Justin Mrkva
Forum Replies Created
-
It’s L_Iris_Horizon, I just made a tiny tweak to the hoop element to keep it moving when the flare goes offscreen (it looked funny when it stopped midway through the animation).
And yeah, mFlare is awesome. I used the After Effects lens flare since way back in CS2 or something, and when I got Motion I was a bit disappointed in the dismal “lens flare” element. But because of that I was forced to look for something else, and even AE’s flare is pitiful compared to mFlare. 🙂
-
A couple things about FCPX that I’m not sure if people totally understand yet here:
The library holds the original file, and will use that original file to render to ProRes422 for editing speed (in h.264, for example, the interpolation causes a major performance hit when going backwards or picking one frame; it has to read several frames before it to compute one desired frame). If you create Optimized or Proxy media, it encodes ProRes422 beforehand for use in the timeline. When you export, the best way is to go to ProRes422 and then encode that master file, being sure to check Original/Optimized media in Preferences. The loss from original to 422 is essentially zero, the loss from anything to delivery format (h.264, etc.) is what’s going to matter.
The only time you should be concerned is if you want to use the conformed clip for chroma keying or something similar, and you have 444 source footage. In that case, you always want to chroma key the original 444 footage because it has the full resolution color data.
Something that might be of interest if someone wants to test this: you can render to ProRes4444 in FCPX as well. Given 444(4) source footage, FCPX will still transcode to 422 in the timeline for editing. But will choosing a 4444 export force it to go through and re-render using the 444(4) source footage? Will that work even if Optimized 422 media is present? That could be an important thing to know, especially if there’s any chroma keying in the project. In that case, rendering to 4444 could subtly improve the quality of the key.
-
Also, remember to try the Match Color option in FCPX, correcting the ND footage to match a similar shot from the footage with better color. It’ll still almost certainly require tweaking but it can get you pretty close very quickly.
-
Actually, moving the text layer above it doesn’t change anything, and neither does the blend mode…
The zero opacity trick works perfectly, though. Now it makes me wonder… what if I had an add blend mode layer that I wanted to have cut off on the floor with a reflection, and then I had another layer that caused the reflection/cutoff to stop working? Hmm… something to investigate further.
Regardless, thanks for the tip. 🙂
-
That’s a straight preset from iMovie. 🙂 Also I believe Aperture or iPhoto can do that as well, don’t quote me on it though.
You could try recreating it, but given your time constraints, render out the globe graphic from iMovie, and presto. Then you can do whatever you’d like with it.
-
You mean the lighting? It’s a plugin called mFlare. Highly recommended, it’s plenty fast even on my 9400M GPU. It’s similar to something like Knoll Light Factory, and it works with FCPX and Motion 5, where KLF doesn’t as far as I know.
-
2D group doesn’t work. See my other post in this thread here for more information, I uploaded a file with an example as well.
-
Nope, it does it even with just a simple generator. I’ve uploaded a file to 5223_testreflectionstage.motn.zip that shows a minimal example. I want the checkerboard layer to cover the whole screen, not get cut off by the floor. Watch when the text starts to fade in, if it works like it does on mine, it’ll fill the screen at that moment. You’ll see what I mean.
I would wait to bring in the stage layer until the text comes in, but I have some other effects I want to have going during that time, and there may be times that isn’t possible so I’d like to figure it out without going that route.
-
It depends on the goal… if the motion graphics are intended to look realistic, then yes, you should try to stick with accurate physics as much as possible. In most cases, that’s probably the look you want to start with, especially if it’s actually the easiest option.
But you’re correct in that sometimes, either for a look or for complexity’s sake, you might want to take shortcuts. For the swinging sign, you could add an oscillate parameter to the damping value to more closely simulate variable damping with air resistance, which would be more accurate, but would also be an extra step that wouldn’t really make too much of a visual difference.
The correct answer? Do what works! 🙂
-
Actually, to be accurate, the oscillation speed of a swinging pendulum (your sign) doesn’t change over time. Physics nerds will cringe if you speed up the oscillation as the motion dampens. And it avoids the extra step of adding the ramp behavior to the speed.
If you want to confirm this, tie something to the end of a string and swing it, noting how long it takes to swing (heavier objects take longer and are thus easier to measure). You’ll notice that even when it’s nearly still and only swinging a few inches, it still just takes just as long between swings.
The same goes for a spring – the oscillation speed should not change as it dampens down, only the amplitude.