Kalleheikki Kannisto
1
COW Coin
Forum Replies Created
-
Seeing a snapshot of the emitter and other particular settings might clarify the situation. I would expect that to happen if the emitter had a z dimension other than 0, but other possible explanations exist.
Kalleheikki Kannisto
Senior Graphic Designer -
You’re already half-way there, copying the motion path to a shape layer. You can draw the other side of the path, turn the shape layer into a 3D layer and then align the path as closely as you can to the motion.
Alternatively — this should be easier and give you a more accurate result — you can first draw the shape layer and then copy one edge of it for the motion path and align it to the shape layer. In fact, if you draw the shape in 2D view without a camera and create the motion path and align the two before turning them into 3D layers, you should be able to keep them perfectly aligned by parenting one to the other before doing any 3D transformations.
Kalleheikki Kannisto
Senior Graphic Designer -
I’d go with a masked 3D shape layer with the full trail shape on it, revealing the shape as the box moves.
Kalleheikki Kannisto
Senior Graphic Designer -
I got curious as to whether that would actually work, and it indeed does. Linked here’s a rather basic application of that idea.
When a card is moved off the original position, the rest of the cards above it move down by one level (or index value) and thus drop down a position. There’s no tweening here, that’s another layer of complexity, but could certainly be added. It would also make more sense if the index was changed when the distance of the card from the original position exceeded the card dimensions.
While doing this I discovered that the expressions evaluate correctly only if the text layer is below the card layers. There’s likely an order-of-execution explanation to this, which I wasn’t aware of. I thought the property values are evaluated post-expression no matter the layers sequence but it seems this is not the case. At least not here. Maybe someone in the know can shed light on that.
Kalleheikki Kannisto
Senior Graphic Designer -
Noise reduction is a balancing act. A metallic or robotic ring is a side effect of too much noise reduction. The algorithms are not perfect, so trying to get rid of all the noise is not really the objective, the objective is to reduce noise without introducing perceivable artifacts. Try dialing down all the settings to half from the point where you start noticing the ring. That’s closer to where you want to be.
Kalleheikki Kannisto
Senior Graphic Designer -
A different approach:
Use a text layer to collect (in comma separated form, for instance) an array of only the cards that are visible in the stack and use the index number of the card for y position. That is, first generate the text layer content from some check as to whether the card is still in the stack and then for each card check if the card is in the text array and at which index position.
Kalleheikki Kannisto
Senior Graphic Designer -
Ok, I get the idea.
Perhaps the position expression should be such that each card layer (apart from the bottom one) looks at the one below itself and adds +1 to the y position of that previous layer for the current layer, except under some predetermined condition, where the layer (card) below it has been removed (x or y position goes beyond a given range?), in which case it would look for the next layer down for y position. Since each layer will look at the result of the other layer’s position this might ripple correctly through the stack. At least I think it would.
Kalleheikki Kannisto
Senior Graphic Designer -
Hi, fortunately I wrote it in the file description ☺
DeFlicker High Speed, Time Window 8, Coarse Color Match, Large Scale
Kalleheikki Kannisto
Senior Graphic Designer -
Oh, ok. Just add “.value” at the end of the expression, and it should read the text value rather than the text property.
Kalleheikki Kannisto
Senior Graphic Designer -
If you explain the purpose of what you are trying to achieve, maybe there’s a different approach to the issue.
I mean, I guess you could calculate a new “index” number for each layer based on their x,y position. But I’m unclear as to why you would want to do this, or if it is even a good idea, not knowing the final objective.
Kalleheikki Kannisto
Senior Graphic Designer