Paulo Jan
2
COW Coins
Forum Replies Created
-
Thanks. I had thought of this, but the way I see it, Xpresso is mostly useful in situations where there are many materials with different properties and one (or a few) common features; in that case, you can take those few common features and wire them up to a single control. (In your video, for example, there are 3 materials with different colors, but with Xpresso you can control the same noise map in all of them).
My case, OTOH, seems to be the opposite: all my materials are identical, with only one feature distinguishing them (the alpha channel), so if I had to control all of them with Xpresso, and since I don’t know beforehand what I’ll want to change, I’d have to wire up all their properties (diffuse color, specular, luminance…).
-
It depends on your workflow. If you are going to do no grading (or minimal grading) on your footage, then it’s not worth it. If you are going to do extensive color grading, though, it can be useful.
Cinestyle, in my experience, is most useful when it comes to preserving shadow and highlight detail. Try importing a neutral image and lifting the shadows, and see how much detail you get. Now try with a Cinestyle-shot image. You’ll be surprised. The same goes with noise. Try zooming into an area of your image that is dark but textured, and compare the amount of compression artifacts that you get with the “normal” profile and with Cinestyle.
The drawbacks? Giving more detail to shadows and highlights means less bits for the midtones. If you shoot stuff with big, flat areas with subtle color gradients (a wall, for example), you’re more likely to get banding with Cinestyle than with the regular styles. Also, you need to know how to grade and treat log footage. This post might be useful:
-
Read this. It explains it all in detail.
https://prolost.com/blog/2009/12/3/you-didnt-believe-me.html
-
Actually Cinestyle isn’t made by Canon, but by Technicolor, and its point is precisely to output an image as “flat” as possible, to make it easier to grade later. That’s the reason why I was using it:
https://www.technicolor.com/en/hi/theatrical/visual-post-production/digital-printer-lights/cinestyle
I haven’t had time to try Keith’s suggestions yet (other work came up), but I’ll see if I can do it this weekend.
-
The footage is shot using a Canon 550D DSLR with the Technicolor Cinestyle curve, hence the unsaturated look (and probably the chromatic aberration you’re seeing: 4:2:0 compression and all that). Despite that, it keys quite well: the screengrab I uploaded was done just with the eyedropper in Keylight, without touching any other control, and as you can see, the background and foreground are well separated (I’ve certainly seen much worse). The only issue is, as I mentioned, with the lower part of the arm.
I just tried doing as you said, adding the Cinestyle recommended S-Curve before the key (using LUT Buddy) to restore it to a more “normal” look, and the resulting key is practically the same, with the same problems.
-
Well, certainly, but I was just looking for a quick and dirty way to view the slow-motion files I’ve just recorded, without having to go through the hassle of transcoding, reconforming, etc. Transcoding can take a while, after all.
-
It just happened to me too. I had some clips shot at 50fps that I had conformed (via “Interpret footage”) to 25fps, in order to make them play at slow motion. I added those clips to the timeline with their sound (which presumably has also been slowed down)… and Premiere started crashing. I removed the sound clips (while keeping the slow motion images), and the problem went away.
(…which I’m writing down here in order to help those people from the future that might arrive here via search engines 🙂 ).
-
Thanks for your answer. But in that case, I take that there’s still some other gamma corrections going on in the system? The books I’ve consulted all explain it in a relatively simplistic way: “oh, the monitor has a certain gamma value, so the software adds a gamma correction to compensate for that”, which makes it sound as if there was only one gamma correction working in the entire OS/software stack. But judging from AE’s behaviour, it’s clear that there must be something else going on. After all, even if you click “Linearize working space”, the monitor will still be darkening the images, so AE will still have to compensate for that somehow…
-
Well, sure, but what if you need to make changes to the edit after the grading? (“Locking an edit”, as we all know, isn’t always possible with clients; there often are last-minute changes, some of them even justified). If you send the sequence to AE using Dynamic Link, it gets turned into one big clip in Premiere, making it impossible to trim, replace clips, etc.
-
Maybe. The default setting for the data rate is “automatic”, and I didn’t touch it. However, my understanding was that the deinterlacing in Compressor didn’t simply discard one field, as most crappy deinterlacers do, but did blend both fields to create the new image.
Paulo.