Michael Hancock
Forum Replies Created
-
That would make sense that it took a while if it was analyzing all the audio for syncing. but 4 hours is a long while. I would have given up after 15 minutes!
—————-
Michael Hancock
Editor -
How did you sync? Did you let it auto sync by audio? If so, I’d be curious if it would be faster to find a sync point and sync by markers, or if it still does some audio analysis for syncing in the background.
—————-
Michael Hancock
Editor -
Perhaps nobody knew the answer to your question, which is why nobody posted.
You spoke to Apple and got an answer – what did they tell you? How do you download an earlier version?
—————-
Michael Hancock
Editor -
[Tom Sefton] “6K HDR, 1000nits with colour accuracy for grading, and £5k – are you kidding?”
I would hold off on being too excited about using the monitor as a proper color grading monitor until it’s in the wild and has been tested for color accuracy. With no SDI in, there’s no real way to get an untouched video signal into it.
—————-
Michael Hancock
Editor -
Funny enough – I’ve used Resolve to trim FCPX projects before. Export FCPXML, import to Resolve, check that everything came in ok, then render a flat pass with handles. In the future, it might make more sense just to edit it in Resolve!
—————-
Michael Hancock
Editor -
Do you have more than one file selected in the browser window? If you do, you’ll only have Spatial Conform in the inspector.
—————-
Michael Hancock
Editor -
You can select your mask, then open your motion tracker and it will track the mask to your footage. No need to make nulls. Then invoke Content-Aware Fill. Crazy that it took overnight to render though. What are your machine specs? I’m on an 8-core MacPro and tested a 5 second 4K clip the other day and took less than 10 minutes.
—————-
Michael Hancock
Editor -
Select your video layer.
Put a mask around the area you want to Content Aware Fill and set the mask to Subtract.
Track the mask to the footage if you need to.
Select your video layer and open the Content Aware Fill panel.
Set your Fill method (I don’t know what the difference between them is, I haven’t played with it that much).
Click Generate Fill Layer.
Wait.You don’t have to make a reference frame in Photoshop unless you want to. If you don’t, I assume it uses AI to generate one.
—————-
Michael Hancock
Editor -
It looks like it creates a png sequence, and in the Content Aware Fill settings you can have it export a Photoshop sequence, so you can go in to each individual frame in Photoshop and tweak them if need be. Kind of neat how it does it.
In addition, it creates the png sequence and places it above the layer you’ve masked out, so you can always drop it below the layer and adjust the feathering of your original mask to blend it a little better. So there are a lot of options. That said, I’m an editor and not a compositor so it’s enough control for me, but for someone who actually knows what they’re doing they would probably want more!
—————-
Michael Hancock
Editor -
I tested it on a few clips this morning. Overall, it’s been really impressive.
I had an interview shot outside with shifting daylight and a textured shirt with a logo embroidered on it. I set it up to remove the logo, and it got about 90% of the way there. The issues was mostly subtle color shifts as the talent moved and as the sunlight changed.
Then I tested it on an ideal shot – talent shot on white cyc, wearing a dark solid color shirt with a name tag. I removed the name tag and it took about 10 minutes to set up, analyze, and remove, and it was about 99% of the way there.
First impressions – it will be super useful for quick, simple logo/blemish removal, and possibly for bigger stuff (or at least getting it most of the way there with the bigger stuff). Looking forward to using it in a real project.
—————-
Michael Hancock
Editor