brilliance, clarity and color, ooohh, it was an just an elusion and “deep fakery”. ;-)
__________________
|
|
#540
03-09-2021, 11:03 PM
|
||||
|
||||
|
Quote:
By the way, I want to mention other flaws like film weave, dirt, negative shrinkage and resulting misregistration, etc. which are completely legit corrections, as Technicolor strove to minimize them in the first place. The correction of registration shows detail that was present in the negatives and was muddied by the printing process on film.
|
| Audiokarma |
 |
|
|
Linear Mode
