Joseph Owens
Forum Replies Created
-
[Shane Ross] “Apple no longer believes that DVD delivery is needed…”
Really? Not even in a “Share” dialogue? I thought there was provision at least for a BluRay output????
Why not upload the video to FaceBook and then download it to everywhere… isn’t that the way Apple wants it to play?
Oh, well.
jPo
“I always pass on free advice — its never of any use to me” Oscar Wilde.
-
[Joel Arvidsson] “I know I only have one graphic card in it right now”
That you’re getting 10 fps is still doing surprisingly well. No red rocket, and no GPU other than the one that is also supporting the GUI?
You should see some gains with multiplying the CUDA-enabled cores with more cards.
I can get 24 fps with 4K RedOne RDC playback (3 GPUs plus the Q4000 GUI card) at quarter Res with no red rocket.jPo
“I always pass on free advice — its never of any use to me” Oscar Wilde.
-
[Sascha Haber] “And the trouble starts when they like your UI better.”
(this should be read as a LouisCK-ish rant, so proceed with caution)
Hmmm… sounds familiar… where have I heard this before…. oh , yeah, nearly every day for almost 20 years. From before we had dependable monitor-matching… to now when there are almost no standards in place that anyone pays any attention to, or are possibly not aware of, plus we are no longer grading for a well-defined target, which used to be strictly broadcast. And everything, at least here, was 525i. Plus ça change…
Sounds like the UI display device should be… maybe at least a calibrated, characterized Dreamcolor, rather than any random $75 Insignia sRGB graphics display from Staples or Future Shop. Okay, the colorspace may be similar to, but it isn’t 709.
And I somewhat agree that the UI should probably be monochrome. I use it for conform, tracking and setting up power windows. That is all. Grading with your nose up against the glass is no way to get a sense of the picture values. Close enough, all you see is pure RGB dots, let alone a nuanced, contrast-balanced rendition. It occurs to me that someone who is re-painting every single element in a scene is a graphic artist, not a colorist. But this may be an out-moded world view, which is related to an also possibly dated interpretation of what DP/Cinematographers do.
I mean, how on earth did we possibly function with the text-only displays we had with daVinci Renaissance? event-beep. {static][-1][preview][dissolve] = auto dynamic scene ripple…. woooooo….(the little three-eyed guys from Toy Story, who obviously see in Technicolor).
“I can’t work this way!”, “no brown M&M’s…”

jPo
“I always pass on free advice — its never of any use to me” Oscar Wilde.
-
[John Burkhart] “any suggestions”
Maybe one way to work around the additive R+G+B overexposure, and perhaps start coming back to the original process would be to think about it as a negative effect which is sort of where you’re going with the CYMg filters– flip the image black-for-white and primary hues for complementary secondaries, which really is what is going on at the three-strip film origination.
That way… whites are at zero and… zero+zero+zero is still zero. Blacks would be really solid at the other end… and that might be a corollary to the subtractive process.
jPo
“I always pass on free advice — its never of any use to me” Oscar Wilde.
-
[Lynette Gilbert] “I convert to ProRes 422 all the time – what have I done wrong?”
Are there any differences between the basic attributes of the HDV and ProRes files?
There should be a few key differences, though: FCP prefers that ProRes files conform to its orthodox attributes — square pixels, 1920×1080, 23.98 or 29.97 fps, and so on.
ProRes also requires about 10x the basic bit transfer rate as it is a much denser codec — although processing is lighter than the CPU-intensive Long-GOP of HDV.
Seems like too basic a question, but — does your sequence match the codec?
jPo
“I always pass on free advice — its never of any use to me” Oscar Wilde.
-
[Sascha Haber] “I have the feeling there is more to it than just the pure math.
Because it doesn’t look much like it :)”Same gut response here… and/but also completely fascinating.
I can’t convince myself that there is a linear correspondence between the decoded RGB values of a digitally-gathered file, possibly (or probably) matrixed as it is for 709 space, and mechanically filtered light recorded on black&white stock, which will have its own spectral response. What are the center-wavelengths of the printed gelatin (complementary) three-strips and printing densities?
Too bad we don’t really have the same Y’CbCr extraction and conversion tools as SHAKE, where the split and recombine nodes are a click away. and….
Its one of the few things I miss about AppleCOLOR/Final Touch.
jPo
“I always pass on free advice — its never of any use to me” Oscar Wilde.
-
This might change shortly, as the 9.1 version of Resolve includes support for the Decklink 4K.
jPo
“I always pass on free advice — its never of any use to me” Oscar Wilde.
-
[andrew smith] ” at the end of the footage the DP was kind enough to include a few seconds of the color chart! “
Shot under what conditions, with which lens, exposure, … etc.?
Its been 6 or 7 years since I last saw a chart… and as it turned out, wasn’t exactly scientific, and really turned out to be just a sort of suggestion that the other 75 or 80 LUTs that I developed for that particular project vaguely resembled. Really vaguely. And that’s after 12 years of film dailies before that, some with a chart per setup, some with no charts whatsoever. Some were shot very rigorously, in-scene, with the lens that was in use for the setup, with the lighting in place… those are very useful for setting up white point and grey scale, gamma, anti-log curves, you name it. In a practical sense, though, that is rare mostly because there simply is not enough time to do it between shots. (Notwithstanding in the olden days that stock was expensive.)
A chart will tell you about white-point (color temperature) and gamma, if you pay attention to the grey scale — if it has a reflectance-step, like a Macbeth-Greitag or a Kodak Q-60. But it is only valid for the conditions it is shot under. If somebody shot it as an afterthought in the truck, and maybe even changed lenses (because of some reason that seemed like a good idea at the time), it is totally useless, actually misleading.
jPo
“I always pass on free advice — its never of any use to me” Oscar Wilde.
-
[Eric Sternberger] “would it work with two iMacs connected via thunderbolt??”
No… see above…
jPo
“I always pass on free advice — its never of any use to me” Oscar Wilde.
-
[Matt Perrin] ” I wanted to see if anyone has some tips on the best way to set up an FCP project so it is easier to work with in Resolve.”
Two months ago:
https://forums.creativecow.net/thread/277/21965A little primer and fill-in-the-blanks form. These tip sheets are always a work in progress since the base application is constantly evolving, and approaches to things like ie. mixed framerates, change.
A couple of quick, but by no means exhaustive, notes that aren’t mentioned in the document….
Media Manager creates files name-v, name-2-v and so on without a .mov suffix. FCP will recognize these, but Resolve will not. You have to add the suffix and re-link… and then FCP will lose them.Also any file names that have illegal characters, HTML or path-name characters will confuse the XML and directory pointers and will result in crashes, hung renders and incomprehensible, if not misleading, error messages. A “/”, for example, in a file name will cause a render queue to halt and freeze the application if it is imbedded in a job with other clips; rendered singly, it will cause the render to halt and display the error “codec not opened yet”.
There are others……………………..
jPo
“I always pass on free advice — its never of any use to me” Oscar Wilde.