Fusion or Compound Clip to Reveal Offscreen Elements Upon Transform
I'm trying to compose several video clips with transparent/black backgrounds against the default black background in Resolve. Then, when they are positioned correctly with respect to each other, bake them together into a single Compound Clip or Fusion Clip, go back to the Edit screen, then apply Dynamic Zoom and some Pitch and Yaw transformations to that entire thing so it moves and transforms together as one object.
The problem is, some of the clips are positioned off of the visible canvas at first, and will only be revealed during the dynamic zoom and other transformations.
To help you visualize: let's just say I've got a totally black background and a 5 identical video clips of water (as if from a fire hose) shooting into the middle of the screen, so 5 streams of water. Then I get a still image of an actual fire hose, copy it 5 times, and position those hoses so their edges are just offscreen, where the water streams are starting from. I combine these into a Compound Clip and go back to Edit screen.
I apply a Pitch of .5 so it "pushes" the top half of the entire clip backwards into space, thereby revealing the hoses which were previously offscreen, and of course the water looks like it's coming out of the hoses.
I tried both ways, using Compound Clip and Fusion Clip. This problem occurs in both: that it's not actually displaying any of the offscreen elements when they should be revealed because apparently the "infinite canvas" you are working with in Fusion or in the Edit Screen becomes a cropped edge once you go back out of Fusion or create the Compound Clip. I can bring a little bit of the hose into the visual area just to test it, and when I apply the Pitch so it should reveal it, I can see just the tip of the hose. The rest of the image is cut off.
I'm sure there is some way to do it, but the main thing is I need to go back into the Edit screen to apply transformations there, rather than applying transformations within Fusion while I still have the infinite canvas. I'm just trying to use Fusion Clip or Compound Clip to bake in all the elements so they are treated as one object in the Edit screen.
I have too many clips to go and apply math to each individual one within the edit screen, so I definitely need a way to position a bunch of things together, bake them in, and transform that huge object as one giant plane, not 5 or 10 different tiny planes independently.
So how can I compose lots of elements, including some off screen, into one object to be transformed, but still include those elements off screen so that when the giant object is transformed, those parts hanging out in the wings will be revealed?
I suppose a potential workaround would be to compose it all within the visual area, and then start off my Edit screen transformations zoomed in so they are hidden, but this would result in effectively lowering the source resolution of my clips during the "baking in" process.
Part of your problem is the fusion clip from Resolve, I believe, is stuck at the resolve frame size, so that is where your cropping is probably coming in.
If you launch stand alone Fusion, and start there, you will not have the same issue.
Thank you for your reply. You are correct that both Fusion and the Compound Clip are sort of performing a "mini export" at the timeline resolution, but believe it or not my timeline resolution is 16K.
So even though I'm working with a width of 15,360 pixels, I still have elements outside the perimeter. Yes I could shrink everything to fit within the frame, but my main focal point in the center of the frame is in a 16k quality - it stretches all the way from top to bottom, but it is 13,000 pixels wide, so that leaves black bars on the side
It's these black areas that I'm adorning with some 1080p video overlays. It's just that I want the overlays on the perimeter to be outside the visual scene at first so they can be revealed upon transformation.
The purpose of maintaining full source resolution is to allow me to zoom in by several magnitudes without any loss of sharpness. So I figured, maybe some expert out there knows of a way to do it, or if there is some setting I have missed, since I am admittedly a complete beginner to this software.
If not, my workaround of compositing everything within the frame, and then doing all my edits from an initially zoomed in place, would work, but it would partially defeat the purpose of keeping everything 100% source resolution. In other words, I would be scaling down all the clips to fit them in, and, as you mentioned, upon returning to the Edit screen, Fusion or Compound Clip would export it in the Timeline Resolution.
I realize it sounds ridiculous to suggest that 16k is NOT quite enough space to work within, but when my central display feature is in fact 16k pixels, it's worth inquiring.
It would also save me alot of work, because doing it the workaround way would require programming a zoom over and over and over just to get to my "default" view.
Also, I don't have the standalone version of Fusion. I'm working in Davinci Studio 16.2 (the paid version). So far, I haven't even had to use Fusion at all, since the Compound Clip does work very nicely for what I'm trying to do, other than this little detail of the elements which are totally or partially offscreen.
Working with it a bit more, I think what I can do, for anyone else interested, is instead of trying to do it all in 1 Compound Clip, I'll make 3. 1 for the left perimeter, 1 for the right, and then the middle content.
Then I can apply the transformations to just 3 giant clips instead of 10 or 15, and that's manageable. And since I'm just compositing those in the Edit screen, and NOT combining all 3 into a Compound Clip (which would cut off the parts outside the viewing area), I'm still working with an infinite canvas, so as I zoom in and pitch and yaw and rotate left and right, I can see the other elements in my virtual space I've created.
The fact that you are working at 16k is irrelevant. You could be working at 5000k, but as soon as you send to fusion from within Resolve, it will crop.
If you are trying to punch in on 16k footage without losing resolution, you should actually be working at a smaller resolution, but telling Resolve to keep original size on the source clips. That way you are looking at a 16 image in (for example) HD. This allows you to zoom out on the16k clip to frame it to fill the screen with as much of the picture as the aspect ratio will allow, as well as zooming into the clip (all the way to 100%) without losing resolution.
But your initial problem is still that you are sending clips to fusion that are clipped at the edges. You might be able to center them all in the frame, and then send to fusion...but again, working at 16k (If i'm understanding what you are trying to do) is not going to help. In the end, you are in 16k, zooming in on 16k images....not too much different than working at HD, zooming in on HD images...now working at HD, zooming in on 16k images...that is what I think you want.
Or as I said originally, launch Fusion stand alone, and load your 16k images into an HD frame, and move around and be amazed at how much you can zoom into an image without losing resolution.
I'll try it thanks.
But in my non-fusion workflow, I'm working at 1080p and switching to 16k before render in order to avoid that pixelization, which definitely does happen if I don't do that. For example, take a 16K still image, zoom in 20x, you can see the difference.
But I'll download the standalone Fusion as soon as I get off this metered internet connection. So I truly haven't been able to take your advice yet.
However, after working with it some more last night, I overcame the problem in the way I said, where I just made 3 different Compound Clips, and it ALMOST worked. The only problem was, the two clips on the side, since they were slightly offset to push their elements offscreen by 180 pixels, it did cause a slight incongruity in the Pitch transform such that the lines of perspective between the row of elements on the perimeter did not match up to the lines of the center element.
So I'm just going to go with the path of least resistance for now and put all 3 compound clips into 1 compound clip so it all moves perfectly, and just proceed ahead with the fun part of programming my transformations.
Once I'm able to get Fusion, I'll definitely try it again as you said, and let you know.
Thanks again for taking the time to reply.