Kalleheikki Kannisto
1
COW Coin
Forum Replies Created
-
Would more than likely be way cheaper (your time is money), easier and faster to rent a place with a good piano and shoot it with a couple of cameras. You’d be done in a day with the footage itself. As opposed to several weeks or even months with AE… to possibly end up looking like you faked it in AE. In terms of time, it is about 1:100 ratio between shooting something or doing something with VFX. So unless something is (next to) impossible to shoot, just shoot it.
-
My theory follows.
I think you’d need a whole bunch of Animator effects applied to the text, as many as the maximum number of letters you will have in a line of text. Then you can set up (sliders and) expressions for animation.
The first expression you need is one that will automatically apply the range selector based on how many characters you have in the text. You can count that or use a slider to control it, the first option being the more automatic one.
Then you need a slider to set the length of the “tween” animations (how long do the horizontal or vertical animations take), possibly another two for maximuma offset horizontally and vertically. And perhaps one more for randomseed, in case you need to change it for a layer.
Then you need that one expression that ties this all together for one letter; range selection, in-between X and y positions, starting x and y positions, the animation first along one axis, then another and repeat for the second set.
I’d imagine it is possible to base the range selector on the animator name (so that Animator 1 and Animator 2 select a different range based on “1” and “2”, etc.), but if that’s too tricky, you can do that once manually for every Animator for the longest line of text. After that you can copy the layer and change the text.
That’s how I’d appraoch it.
-
If you have several separate 3D EFFECT layers, that are actually 2D layers with an effect, the will not interact in 3D, but follow the layer order.
In Element 3D there is a Unified Render option to stack several E3D effects on one layer with correct z-depth.
With other 3D Effects, you may need to use track mattes created from z-depth or world position pass to make front layers disappear “behind” other layers.
-
The simple approach would be to construct the room from matrix footage layers turned 3D.
For the people it would be simplest to key them out (shoot them on green screen) and use the key as track matte. That’s not going to be very 3D, but you could use a blurred copy of the key as a displacement map to give it some apparency of 3D. Or two displaced layers for front and back sides if you want the people “see-through”.
Other details like those hanging wires in your sample image add to the 3D effect.
-
Particular already follows the comp camera as a 2D layer with the effect applied. Turning the layer into 3D results in “double” 3D transformation, your Particular effect is already 3D as the Particular camera matches the comp camera, and then you add another 3D transformation with the comp camera for the layer if you turn it to 3D. That second 3D transformation is extra, unnecessary and will actually result in the effect NOT following the camera. You end up moving the Particular effect twice as much as the camera while doing a 3D to 2D to 3D conversion in between.
-
I would do such a thing with Filter Forge, wherein you can take a bunch of leaf images and scatter them in a given area.
Genetica is another good texture generation plug-in.
Both have 30-day trials. There is a learning curve involved with both solutions.
-
Me neither, but when I have to look at something five times very carefully to still be unsure if it is 3D or 2D, it is probably not critical in achieving the look.
-
Why are you turning it into a 3D layer?
-
Looks like shape layers rotating in 3D space with high depth of field setting.
I don’t see anything organic nor anything that needs Particular there.
-
it is a color correction question and the real question is whether it is possible at all.
It is probably possible to make BMCC footage match Canon footage or try and split the difference, but this way round, doubtful.
It is possible to make Canon footage look somewhat like BMCC, but just how close you can get depends on the quality and format of the original footage.
Which Canon model are we talking about? Do you have some footage you can upload? Meaning the final graded BMCC footage that you want to match AND the raw footage from the Canon.
What color correction software do you use?