[Joseph Owens]

[Peter Robertson] “my timeline contains many video formats : HD Prores 422, 4k mp4 and 8k mp4. That all works on my new 27 inch iMac i7”

Not sure what “works on my… iMac” means, but

First thing that comes to mind is that possibly the Long-GOP mp4 high-resolution clips may be overtaxing your system’s cache/buffer/CPU/UI display buss when being decoded in Resolve. Its image engine is not the same as the playback-only function of most Quicktime-like applications.

jPo, CSI

\”I always pass on free advice — its never of any use to me\” Oscar Wilde.

[Joseph Owens]

[Robert Withers] “Seeking more info on “advisability” and “possibility” of having two versions of Resolve on this computer. I don’t know what the requirements for 15.2 are and if it will run on my MB Air.”

Two versions is a no-go.

BMD has a configuration guide on their Resolve Support partition that will give you some insight re your Air. You may be able to edit HD on it, who knows?

jPo, CSI

\”I always pass on free advice — its never of any use to me\” Oscar Wilde.

[Joseph Owens]

Probably a combination of low colour resolution (8-bit?) and Long-GOP compression macro-blocking.

There are some OFX plugins that try to “undo” mpeg artifacts- mostly they just blur the offending area.

jPo, CSI

\”I always pass on free advice — its never of any use to me\” Oscar Wilde.

[Joseph Owens]

You should investigate $pectraCal CalMan plus a C6 spectrophotometer.

They have signed deals with LG, Sony, and Samsung to access the internal color management within the 2018+ OLED monitors to do automatic alignment (DDC) Direct Digital Control. This is stated to be superior to LUT approaches as it offers native internal processing at a finer step resolution than a (for example)17-point cube.

jPo, CSI

\”I always pass on free advice — its never of any use to me\” Oscar Wilde.

[Joseph Owens]

[Richard Swearinger] “I was taught that changing only the luminance would help prevent color shifts”

Seems like an incomplete understanding of the principle of “constant luminance” which is a strength of the (small-d original) daVinci-originated algorithm. If you simply adjust RGB without regard to what that is doing to the YCbCr matrix, you will get a non-uniform locus owing to the power-function distribution within the colorspace. Think about how Y’ (gamma-weighted) is calculated as a single weighted-value based on the contributing tristimulus components and you will start down the right path toward grasping how you would really need to adjust the overall balance of all three in order to keep their relative vectors coherent in order to achieve the same hue rendition within a changed brightness range. Luma/gamma weighting also affects the perceived saturation of a color value, so it isn’t just a mathematical problem, either.

Color correction is one of the most non-linear pursuits that you can attempt. Most people think of the CIE colorspace as a horseshoe, but that is just in 2-D representation. Luminance-weighted, it looks like a lopsided ski-hill full of moguls.

jPo, CSI

\”I always pass on free advice — its never of any use to me\” Oscar Wilde.

[Joseph Owens]

I expect their word processing will improve, given that this is Subtitle0.1… As a free add-on, no… it certainly doesn’t do what MacCaption or CaptionMaker are capable of, Which is exactly what you are asking it to do, but if you have explored any dedicated caption/subtitle tool, you will find that they have hefty investment antes and continuing support fees.

The fact that the Resolve timeline subtitle cells are timeline based is a VERY significant step forward, which should really give the application developers at Telestream something to think about. There is absolutely nothing graphic or gestural about the tools that we have been using from them for over a decade. However, it works just like a world processor, fully searchable, replaceable… all those things that you take for granted in Text/Word… etc.

The other really interesting aspect is being able to tie a text cell to a clip… that’s something that makes it all editable, which is what just kills your work if things get re-versioned for various deliverables.

I did get a chance to mess with a feature-length project a few weeks ago that needed a lot of re-positioning, italicization and in/out trims (in Chinese, SimSun font) but it was way easier than doing the re-version in the original application. The workflow was bringing in an srt that I had built for a Blu-Ray release, which originally had three subtitle streams. This retains most of the markup instructions, and offered far better stylistic treatment in terms of letter attributes like HSL and drop-shadow. A step forward, and I expect ore to come.

jPo, CSI

\”I always pass on free advice — its never of any use to me\” Oscar Wilde.

[Joseph Owens]

Projection in a commercial-run theatre or…?

jPo, CSI

\”I always pass on free advice — its never of any use to me\” Oscar Wilde.

[Joseph Owens]

[Michael Gissing] “The new RTX cards also have a way to link two cards”

Hope it’s not just VLC… Resolve doesn’t see that threading.

Also, those who wait… (ah, that’s not really a strategy…) there will be some reason I won’t be able to use these cards and jump straight from my Titan Blacks over the 1080Ti’s…

jPo CSI

\”I always pass on free advice — its never of any use to me\” Oscar Wilde.

[Joseph Owens]

[Mike Krause] “Is there a way in Resolve to “reinterpret” those values?”

ACES.

jPo, CSI

\”I always pass on free advice — its never of any use to me\” Oscar Wilde.

[Joseph Owens]

[veronika koubova] “Gamma is Input: 2.2 and Output: 1.0.”

… might be one of the only useable pieces of information here… (and invalid for cinema production anyway, as the television industry has already moved on to 2.4.

It seems like you are getting a purely graphical output with no regard for display-referenced motion picture colorspace.
In a way, the same issue as a musician providing a sound track referenced to “full” digital gamut – the way music CD’s are often produced at full excursion “to provide maximum dynamic range” — when the industry standard for television delivery sets a reference at-20 dBf.
If there is a post-production supervisor on this project, they are going to have to negotiate a delivery that takes the *deliverable* into account, if you want to make your life any easier. If the graphics company is producing something that has been artistically deemed to be the desired result, then your only option is to “make it work.”

When you state that some values appear “oversaturated”, the question is, compared to what?
If the onus is on you to reconcile the values, you may have to use your Input settings like looking through the wrong end of a pair of binoculars to make them smaller in order to fit into your grade.

Apparently they are just fine to whoever is producing them, and how they are being evaluated. Probably no scopes or other measurement values involved.

Were this a problem on my system here, the first approach would be to make sure that the files are being reproduced correctly — full digital sometimes means over-riding the Resolve “Auto” setting and forcing the choice to the correct method. Then opening up the source media import colorspace — ACES is the largest universal gamut/workflow framework – and picking an Input Transform that will reconcile the apparent image with the Output that you are aiming to deliver in. You have, in a sense, the opposite issue to using a Transform to reconcile Log-sourced footage in a Linear workflow, where you have to expand the values so they reconcile with your display-reference. In fact, you appear to have to “compress” the picture values so that they are compatible with the colorspace of your working environment.

jPo, CSI

\”I always pass on free advice — its never of any use to me\” Oscar Wilde.