Expressions: Working Smarter in Adobe After Effects

1. Expressions - Trailer: Hey, I'm Jake Bartlett and this is expressions working smarter and Adobe After Effects. I've been working with After Effects professionally as a motion designer since 2010 and playing around with the program for years before that. If there's one thing that I've learned in all of that time to work more efficiently as a motion designer, It's how to use the expressions, expressions of the coding language that After Effects uses you can apply them to any keyframe of all property and allows you to control and manipulate those properties in many cases, making complex or tedious animations, very simple and easy to manage. And if you don't believe me, take a look at my class project, this entire animation was driven by four keyframes on a single property. Seriously, and I'm going to show you how I made every step of this project from scratch. We'll start by learning all the basics of the expressions that we're going to use to recreate the animation, how they work, and how to control them. Then I'll show you how I created my project from start to finish, teaching you how to take advantage of all the expressions that we cover, building custom controls for making animation easier and ending up with the final project. You can create your own artwork for a unique class project, or you can use the provided after-effects project with my artwork all ready to go and just follow along with me. This class is for anyone with a basic understanding of After Effects. If you've never used the program before, then I would highly encourage you to take one of my beginner's classes first, like the beginner's guide to After Effects or the beginner's guide to Animating Custom gifts. I know the thought of expressions can seem intimidating, especially if you've never used code before. But once you get a basic understanding of how to use them and how to take advantage of them in your own projects. It'll unlock a world of possibilities that you never knew existed. You won't regret it. I'll see you in class. 2. Class Project: For the class project, you'll want to make something that's driven mostly by expressions. My project is driven entirely by expressions all tied to one property with four keyframes. If you want to just learn about expressions and get right into it than just download my project file. Go to the Project and Resources tab, find that After Effects project file with the artwork already created a shape layers and you can follow along with me as you watch the lessons. Otherwise, you can create your own artwork and make a unique project applying the expressions that you learn in the next few lessons. And that's exactly what we're going to do next. We're going to learn about the expressions that I use most in my own workflow, how they're written, how they function, and how you can control them so that you can start working a little bit more intelligently inside of After Effects. If you have questions at any point throughout this class or you're running into errors, feel free to leave a discussion. I will get to that as soon as possible. I check those almost on a daily basis and you can be sure that I will answer your question as soon as I possibly can. 3. linear();: The first thing we need to talk about is expressions because that's gonna be the driving factor behind the entire animation of my class project. There are so many different expressions and they can be used to do so many different things that it's kind of intimidating to know where to start with them. And expressions are actually just JavaScript, which is a coding language that's been around for awhile and it's used extensively. And that means that there's actually a ton of resources online for learning any expression, because it's exactly the same as JavaScript. So if you're ever curious how you could do something in After Effects using expressions, It's very likely that somebody has already asked the question and received an answer online. It just may come through a JavaScript forum instead of an aftereffects specific expressions for them. But that doesn't really matter because everything that you learn about JavaScript can be applied to expressions in After Effects. The first expression I want to talk about is easily my most used and probably favorite expression just because of how much you can actually do with it. And it's the linear expression. What the linear expression allows you to do is map any range of values to another range of values. And I'll show you exactly why that's so powerful. I have a square here and let's say I want to add an expression to the rotation to automatically rotate as I move the position on the x-axis. Well, I'm going to bring up the position and rotation by pressing P and then holding Shift and pressing R to also bring up the rotation. The rotation property is what I want to be driven by the x position. To make this easier, I'm going to right-click on the position and go to separate dimensions. That way the exposition has its own property. We could do this with both the x and y position in a single property. But that's something we're gonna get to a little bit later. For simplicity, let's just focus on the rotation and the exposition. So with both of those selected, I just Control or Command clicked on both of those. I'm going to double-tap the S key to solo those properties that we don't have to think about the Y position. To add an expression to any property you need to Alt or Option, click on the stopwatch. I'm going to click on the stopwatch while holding Alt for that rotation property. And After Effects is going to automatically fill in the expression it needs to reference that property. So if I were to apply this expression by clicking off, it's gonna be exactly the same as before, except my numbers are now read. This means that it is being driven by an expression. But because the expression code that was written in here is just referencing itself, I can still change the rotation to whatever I want and the expression isn't really doing anything for me. Alright, let me go back to zero and click on that expression to start writing my own. Now with expressions, you can type everything out by hand, but there's this pick whip right here. That's specifically for expressions that I'm gonna be using a lot and that you should get comfortable with. It's a lot like the parenting pick whip or the property pick whip. But it works while editing an expression just like the parent pick whip, compare it one layer to another and the property pick whip, compare it one property to another. The expression pick whip can grab any property that you select and add it into your expression. So with everything selected in here, I'm going to use the expression pick whip, and I'm just going to drag it up to the exposition and let go. Now this expression is saying to look at the transform controls of the layer and then the exposition property value. And that's what it's going to plug into the rotation. So if I click off, we now have a different value than zero on the rotation. And the reason it's kind of a weird number is because once this gets up to 360 degrees, cycles back down to zero and adds one revolution. And since the x position is measured in pixels and the rotation is measured in degrees, That's why the numbers don't line up specifically. But if I zoom out a little bit and crank the exposition down as I do that, check out how the square is also rotating. This is a direct link between the values of degrees and pixels. And that's pretty cool except it's rotating a lot more than I want it to. If I wanted it to look like it's actually kinda rolling than it needs to rotate a lot less frequently as this exposition is changing. And we could do some simple math in this expression. If I just click into edit it and go to the end, I could say maybe divided by ten. That's going to take the exposition whatever it is and divide it by ten and then plug it into the rotation. So let me click off and see what that does. Now it's rotating a whole lot less, in fact, not enough at this point. I could go back and forth with this and modify my calculations until I get something that's about right, but that's not actually what I wanna do. Instead, I want to show you how to use the linear expression to isolate a specific range of values of the x position and remap them to another specific range of values on the rotation. So to write the linear expression, Let's go into that expression and just clear it out. The first part of basically any expression is the method which in this case is linear and After Effects is going to auto-fill the rest of this so I don't even have to finish typing. I could just double-click on this. And it's going to put two parentheses in-between. These parentheses is where I have to put my argument. The first part of the argument is what we want to linearize or where we want to take our input range from. In this case, it's the exposition. I want to take a range of values from the exposition and remap them to an output range for the rotation value. So I need to reference that property and I'm gonna use the expression pick whip to do that with my cursor right there between the parentheses I'm going to select with my expression pick whip. The exposition were telling the expression what value to look at for the input range. Then I'm going to put a comma. And we need to put two numbers. This is our input range. So the beginning and end of the range of values for that property we just selected. So why don't we say 100 for the input, then we'll put another comma and now choose an input maximum value. So let's say 500. So we want to look at the exposition and use the range between 100 pixels and 500 pixels. Then we need to put two more numbers. So I'm going to put another comma here. And the next two numbers are the output range or what values we want these two numbers to be remapped to for the property we're applying the expression two, in this case the rotation. So another way to say this is when the exposition is 100 pixels, what do we want the rotation to be? Well, let's just say zero. We want no rotation at 100 pixels on the exposition. Then we'll put another comma and choose the output max. Or in other words, when the exposition is set to 500, what should the rotation be on the same layer, let's say 360. Because remember this property is measured in degrees. And that's another thing I love about the linear expression. We're basically converting units here. We're going from pixels to degrees. Now with any expression at the end of the line of code, you always finish it with a semicolon That's proper syntax. Sometimes After Effects can know what you want to do and not give you an error. But it's good practice to always keep your code clean and follow that proper syntax. So I'm gonna do it here. Now if I click off to apply that expression, we can see that the rotation is already set to one revolution or 360 degrees. And that's because my exposition is higher than 500, our input max. But let's put the exposition down to zero. So the squares off to the left side of the comp and start increasing this value. As I do that, you'll notice nothing is happening, but watch what happens as soon as I cross that 100 pixel threshold rate there, the square starts rotating. Nothing happens before that 100, but as soon as I cross that threshold, then this rotation property is being mapped directly to these values that I've set. I'm going to increase this all the way up to 500. And as soon as it crosses five-hundred, it stops rotating, it's finished its cycle. It's reached the input range of max value and nothing happens after that, but 100-500 or rotating. And I could change this range to absolutely anything. So I could say 250 and maybe 800. And now it's going to take longer for it to rotate as the square goes across the screen. Now this particular example isn't necessarily all that useful, but I think it's a great demonstration of how you can map one value to another value, even if you're using different units, we're going to use the linear expression a lot with the project. And like I said, I use it all the time in my work. So it's something that you should definitely get familiar with. So let's review the structure of this one more time before we move on. When writing a linear expression, we first write the function linear. Then we put open and close parentheses. Within those parentheses, we need to write the first part of our argument, which is telling the expression what property or value we want to linearize. Then we choose the input minimum and maximum value range separated by commas and then the output minimum and maximum range again separated by commas. So the first number is the input minimum, and that corresponds to the output minimum. The second number is the input maximum that corresponds to the output maximum. Now remember we separated the dimensions of the position. If I combine those again by unchecking separate dimensions, the expression is actually going to break and that's because our position is no longer a single value. It's made up of two. This is called an array. Whenever there are multiple values on the same property, it's an array of values and there's a specific way for targeting these arrays in an expression. So I'm just going to put the rotation on pause for a second and actually add an expression to the position property. So let me alter option, click on that and show you how to write an array. It's actually pretty simple. We just use square brackets. So I'm going to start by typing an open square bracket and a closed square bracket. You can put any number of values in an array as long as you separate them by commas. So in the case of this position property, the way to write an expression for each of these values is by writing one value and then comma, which we can see a comma right here as well. And then the second value. So if I said 100 comma 200, finished that with a semicolon and clicked off than that array of values is going to be plugged in directly to the position x and y values. It is now 100 by 200, just like I have written out here, if my layer was 3D than it would have three values in the array, I could go back into my expression at another comma and add a third value. So I could say 100, 200, 400. Alright, I'll turn off that 3D. And when counting array values, they begin at the number zero. So the first value in array is indexed at zero and then 12 and so on. So if I wanted to target a specific value in an array property, I could do that by selecting its index, and that's exactly what I wanna do with the rotation. So let me get rid of this expression. So our position goes back to what it was at and then go into our rotation expression and edit the argument where we have transformed our x position. Instead of that, I want to just use my expression pick whip to grab the position property. So it's now transformed opposition. But I again want to target just that exposition. To do that, I'm going to write an open and closing square bracket indicating that I want part of an array and then choose the index value for that first property, which again is zero. We start at zero and then one. So I'll just type a zero in-between those square brackets to let the expression know that I want the transform position property and the first value in that array. Now if I click off and move that exposition around, we're back to working exactly the same way as before. If I were to change this 0-1, then we're gonna be looking at the index number one, the y position. I'll apply the expression. And now as I move it up and down on the Y position, the rotation is being triggered. Nothing happens on the exposition. So that's the basics of how the linear expression works and how arrays of values also function inside of After Effects. Like I said, the linear expression is one of the most powerful expressions and most used expressions by me simply because of how many things that you can do with it, you can literally tie any value to another value and be very specific about what range of values are being translated. 4. Expression Controls: Now another thing that we can do to make expressions way more flexible is instead of putting in hard-coded number values, control them with expression controllers, expression controllers are actually affects. So if we go up to the Effect menu, down to expression controls, every one of these effects will give us values and in the same format as all of the properties that we have in after effects, e.g. the position property is an array of two values. We can get this exact same kind of an array if we go into the expression controls and grab the point control, that gives us an array of two values that is affected does not do anything in terms of visuals. It doesn't render, it doesn't affect anything on the layer. It just gives us the values that we can reference with expressions. So if I were to grab the property, pick whip and select that point control on the position property. Now the x and y position are being directly driven by this point control. So I can move it around and it's going to drive that position property. And since our rotation is still intact, That's also interacting with that controller. Alright, I'll get rid of that point control and remove this expression. There's a keyboard shortcut for that. If you just select a property and do Alt Shift E on a PC or Option Shift on a Mac that will remove the expression from the property for rotation, we could grab an angle control. So let me apply the angle control expression controller. And now we have an angle property. Again, it doesn't do anything on its own. But if I grabbed that property, pick whip and select the angle that will fill in the expressions for referencing that effect. I'll click OK to apply it. And now the rotation is going to be driven by that angle control, even if you're just using it to directly drive the transform properties. Expression controllers are great because they show up in the effect controls. And I'll be able to modify these things without opening up the layer. I just have it selected and the controls are now effects. Now I would say the most used expression controller in my workflow is the slider controls. So let me drag that out and I'll get rid of the angle control and double-tap the E key. This is the shortcut for revealing expression, so I'll double-tap E. That brings up all the expressions. And again, I'll remove the expression from the rotation by holding Alt Shift or Option Shift. On a Mac, a slider control as a single value, so not an array, and it can be used in all kinds of ways. Let's bring up the position by holding Shift and pressing P and starting to expression on the rotation one more time. Let's again linearize that position property and our Input Range, we'll just say 200 to 600, and our output range zero to 180. Now I do want to choose x or y position. I could write out an array, or I could just select all this text. And instead of going to the position property, choose the value in the array that I want to reference. So let's just say x, I'll let go and it automatically fills that in. I write as little expressions as possible and rely on the expression pick whip to do everything faster and more accurately. I'll apply this expression and we're back to where we were before. It's going to rotate between the values 200.600 on the exposition. But what if I wanted to change that input range without having to go into the expression, well, I could do that with this slider control. So instead of having 200 hardcoded in here, I'm going to double-click it to select that number. Then use my expression pick whip to select that slider value that will automatically fill in my code. My expression is getting longer and harder to look at, but this is what we just added. It's just referencing the slider control slider value. Now if I click off of this, my minimum input is going to be whatever the slider control is set to. Currently it's set to zero, but I could change this to say 250. And now when the x value hits 250, it starts to rotate. Or I could just interactively click and drag this down to a random number and then it'll start rotating from that point without having to actually update the expression. Our expression controller is what's updating that expression because it's referencing that value. So why don't we rename this slider input Min. And then let's duplicate that slider and rename it input max. You can probably guess what I'm about to do, but let's just change the value here and then go into our expression and find that input maximum spot, which is this 600 that we hardcoded in. And use our expression pick whip to choose that input max slider. Now with that applied, I can set my minimum and maximum values with these sliders rather than going in and editing the expression. Now, this range is very small right now, but I can increase this past 100. So if I make that much larger than we're going to see that rotation go across further distance. And I can even customize this range. If I right-click on the slider and go to Edit Value, I could change this range 0-100. The minimum and maximum values are huge. It's 1 million in either direction. So I could change this to 1080, and this will be the range of pixels of my comps resolution. So I could click Okay, and I can know that if I drag this all the way to the right, that's the right edge of my comp. If I drag this all the way to the left, that'll be the left edge of my comp. And now it's going to rotate 180 degrees from the left to right edge. But we don't have to stop here. Why don't we duplicate both of these, drag them down and rename these output Min and output max. Then I can go into my linear expression and grab that output minimum value expression. Pick whip that output men slider. Grab the output max value and expression. Pick whip the output max slider. I'll apply that expression. And now I have four individual controls for driving that linear expression. So not only can I control the input range of pixels for when did the rotation is triggered? But I can also change how much it rotates. So if I said 360, that'd be 360 degrees. Now it's going to rotate 360 degrees, 0-1080 on the input. But remember we had an angle control that would probably make more sense to drive the rotation. So why don't I get rid of those two sliders and bring out the angle control and point out that my expression did break because I deleted effects that I was referencing in the expression. The expression is no longer valid and it's not going to evaluate. But I can link these backups. So let's say output Min, duplicate, output max, go into our expression and find the parts of the expression to link up to these angle controls. Now, like I said, this is getting a lot harder to read since we have so much code crammed into this expression. But what we're looking for right here are the last two values, which are these two, I'm going to select between these two commas for the third value or the output minimum, and select that output Min angle control. Then grabbed the last value, making sure to leave that last parentheses because this parentheses belongs to the linear expression and grabbed the output max. Now those angle controls are linked up and it makes a little bit more sense. I could change the angle revolutions to one, and we're going to get the same thing as before. I can also change it to say negative one, and it's going to rotate in the opposite direction. But now everything is neatly organized in these effects rather than down in the expressions and makes it much easier to modify it after it's already been coded. And that is a very smart way to work with expressions to make them more accessible and even keyframe bubble, that's definitely something we're gonna be taking advantage of. What the class project. 5. Random Values: If you're filling a little bit overwhelmed by these expressions, don't worry, that was definitely the most complicated set of expressions that we'll be using in this course. And when I go to recreate my class project, I'm gonna be walking you through how I use each one of those very slowly in a lot of detail. So that by the end of the class, you should feel extremely comfortable with these expressions. But let's take a look at the next expression which is random. This is going to be a way to generate lots of random values. And then we'll take a look at how we can actually harness it and control it a little bit. So it's not completely random. So I'm again going to press R to bring up the rotation of this square and add an expression to the rotation property to generate a random value. We're just going to start by typing random. And After Effects will give us an autofill random with an open and close parentheses. Now there's some other expressions that deal with randomness, but this is the one we're going to focus on and it's pretty much the most simple. So I'll double-click on that to auto-fill it. Again, random is the method in, in-between the parentheses is where we put our argument. That random expression is actually pretty easy to understand. We just need to put a value between the expression. So I'm just going to type out 500. What this is going to do is generate a random value 0-500. I'll finish it off with a semicolon. Click off. And now if I hit Play, the random expression is going to generate a random value on every single frame of my composition between the values of 0.500. If I just step one frame at a time, you can see that this is changing completely randomly. And if I click this little graph button right here, It's called Show post expression graph. Actually, I'll open up my graph editor first with the rotation selected, you'll see that it's just a straight line, meaning the value is not changing. But as soon as I click on that post expression graph, we're going to see what that rotation value is after the expression has been evaluated. So we're getting a completely random number, 0-500 for every frame of the animation. Now if I come down to this menu right here, I can say show expression editor and that's going to show me the expression written on that as well. And I can even go in here and edit it. So instead of saying 500, why don't I say 25 and apply that. Now my range has changed and we're going to get a random rotation, 0-25. But let's say you wanted to devalue to be between two specific numbers, not zero and whatever you type in here. Well, we can actually do that by just going into the argument and putting two values in. So I'm going to put a comma right after 25 and say 250. And now the expression is going to give me a random number, 25-250 for every frame of the animation. I'll click off of that. And again, my values have changed. If I play this back, it's going to give me that random value between these two ranges. And I could do negative numbers as well. So I could say negative 18180. It's totally customizable to whatever values you want to work with. But what if we didn't want this to be updated on every single frame? Because this is really random. It's very chaotic and maybe not quite what we want. Well, let me get rid of this graph editor and go back into our expression. And I'm going, and I'm going to drop this expression down one line above it. I'm going to write another expression called posterize time again After Effects will auto fill. So I'll double-click on this. The method is posterize time and the argument goes between the parentheses, just like everything else we've looked at so far. What this expression is going to do is tell this code how often to evaluate, and it's measured in frames per second. So I'm working into 24 frames per second comp, but if I say six between those parentheses and finish it with a semicolon, then whatever I put in this expression is only going to be evaluated at a rate of six times per second. So if I play this back, you'll notice that the randomness is much slower. The frame rate of my comp hasn't changed, It's just the frame rate of that expression and that number can be whatever I want. So if I wanted a little quicker, I could say 12. It also be clever and add in a slider control, rename this FPS. And instead of putting a number here, I could expression pick, whip the slider. And now whatever I put in here is what the frames per second is going to be on the posterize time. So since it's set to zero, it's just going to give us this random number and hold onto it. Or it could say ten. Again, controlling these values with the effects, instead of having to go back into my expressions every time, it's just a smarter way of working. Let's go back to the graph editor and you can see that the posterior is now holding onto those values by whatever frame rate we have. So again, if I turn that down to six, that's going to get less frequent. If I turn it up to 20, We're gonna get a value that changes a whole lot more. But what if we change our expression a little bit, instead of being just random, I could type out another expression. So let's say Gaussian random. And I want to point out, since I didn't finish typing and I left part of the expression. If I double-click to auto-fill, it's going to generate those two parentheses. And I don't need those anymore because I already have my values over here. So let me get rid of those two and then apply the expression. Now it might not look like there's much of a difference here. But if I put this up to a rate of 24 frames per seconds so that it matches my comps frame rate. And I zoom out a little bit on my timeline. The difference between Gauss random and random is a little more evident with random or getting an even distribution of these random values between our range. But if I undo back to Gauss random, you'll notice that a lot of those higher and lower values are not showing up as often. The easiest way to think about this is that it's basically just choosing random values from the center of the range a lot more often than the extreme ends of that range. So a randomness isn't as evenly distributed. Now, I actually want to keep it at the regular random. So I'm going to undo until we're back to where we were. But I just wanted you to be aware of that expression in case you ever wanted to play around with how random things were getting. Alright, let me get out of that speed graph. And let's say that you're doing more than just getting a random value in this expression just before Random. Why don't we say time, times 20 plus our random expression? Now what's going to happen is it's going to take the time value and multiply it by 20, add it to the rotation, and then add our random number to it. So if I click off to apply and then play this back, it basically looks exactly the same. The rotation is totally randomized. But if I were to get rid of this part of the expression right here, I'll just cut that out. And now we just have time, times 20. Then we're gonna get a slowly rotating square. And if I change my posterize time value down to zero, then we're not gonna get any rotation at all. Because this is globally changing the frames per second on any expression written in here. But let's say I wanted the rotation to be happening, but add a random value that does not change to it. Well, that's where the seed random expression is going to come into play. So let me drop my expression down and separate it from the posterize time a little bit just so we can have a visual separation. And I'll paste this expression back in. So we're back to having a random number implemented. But because our frames per second is set to zero, we're not getting the rotation driven by time. Well, let's go back into the expression dropped down a line and type a new expression as seed random again, After Effects will auto-fill. Seed random is the method and the argument goes between the parentheses. What this expression is for is choosing a starting point for the randomness. Because After Effects is a program on a computer, it's actually really difficult to generate truly random numbers. Anytime you generate a random expression, it's all based on a seed value or a starting point. Putting in seed random allows us to choose that starting point. If I'd said seed random one, this could literally be any value I want. I'll finish that with a semicolon and click OK to apply. Then my square is gonna be rotated at a specific random rotation. But it can actually do more than just give us a random starting point if I change my frames per second backup to 24 to match the comp, we'll get that randomness. But with this seed random, I can add to our argument. So I'm going to separate that value by a comma. And this next section is the timeless part of the argument. Timeless means. Is it going to update on every frame or not update on every frame? By default, it's set to false, which means that it's going to update on every frame even if seed random isn't part of my expression. But because it's here, I could type true, telling the timeless argument to be true, meaning it will not update on every frame. So if I click off now, even though my posterize time is set to 24 frames per second, my randomness is not going to update and we're going to get the time times 20 part of our expression added to a random value between negative 18180 and that value doesn't change. So if I go to the beginning, my random value is negative 10.2 and that's being added to the rotation. And if I duplicated this square, we're gonna get a different random number. But there's still both going to be rotating at that rate of time times 20. I could do this as many times as I want in each one of these layers is going to get a random starting point. And let's say I don't like the starting point of this square. I could go into the expression, change my seed random value to something completely different. Say 64, apply it and it's going to have a different starting point, whatever this value is. We'll give that randomness a different starting point. And again, I could link this to a slider instead. So I could say random seed, grab that value and use my expression pick whip to grab the random seed slider. And now we can update its starting point randomly with a slider instead of going back into the code. So that's how we can generate random values, add them to other things, slow them down and even tell them not to change on their own. Random values aren't necessarily all that useful or you can't use them for that many things. But when you start to implement controls, then you can really dial in these random values to be whatever you want them to be. 6. wiggle();: The last expression I want to look at here is probably one that you've run into before. It was definitely the first expression I learned how to use. And it's the wiggle expression. So I'll bring up the position property by pressing P with that layer selected and Alt or Option, click on the stopwatch to add an expression. To write a wiggle expression, you start by typing wiggle, and then I'll auto-fill those parentheses again, just like before we have the method of wiggle and in-between the parentheses is our argument. And they're actually can be a lot of possible values that go between these parentheses. But we're going to start with the first two that are the most common, which is frequency and amplitude. Another way of saying that is how often this will wiggle and how much it wiggles. The frequency is measured in times per second. I'm going to just say 141 times per second. Then I'll separate that with a comma and the second value, the amplitude is how much it can wiggle. So let's say 250, and this is 250 pixels. Since we're working with the position property. I'll finish that off with a semicolon and apply it. And now what's going to happen is the layer is going to wiggle around. And this is different than just a random value, 1-250, because it's actually changing value. It's not just putting a random numbers in every single frame. It's sliding the values around at a rate of one times per second with a maximum range of 250 pixels in either direction, either positive or negative. And that's why it looks like the layers actually shifting around instead of just randomly sporadically appearing all over the place in the comp. If I wanted this to move a little bit faster, I could change it to two times per second. If I wanted it to move slower, I could say 0.5 times per second. If I wanted it to move less, I could change this down to say AT, and maybe we'll say five times per seconds. So now it moves quickly, but not as far. Now if I cut this from the position and instead added it to the rotation, then we're gonna get this wiggle applied to the rotation. We just have to remember that we're changing units. So we're going at five times per second and a value of 80 degrees. Now what if I only wanted the Y position to wiggle? Well, I could separate the dimensions and add the expression to just the white position. But I want to show you how to do this on an array. So let's cut that expression off the rotation one more time. Go back into position. And what we need to do is write our expression in an array, which if you remember, starts with an open square bracket. For the x value, I want that to be whatever I set this value to be right here. Instead of coding and a value, I'm just going to reference this value. So I'll use my expression pick whip to select that specific property, separate the array with a comma, and then paste in my expression. Now I need to be careful here because there's a semicolon at the end that I also cut and paste. I need to go just before that and put a closing square bracket to finish off my array. That turns green and it highlights the first square bracket letting me know these two correspond together and that it's valid. Now, if I click off to apply the expression, I'm actually going to get an error. Clicking on this tells me that there's an array piece that can't expand to more than one value. This is something that's unique to wiggle. You saw that I applied wiggle to this property that has two values in its array. And it worked, it wiggled on both the x and y position, and I applied it to the rotation and it also worked there where it's a single value. So the wiggle expression is a little bit more universal in that way. But if I want to target a specific value in an array, I have to add one more thing after the wiggle expression, which is another open and close square bracket. And then the index of the part of the array that I want to target. So in this case it's a value of one. Remember we count from zero and then one in an array. I want to target the y value indexed at one. I'll put one between those square brackets. Click off to apply. And now the value is only going to wiggle on the Y position. I can move the exposition around still since I referenced that code right here. And it will still wiggle. And that's really all there is to the wiggle expression. Again, just like before, it's a smart way to work to tie these two sliders. So if I go into my expression controls and add a slider, I'll rename this frequency or just FREQ for short. I'll duplicate this one, call it AMP for amplitude. Then I can go into here and choose the frequency value for the wiggle expression. Use my expression pick whip to grab that slider. The second value is the amplitude. I'll use my expression pick whip to grab that amplitude. Apply the expression. And now there's no wiggle because both of these values are set to zero. So if I change this to two and I change my amplitude to say 250, then it's going to update at a rate of two times per second and 250 pixels. Now, this expression just got a lot harder to read. And in the next lesson we're going to take a look at variables which will help us make this much more readable. But before we do that, I'm just going to rewrite this expression back to the way we had it originally. I'm going to say wiggle and then say one times per second and 250 pixels and the max, if you remember, I said that you can write more than just these two values in this argument. And it can be difficult to remember what all of these possibilities for arguments are in an expression. And there might even be expressions that you don't realize have more possibilities than what you've been using. But this little arrow right here and the circle is the expression language menu. And if I click on that, this is the entire directory of possible expressions that you could write in After Effects. And if we go to the property menu, that's where wiggle is. So let's take a look at what's written here because it's a lot more than just wiggle. At the beginning we see frequency and amplitude and those are the two values that we put in. But after that is another comma, and we have octaves equals one, another comma amplitude multiple equals 0.5, comma t equals time. What in the heck are all of these values? Well, honestly, they're not used very often and I'm not going to get into exactly what to do with them here. But this is a perfect way of being able to see what each part of an argument is actually four, because if we go to say the random numbers category, then we're going to see the random expression. And there are a couple of ways of writing it. The way we've been using it is the minimum value or array and the max value or array. And at the top we have the seed random where we first put in the seed value and then we put the timeless value true or false. So this is a great reference for if you can't remember, what values should be in the arguments and what part of the arguments they should be in. If you're curious what all these other wiggle arguments can be, then you can just Google the wiggle expression and find a guide that explains what octaves are and what the amplitude multiple is. But like I said, those arguments are very rarely used or needed in after-effects projects. But that's it for the wiggle expression. It's something that comes in very handy for generating random motion that isn't so sporadic, like the random expression. 7. Variables: Alright, I've gotten back to where we had this expression that is much harder to read for the wiggle animation that we added to this square. Remember the frequency and amplitude are being driven by these slider controls that I added up in the effects panel. I would like to make this much easier to read and interpret and even modify by using what are called variables. So if I go to the start of the expression and drop down two lines, this is generally how I like to organize my code. I start with variables, I separate it with a blank line of code, and then I write my expression. I think the easiest way to think about a variable is that it's shorthand for the rest of my expressions. And we start writing a variable by typing VaR, which is short for variable in After Effects, is going to highlight that in blue because it knows what you're trying to write here. So VAR, now I need to actually define the variable and give it a name. Variables can be named anything as long as they don't have reserved values in it. Like I couldn't name a variable wiggle since that's already an expression method. But I could say VAR FREQ for frequency, and that's a perfectly valid name. And then I'll put an equals sign. And now I have to actually define that variable. So I've given the variable a name, but I have to say what that name is going to represent. Now. Well, I want frequency to represent the frequency slider. So I'm gonna use my expression pick whip with my cursor right there, grab that frequency slider and that affect expression is now going to be referenced anytime I type in frequency or FREQ in my expression. So I'll finish that line with a semicolon that is very important. And then I can go down to my wiggle expression and find where I have that same code right here. And instead of using that code, I can just type FREQ for frequency. Now it's very important that I follow the proper case because frequency with an uppercase F actually is part of an expression. It's part of JavaScript. As long as I'm using the same case, then it's going to know I'm referencing this lineup here. So that's great. What if we define a variable for the amplitude as well? Well, let's drop down another line. Type, VAR AMP for amp equals, and then use that expression pick whip to grab the Amplitude. Finish that off with a semicolon. Grab that part of my expression that was referencing that same slider in instead of that code type AMP for amplitude will already has gotten a lot easier to look at, but we can take this a step further and let's go to the top, dropped down a line, and write another variable, VAR. And we'll call this x for x position equals. Use the expression pick whip to select the exposition finished now with a semicolon, and then select that part of my expression and replace it with x. Now if I apply this, the expression evaluates to the exact same result, but it is so much easier to look at. But I don't even have to stop there. Let's go back into our expression and add another variable, VAR. And I'll just call it W for short for wiggle equals. And then I'll cut this expression right here, leaving that final square bracket for the array. I'll just cut that out and paste it into that variable, finish it with a semicolon. And then down here in my array type W. So now my array is literally two characters, x, W. I'll apply it. And again, we have the exact same result, but it's in so much more of a structured format that it is so easy to read, I can just look at what is my frequency. Oh, it's the slider for frequency. What is the wiggle expression? No, that's right here and it's isolated so I can modify things. And you'll even notice that I'm using variables within variables, which is another extremely powerful use of variables. It simplifies your code so much, makes everything more organized, easier to read, and easier to update. I use them all the time and I'm constantly finding new ways of making them even more useful. But we're for sure going to be taking an advantage of this technique when we go to animate the class project in the next lesson. 8. Global Controls: Now that we've learned how to use a handful of expressions and control them in an intelligent way. Let's start actually animating our class project. Now if you want to just use my art work and recreate my project, that is totally fine. Just head over to the project and resources tab. Expand the little menu down. Look on the right side and you'll find it the After Effects project with all of the artwork already set up exactly like this, you can follow along with me. Or if you'd rather create your own artwork, then you might want to watch through the rest of the lessons just to see how I handle my layers so that you can be more informed when you go to create your own. But let's just quickly review what the final animation looks like. There's a lot going on here. The tray lifts up off the ground, all of the pieces of sushi and all the other things on the tray float up even higher. Each layer is randomly distributed and they kind of levitate there hovering up and down, as well as rotating before all coming straight back down in resting on the tray, back down on the ground. There's also an animated texture that wiggly boiled edge that's applied to every single layer. And yes, we're going to derive all of this animation with a single set of four keyframes. So let's get started very focused and just animate this tray. I'm going to solo that layer along with the background and the shadow so that we're not distracted by the other layers. Now obviously I could animate this with key frames directly on this layer, I could set a position keyframe at the beginning, go forward a couple of seconds and have it lift off the ground. Easy ease these keyframes, then copy them, paste, right-click Keyframe Assistant, time-reverse keyframes. And then the tree is going to lift off the ground and drop back down. But if I did this for every single layer, then I'm going to have keyframes for every single layer and not just on the position property because I also want to animate the rotation with 15 different layers that all need keyframes on multiple properties. This could get really complicated really quickly and be very difficult to make adjustments to. On top of that, if I move a layer around with the position after it's had keyframes applied, it's going to add more keyframes and it's more difficult to make adjustments to all of those keyframes at once. So not only am I not going to add keyframes on every single layer, I'm not even going to animate using the transform controls for those layers. Instead, I'm going to use the Transform effect and I'll show you why. Let me search for transform over here in my effects panel and apply this. This gives me a set of transform controls as an effect instead of the actual transform controls for the layer. So I can recreate that same animation. Just add a couple of keyframes, shift this off the ground. Copy paste, copy paste and easy ease. And we have the exact same looking animation. But the benefit is now, let's say I wanted to reposition this. I could grab the layer because you can see the outline of the layer is right down here. Still. Move it around wherever I want in that animation is all relative to the layer because the transform controls aren't affecting this transform effect. I can also copy and paste this to all the other layers. So actually let's do that right now. I'll undo back to where that tray should be un-solo. These layers, copy this transform effect and paste it across all the other layers. And now they're all going to move together. Now if I press U, we're still in that same issue of having keyframes on every single layer. That's not at all what I want. But the point is I set one set of position keyframes and applied it to all of these different layers. And it didn't mess up any of their placement because the effect is relative to the layer that you're applying it to. Alright, let's undo back to where we were. And again, I don't want to derive this specific effect with keyframes. Instead, I want to create a global controller that has all of my expression controls and tie all the properties that I want to animate to it. So I'll get rid of those keyframes and go up to Layer New Null Object. If you're not familiar with the null object, this layer doesn't render, it, doesn't do anything on its own. It's just there as a container for controls. Basically it has all the same transform controls as any other layer. But I could use it to parent other layers to it and then transform that layer and it will control everything that was parented to it. Or we could use it to hold expression controls and then tie individual properties to those controls. So I'm going to rename this global controls. And I'm just gonna change the label color to green and slide this off to the side. It doesn't have to be up to the side, but I figured it's less than the way if it's not over top of my artwork, I'm going to give myself a little bit more room down here in the timeline since we're gonna be working with all of these layers. And one trick with expressions that can save you a lot of time is a feature called copy with property links. So if we go up to the Edit menu and down to Copy with property links, what this is going to do is copy whatever you have selected with all of the properties inside of it, tied it to the source via expressions. So let me add the transform effect to my global controls null object. Again, it's not gonna do anything visually because the null object is not visual. It's just there to hold the information. But now that, that transform effect is applied to the null, if I come up to Edit and go down to Copy with property links and then paste that on top of the tray. It's going to replace the effect that was already there. But all of the properties of that effect are now tied to the instance that's on the global controls layer. So I can move the position around, I could rotate it and it's all tied to this global controls null. The reason that's so powerful is because if I paste that to all of my other layers and un-solo everything so we see all of the artwork in my comp. Then I can go to my global controls null and adjust a single position control to modify all of them. Now again, this isn't that impressive because I could have just parented everything to the null object and move the position around. But the point is, copy with property links is a huge time-saver for linking up properties to a controller. Now, I know for a fact that this rig is going to get pretty complex with lots of controls and the transform effect takes up a lot of space. So to make this cleaner, I'm going to use expression controllers for custom points of control and then just link the properties that I want to animate to them. Instead of having all of these transform controls, taking up so much space in this layer. So I'll undo again to get rid of the transform effect on all of those other layers. And I'll even remove it from these two layers with them selected, I'll press Control Shift E or Command Shift E to remove all effects. Then I'll bring the Transform effect of back out onto the tray. And on the global controls null. I need to add a slider. So let's go to the slider effect, apply that. And this is going to be the y position. So I'm just going to call it y POS for short. I'm going to abbreviate as much as I can while still making things descriptive. And this is what I want to drive the position of the tray. So let's open up the transform effect. And I'm going to solo this position property. So with it selected, I'll double-tap the S key and then I'll press E to bring up the effects for the global controls so that that slider is visible as well. Now I want to write an expression on the transform property. So I'll Alt or Option, click on it and start writing my array. Since this property is an array, and just like we learned, I'm going to use variables to make this easier to read. So I'm going to start with VAR x equals. This will be my x value. I want this to be whatever it's set to. So I'm just going to use my expression pick whip to grab that x property and autofill in that code, finished that with a semicolon, drop down a line and type out VAR y equals. And I want this to be the Y slider that's on the global controls null. So I'll use my expression pick whip to select that slider, finish it with a semicolon, and then dropped down two lines because I want to actually write my array. Now, I'm gonna give myself a little bit more room here. And we start writing our array with the open and closed square brackets. Within those brackets, I'm going to say x comma y, and finish it with a semicolon. So now my array is extremely simple. All the code is stored in the variables and it's very easy to identify what's going on. So if I click off to apply, my tray is going to lift off the ground and it's set to a value of zero because my y position slider on the global controls null is set to zero. But if I increase this or decrease this, it does drive the position of that tray on the y-axis. Now I would like this to start at a value of zero, but not have it lifted the tray off the ground. And I can actually do that if I modify the anchor point. If I disable this expression just by clicking on that equals sign, will see that the anchor point in position are 540 by 540 because this is a shape layer or Vector Layer and vector layers and after effects are basically working in a coordinate space that's relative to the comp size. So even though the tray is down here, the effect by default centers the anchor point in position which to it is in a space that is 1080 by 1080. If I change my anchor 0.200 and my position to zero-zero, then they're both going to be in the top left corner of the comp, but my tray is going to be left exactly where it was, so long as the position and anchor point are the same, that's what will happen. Now if I turn my expression back on, my values start at 00, but the trays rate where it's supposed to be, That's exactly what I wanted because now I can just adjust this value to whatever I want. Know exactly how many pixels I've removed it and zero it back out to get it to where it was at its resting point. Now I could animate this slider if I go forward and just change this value to lift it off the ground a bit. Copy paste, copy paste and easy ease with F9. Then we're back to having a levitating tray and then it falls back down. But again, we're going to be animating all of this based on one set of keyframes. And this is not cutting it. This is not going to drive all of our animation. So instead of that, I'm gonna get rid of those keyframes, set this back down to zero, and add another slider. In fact, I'll just duplicate this one and I'll rename it animation. I'm going to move this to the top of the effect stack because this is what's going to drive everything. This is what will have keyframes on it. And I'm going to think of this slider in terms of percentage, zero to 100% completion. So I'll start by adding a keyframe near the beginning. We'll say ten frames forward at a value of zero. Then I'll move forward a little bit. I'm not too worried about timing right now, so I'll just move it forward a couple of seconds, change it to 100%. Press U to bring up my keyframes. Copy and paste those. Right-click Keyframe Assistant time-reverse keyframes. Now nothing visual is happening. We're just animating a value. If I go to my value graph, you can see that it's changing 0-100 holding and then going from 100 back down to zero. Now we get to start implementing the linear expression. Remember the first thing that we learned in this class to map values to that set of keyframes. Let's go back into the expression for the position and add another variable at the top of the list will say VAR a for animation equals this animation slider finished that with a semicolon, and now we can write out a linear expression to map the y-value to that animation slider. And I'm going to do this inside of another variable, so I'll drop down after variable y. And since this is after that variable, I can actually reference anything before it. So let's say VAR new y with a capital Y equals, and then I'll start typing my linear expressions. So linear and it's going to auto-fill. I'll press Enter to get those parentheses. And what do we want to linearize? Well, the animation slider, this one right up here, which we store it as a variable a. So I'll say a is what we want to linearize. Then we need our input range, so comma and our input range is going to be zero to 100. Because remember, I'm thinking of this as a completion of zero to 100%. So we'll say zero comma 100, and then another comma and our output range. Well, when the animation is set to zero, I also want the position on the y to be zero, so I'm going to type another zero. Then I'll put a comma. But when it's at a maximum value of 100%, I want it to be whatever I've set the y position slider to, which is this y variable that I've stored right above it. So I'm going to type in the variable Y from my output maximum. I'll finish that all off with a semicolon and then change my array down here instead of to just x, y to x a new y. Now if I apply that expression, nothing actually happens, but that's because my y position is set to zero. If I change this value to something else, it's going to lift up off the ground. So maybe we'll say negative 80. Watch what happens now once I scroll back past 100, because we're linearizing that value, it's going to animate between zero and y. So let's add some easing to these keyframes and I'll go into my graph editor and just make it a little bit more extreme so that it kinda hops off the ground a little more quickly and then eases nicely to a stop at the end. Maybe it even eases a little bit longer. Now the tree is going to pop up off the ground, hover for a second before coming back down. And I can easily change the distance that it lifts off the ground just by changing that wide position slider. No matter what I change it to though, it's always going to come back down to its resting point on the ground, a value of zero. And this is the technique that we're going to use to drive virtually all of the animation in this scene so that everything is tied to this one animation slider, but still easily customizable with the effect controls that we build here. 9. Controlled Randomness: Things are looking great so far, we've now dynamically tied the position value to a slider that is then triggered an animated by a different slider. And this is a one-to-one ratio right now the y position value is how far that tray is lifting off the ground. But that's not what I want to happen with all of the things that were on the tray. I don't want it to just lift up that far. I want them all to hover up off the tray, all randomized at different positions, and then do even more after they're up in the air. But let's focus on how we can get everything else off the tray and randomly distributed on that y-axis. Well, first things first, I'm going to rename this transform effect Y position just for organization. Then I'm going to copy that effect, close this layer up and solo the first item on the tray, which is this role to layer. I'm gonna zoom in here so we can see it nice and clearly. And I'm gonna paste that transform effect to it. So now both the tray and the sushi roll are being driven by this global controls null object, those keyframes going 0-100%. Well, I want this sushi roll to not only lift off the ground but also off the tray. So let's take a look at that position expression. In fact, I'll just double-tap the E key to bring up the only expression on that layer so far. And think about what we could do to introduce some random offset. Well, obviously we could add in the random expression, but we don't want it jumping all over the place, so we're definitely going to need that seed random. So why don't we start there, we'll just drop down a line and I'll type out seed random. And in the parenthesis will just say one comma, true. Remember the second part of the argument is the timeless arguments. So if it's true, that means it is timeless, meaning it will not apply. It just gives us a random value. Finish that with a semicolon. And then I'll actually just dropped down another line to separate the actual array from everything else. So that way I just have some visual separators. I have my variables, I have my seed random and I have my expression. Now, seed random on its own isn't going to do anything. We need to actually implement some randomness. If we think backwards a little bit new y is a linear expression, but it's based on this slider right here, the y position, if you remember, we zeroed out all of these values, so they're going from zero to whatever it's set to. And that's actually really convenient because I could use it as a multiplier with a random number. Because at the start of the animation, when the position is set to zero, anything I multiply that by no matter how random the number is, will always equal zero, meaning that the layer will not move until the animation slider's value has increased to something higher than zero. So if I jump into my linear expression and find where we called out that y-variable. I could say y times and then a random number. So let's say random and we'll just say a value of two. So it's going to generate a random number 0-2 and multiply that by whatever the Y slider is set to. So I'll apply that and see what happens. Well, the tray is lifting up and down and the sushi is lifting up and down. But as soon as that animation keyframe has started, then my sushi is bouncing all over the place. And I'm actually glad this happened because it's something that I should have pointed out sooner. Seed random has to happen before any random numbers are generated in your expressions. So I'm actually going to cut this seed random from where it is and put it on an isolated line before all of my variables. That way we're setting the seed random and telling it to be timeless before any of those numbers are generated. So now that we've cleared that up, let's reapply that expression and play it back. Now you can see that the tray and the sushi roll are moving at different rates. But why is the role not moving as much as the tray is? Well, it's because we're generating a number randomly 0-2. So if the randomly generated number is less than one, it's going to be moving that sushi roll less than the actual tray is. So as a safe guard, we could say the random numbers should be 1-2. That way we're going to guarantee that it will move at least as much as the tray, but could move up to as much as double the amount. So let's apply that. And now my sushi roll is higher because we've gotten a new random number that's 1-2. Now, just like everything else, I don't like hardcoding in these numbers. So I'm going to add in two more sliders. Let's duplicate the y position and rename this y position Min. Duplicate it again and say y position max. And I'll set these to the same values of 1.2. Then we'll go back into our expression, add another variable just before the new Y-variable. And I'll say variable y Min equals, and I'm using what's called camelCase in this JavaScript code, where the first part of a variable is lowercase, but you set every corresponding word or separate component of your variables name using an uppercase letter. This isn't necessary, it's just common practice. So variable y Min equals, and then I'll grab that slider semicolon, drop down a line and do another variable this time y max equals and grab that wide position max slider. Now we can call those two variables out down here and say y min and y max. With that applied, I can now set that y position minimum and maximum value on the fly without jumping into that expression. So let's say that I want this sushi roll to be able to go five times higher. I'll set the maximum to five. And now we'll see that lift up much higher as these keyframes take place. Now there's gonna be a lot of meat copying things back and forth between all these layers just so you can see how it's all affecting everything, but you don't have to follow along with that. I just wanted to show it as a visual. So I'm going to grab that y position, which remember is a transform effect and paste it on all of these other layers. And then give myself plenty of room here so we can see exactly what's happening. Let's grab those key frames so we have them visible. Everything starts where it should at the beginning and then lifts off the ground, lifts off the tray and is randomly distributed across that y-axis. But I have the y position minimum and maximum sliders that will allow me to dial in just how far up these things are allowed to lift, as well as how far off the tray they're allowed to lift. So again, if I set this down to zero, things could possibly slide off the trade downwards. It needs to be at least a value of one for the minimum amount of position lift to be still on the tray. But being able to interactively adjust this now is so much more visually apparent what exactly is happening. But let's say I don't like the way that this is all arranged and I want to change the random seed. Well, that would be a lot of work to go through each one of these effects and change that. So instead, I'm going to duplicate this slider and rename it. Random seed will go back into the expression on that first sushi roll. Add another variable for that random seed slider, which I'll do right after the WiMAX, will say variable SR for seed random equals, and then use my expression pick whip to grab that random seed. Now that wasn't expanded so I can actually grab it and that's going to cause an error. But if I give myself a little bit more room and open that up so it's visible. Then I can go back into that SR equals and then grab that random seed. It actually would make more sense to me if I named that our random seed. And I just need to make sure that I finish that off with a semicolon and I can incorporate it now into my expression. The problem is I need to incorporate that up at the seed random line, which is before that variable has been defined. So this is where I need to make an organizational decision. Do I want to move that seed random it just after the random seed variable? Or do I want to move the random seed variable above the seed random? Well, in this case I think I want to do the second option, put the variable up at the top, just above seed random and keep everything else nice and organized down here. I think that's a little bit cleaner and it makes sense to group things together that go together. So I'll apply that, copy that effect one more time over top of all the other layers. And now I have this global random seed slider. So let's give ourselves some more room again and watch what happens when I change this value. Now I can call that variable into the seed random expression. So I'll just replace that one with RS and then copy this effect to all the other layers and then see how that affects things. Now that I have the random seed global control watch what happens when I shift this around. Every single value that I put into this slider is going to generate a different set of random values for the y position, again, all starting and ending at the resting position of zero. But this is a way to be able to randomize all the elements of my illustration if I'm not happy with how things are being offset. So that's all great. But I still don't have that much control over spreading things out manually. So what can we do to add even another level of control? For instance, this piece of sushi is below the rice that it was sitting on top of. So how can I dial that in more precisely? Well, let's find that piece of rice. I think it's rice to double-tap the E key to bring up the expression. I want to add a slider, but not to the global controls layer. I want it to be on this specific layer. I'll bring that slider control out to the top of my effects stack and call this Y offset. And I want to incorporate that into this expression so we need to first put it in as a variable. So let's get enough room to be able to see that slider and the expression. And I'll type another variable just before the new y variables. So we'll say variable Y offset equals and then use the expression pick whip to grab that slider, finished that with a semicolon, and now I can incorporate that into my linear expression. So again, in the part of the argument that corresponds to the 100% slider, I want to add something right after that. The order of operations applies to expressions. First, multiplication and division is going to happen and then addition and subtraction. We're going to take the white slider value, multiply it by the random number between the minimum and maximum values, and then add the y offset. And because this is happening on that part of the argument, it's not going to affect the layer when the animation slider is set to zero. So let's apply that. Look at this comp in full view and change that offset. Now I can very precisely move that better price around to place it absolutely anywhere that I want. And even if I put it above where it should be, when I go back in time, back to where the animation slider is set to zero, it goes back to its correct position. Now that I've set that up for this one layer, I'm going to reset the value back down to zero, press E to bring up the effects on that layer. Copy, and then paste them to all the other layers. But I have to be careful here because watch what happens if I paste it on everything. Alright, I've just pasted and I'll press E and you'll notice that it did not replace the y position transform this time, I've never figured out exactly how After Effects chooses when to replace effects and window paste it duplicates. So from now on I'm just going to be removing all of the effects before pasting. And to do that again with all the layer selected Control Shift E or Command Shift E to remove all the effects. Then I'll paste one more time. And now every layer has the Y offset slider, which means I could randomize this using my global controls null. Let's actually find where that is. So right here I'll move it to the top of the layer stack, go to the Effect Controls and change the random seed. So I'll find a frame where things are distributed fairly well, but one or two things is not lining up that much. So here we are again where this bed of rice is not where I want it to be. So I'll go to rice too. I'll grab that slider and I'll offset it downwards a little bit. It doesn't have to be a lot. This is purely up to me how far I want things to move, but maybe this chopstick is a little low. I think chopstick to that's the one we want. So maybe I'll bring that one up a little bit and then the sauce ball right here, I could grab that layer and push it down more so it's not raising up so much. So now what I'm doing is randomly distributing all of my layers, but confining them to the minimum and maximum values that I've set up on these sliders. Again, I can affect this however I want at any time. And it's going to affect everything. I can reposition individual elements for every single layer to have more precision. And I can completely randomize everything just by changing the random seed. It's all being driven by this single set of four keyframes on the animation slider, starting and ending exactly where I want it all to be. 10. Layered Motion: Now that we've successfully positioned all of these layers and been able to randomly offset them, but still have some control over the placement. Let's move on to the next stage of the animation, which would be having these items levitate, hovering up and down as well as rotating a little bit so that it's not just static from point a to point B, back down to point a. To pull this off, we're going to take advantage of the transform effect. Again. Why don't we just start with the top layer here. So the wasabi, I'll press E to bring up the effects. And I'm going to duplicate the y position, which remember is a transform effect. So I'll duplicate that, rename it to why wiggle, and then go into the position property. And let's just modify this expression. We're probably going to reuse some of the same variables so I don't need to completely remove it, but I'll just zoom in here so we can see that with savvy. And I want this to wiggle up and down, but only when those keyframes, again are not set to zero. So to make this easier to look at what I'm actually going to do is just turn off the y position effect and disable the expression on the y wiggles. So now that we're savvy doesn't move at all. In fact, let's just shut off all the other layers so that we can stay focused on just that was sobbing. What we want to see for this to be just gently bobbing up and down. And we're gonna do that with the wiggle expression. So let's go back into the effects and into the wild wiggle. I want to modify that expression that we have on the position this time instead of being directly tied to this y position slider, I want to tie it to a wiggle expression that has frequency and amplitude sliders. So I'm going to select all of these effects up here just by clicking on one of them, then pressing Control or Command a to select all, and then just collapse them all at once. And then I'm going to duplicate the last slider and I'll rename this y position frequency. Duplicate that and name it y position amplitude. I'll reset these both download. Why don't we just set some default values of one and maybe 100? And let's think about how we're going to change this expression. Well, we know for sure we need the frequency and amplitude in here. So why don't we go ahead and put that maybe right after the y. So I'll say VAR FREQ for frequency equals, and grab that slider. Then VAR AMP for amplitude equals and grabbed the amplitudes slider semicolon. Now we can access both of those in our expression for the y part of our array, we're going to want to wiggle, but only when that animation slider is corresponding to something other than the zero percentage. So we're probably still going to use this new y linear expression. It's just this part of the expression is going to change. Instead of being based on that Y position slider, we're going to base it on a wiggle expression. So I'm going to type out a wiggle expression. So wiggle, then between the parentheses we'll type of frequency and amplitude. So we're using those two variables that we just wrote to drive the wiggle. And remember, because this is going into an array, I need to put a open and closing square bracket and the number one for the number one index in that array, 01. And that goes outside of the parentheses for the wiggle expression. Now there is an issue. This linear expression needs a closing parentheses, so I need to go after those square brackets, put a closing parentheses and then it should be valid. Now to clean this up a bit, I can get rid of that y variable because I'm not referencing it anymore. I also don't need the y minimum or maximum values or the Y offset. And I don't even need this random seed or seed random expression. So I'm just going to select all of that. So I'm left with just the variables I need access to. So what we have now is x, which is the actual x position and y being our new y, which is the linear expression taking the animation slider 0-100 and remapping that value to this position on the why from zero to wiggling that frequency and amplitude value zonal dynamically change from not wiggling at all to this wiggle expression. So let's apply it and see what happens with those keyframes and view down at a value of zero, nothing happens. As soon as it gets up to 100, it starts wiggling, and then it comes back down to zero. That is exactly what we wanted to happen. And now that we know it's working, I can turn the y position transform effect back on, and these two effects stack on top of each other. Now, if we play it back, not only is it going to lift up off the ground, but it's going to start wiggling and then returned to its home base. So let me space this out a little bit so we can see it hovering for longer and see it raises up, wiggles up and down, and then comes back down to a resting point. And these are based on the frequency and amplitude. So let's copy these over to all of the other layers. So with those effects selected, I'll press Copy with Control C, removed the effects from all the other layers and I'll turn them on as well, everything but the tray that is press Control Shift E to remove those effects and then paste one more time. Now the reason I didn't remove the effects from the tray is because remember, we wrote a different expression on that wide position than the rest of the layers so that it's a direct tie into that wide position slider on the global controller. So if I replace that, then the tree is going to be moving differently than I want it to. But let's play this back and see what happens. Everything lifts up off the tray. It's wiggling up and down on the y-axis, and then it falls back down into place. So that's great. Everything is moving more than I want it to, but this is very easy to dial in now because I have the position, frequency, and amplitude that I can just scale back if I want. So let's say I don't want it to happen one times per second. I want it to happen 0.5 times per second, or once every 2 s. And let's also say the amplitude was too much, I could drop that down to 80. Now everything is going to dynamically wiggled differently based on those values that I put. So it's much slower now AT is probably still too much. I'll drop that down to 50. Let's go back to the train now because I do want that y position wiggle on the tray. I just don't want to overwrite the standard y position, so I need to be careful with my copying. I'll bring up the effects of any of the other layers. Grab that, why wiggle? And then show you what happens if I paste on top of what was already there. It just replaces the transform that was already there. So instead of that, I'll undo duplicate y position and with that second copy selected, I will paste. And that replaces that instance of the transform effect. Now I should get the same wiggle on the tray as everything else. There we have it. So that's looking great, but it's still pretty rigid and static because it's only wiggling on the y-axis. Let's also introduce a wiggle to the rotation, which we again will do using another instance of the transform effect. So this time let's start at the bottom and I'll just duplicate the y wiggle on the tray. So I'll zoom in here. Duplicate why wiggle, and then just turn off the y position and the wild wiggle so that we can focus on just the effect we're working with. This is gonna be the rotation wiggle. So I'll rename that effect and then open up the position. We can actually get rid of this expression. So I'll just hold Alt Shift or Option Shift E on a Mac to remove that expression. In fact, I'm actually going to reset this entire effect so that we have default values for everything. And I want to point out something. If I change the rotation, It's not going to rotate around the center of the artwork. And this goes back to what I was talking about with vector layers and After Effects, after Effects is looking at that layer as if it's the size of the entire comp. And so is this effect, which is why the anchor point is right in the center of my comp, and that's where it's going to rotate around. I want that anchor point to always be right in the center of my art work. But if I move the anchor point, the contents of the layer going to shift around with it. And that's because the position also needs to be adjusted. There's two steps to fixing this. First of all, the position should always follow the anchor point. So I'm gonna go to the position property, use my property, pick whip to grab the anchor point, and that automatically adds in the expression to follow the anchor point. Now I can move the anchor point around wherever I want and it's not going to affect the position anymore, at least visually, the actual position property is changing. But now I can rotate this around whatever point I put the anchor point at, but I want it to be right in the center of my layer. Well, if you've noticed that there actually is an anchor point on the layer and that corresponds to the actual transform controls of the layer. If I go into the transform, There's my anchor point. That's that crosshair right in the middle. And that is exactly what I want to target with my rotation wiggles anchor point. So I'll alter option, click on that anchor point and then type out a very simple expression, transform dot position. And finished that with a semicolon. So we're just looking at the transform controls of the layer that this belongs to and the position property for that layer. Apply it. And now wherever this layer goes, the anchor point is always going to be smack dab in the center. Which means when I go to rotate this, it's going to rotate around the center of the layer, just like I want it to. So now that we have that done in all set up, I can actually go to the rotation and implement a wiggle on it. And we're going to just borrow the same expression that we used on the Y position wiggle. So I'm gonna go into that and copy, then collapse that backup, go into the rotation, add an expression to it and paste. Now if I apply this, we're gonna get an error because this property is not an array, it's a single value. We need to change things up a bit in here. First of all, we don't need an x-variable, so I'm going to just select that by triple clicking it and then pressing the Delete key, and then I can get rid of my array. I don't need this first half with the x. I don't need the square brackets. We're just going to leave it to new. Why? I also don't need to specify the index of the array since there isn't one in my wiggle expression. So I'll get rid of those square brackets and the number one. Now if I apply this, our error goes away and this tray is going to start wiggling around. That's great, but it's referencing the same frequency and amplitude sliders that the y position is. And I want unique frequency and amplitude sliders for the rotation. So I'm gonna select both of those, duplicate them. Drag them down and rename them. So we're going to say rotation frequency, rotation amplitude, and then link those up in the variables. So I'll go into here and instead of this line of code for the frequency, we are going to change this to rotation frequency. Same thing for the amplitude. Select that all the way up to the semicolon and then grab the amplitude. Now I can change the rotation wiggle separately from the position wiggle. And I know for a fact, 50 degrees of wiggle is way too much. I'm just going to dial that back down to say 12th. And then I'll turn the other effects back on and just bring up the keyframes so we can see how they align. The tray lifts off the ground, hovers up and down and wiggles in a much more manageable rate. I think that's looking great. So let's grab that one effect, the rotation wiggle from the tray. Un-solo layers go into role too, so we can see the effects. I'll press E and paste. And I did it that way just to make sure that it wasn't going to replace any of the other effects and it didn't. So now I know it's safe to grab all the other layers and paste that as well. Again, just pressing E to double-check that none of the effects actually were replaced. And now I'll play that back and see what it looks like. Okay, So everything lifts off the ground. It hovers up and down and it's rotating, but it doesn't seem to be rotating around the center of each layer the way that we had it set up. So let's take a look at why that is. Let's just grab this sauce mole and solo that for a second. So what's happening is it's lifting up off the ground, but then it's rotating and it looks like it's rotating from somewhere down here. So if I press E to bring up the effects, the reason this is happening is because effects are applied in the order of top-down. So first we're moving the Y position, then we're wiggling the y position, and then we're rotating the layer. And that instance of the transform effect is rotating around this point right here where the original artwork was. So instead of doing that, I first need to wiggle the rotation, meaning I need to move it above the y position effect. Then I'll move it up and then we'll add that wiggle to the position. With that order of effects. I'll copy them and then just to be safe, I'll remove all the effects from everything but the tray Control or Command Shift E, and then paste them again. And then I'll go into the tray as well and move that rotation wiggle up above the other two effects. Now I should be able to play this back and see the type of motion that I actually wanted. Everything lifts off the tray. It's bobbing up and down in all the rotation is happening around the center point for each layer. That's exactly what I wanted. Now the position amplitude is probably a little too strong, so I'm going to drop that down to 25. And the rotation amplitude could probably also be dropped down to maybe eight degrees instead of 12. This is completely up to your own taste. And the great part about building all these controls is that you have the flexibility of just modifying some of these values and seeing how it affects the overall animation. Now, this piece of rice right here again is giving me trouble. So maybe instead of dropping that down, I'll grab the tuna and I'll just bring that higher up in the stack. But overall, this is very close to what I had in my mind for what I wanted this animation to look like. 11. Property Links: Now I've generated all of the motion that I wanted to out of this animation. And I've done it all with a single set of keyframes, but there's even more detail that I could add to this that isn't strictly for the motion of all of these layers. Let's say that you wanted to add some stylized looks to your art work using effects, but be able to control them with the same amount of precision from this one point. Well, we can do exactly that. So let's add a boil effect to all the artwork to give it a texture dancing edge. I'm going to search for the turbulent displace effect and apply that to the tray. If you've never used turbulent displace, this is just going to add random distortion to the entire image and we have a size and an amount to mess with how that distortion actually looks. So I want to turn the size all the way down to its minimum of two, which is going to give this nice textured edge. And the amount, I'll probably have a rate around 50. I want to bring that all the way up to the top of the effects stack. And I'll collapse these other effects just so they're out of the way. But that way the texture is applied before the motion and it will stick with that layer. Now. Now there are a few different ways that you can animate turbulence. One of which is evolution, but this is kind of a more fluid way of animating things. And I want it to be pretty randomized. So I'm gonna go down into the evolution options and there is a random seed value, just like with the seed random expression. Random seed is going to give the displacement a random starting point. So if I change this to any other number, it changes pretty drastically how that distortion looks. So we can add a random expression to this in order to animate it. So let's just press Alt or Option and click on that stopwatch to add an expression. In that expression we'll just type random. And in the parentheses, I'm just going to put a really big number, so I'll say 10,000. And that way on every frame it will generate a random number 0-10 thousand, completely changing the way that that turbulent displace it looks. So I'll play this back and you'll see that now that texture is animating. It's dancing around, but it's changing on every single frame. I would rather not have it changed every frame. So let's go to our global controls. Duplicate the last slider and name this boil FPS for frames per second. Now, I want to make sure that I can always see these effects. So I'm going to click on this little lock button right here in the effects controls panel. That way it'll be visible even if I select a different layer. Okay, so let's go back down to the layer we were working on the tray into that expression. And before that random 10,000, I'm going to drop down a line and add in a posterize time. And in-between the parentheses, I'm just going to select that slider. That way whatever number I type in here is what the frames per second for that expression, that random value will be. So if I change it down to six, it's going to update at a rate of six frames per second. And that's a lot more like what I actually wanted. I might bump it up to 12 frames per second. That's half my comps frame rate. But the point is, I have very flexible control over it now. But what if I wanted to change the amount and the size as well? Well, I could make more sliders and I could tie these two it, or I could do that copy with property links trick. So why don't I do that just to change it up a little bit. I'm actually going to cut this turbulent displace from the tray and paste it onto the global controls and will collapse all of these other effects. For now, I'm going to rename this turbulent displace boil one. Remember this doesn't do anything because it's applied to the null object and not any of these artwork layers. But if I copy it with property links by going up to edit, copy with property links and then pasted across all of these other layers. Now I have this one point of reference and can adjust things. So let's say I want the size to be bigger. I could increase the size here on the global controller and it will affect every single instance of that Turbulent Displace. Now I liked to, so I'm going to leave it as is, but maybe I turned the displacement amount down a bit. Okay, That's looking pretty good, but I want to double up this boils that we have two levels of it. So I'm going to solo these last few layers again, go into my global controls and duplicate boil one. This will be boiled two, and I'm going to copy and paste both of these with property links now. So edit, copy with property links, go down to the tray. I'll just delete the boil effect there and then paste it. And remember, I want these boils to happen before the other Motion Effects. Now that I have two instances of turbulent displace being driven by this global controller. I can dial in the second look. So I want the size of this one to be much bigger and the amount to be much smaller. If I go into the evolution options, the random seed is still being animated by my boil FPS slider. So that's still applies. It will update at a rate of 12 frames per second. So let's play this back and see what it looks like. Now we're getting a larger distortion over the entire layer. So let me turn off boil one just so we can get a clearer picture of what this second distortion looks like. So it's subtle, but it just adds a little bit of personality overall, especially when combined with the first instance of boil. Alright, let me un-solo those layers and since they're all linked up with property links, I can copy and then paste it onto another layer. Press E to bring up those effects and make sure that they're above the motion transform effects. I still want the Y offset at the top, but now that that effect stack is set, I can copy it. I can go into my other layers, select them all and remove the effects and paste one more time. And now everything has that double boil applied. I can unlock this so that we can see more than just this layers effects. If I zoom in, it's very easy now to dial this in to look however I want. So if I want more of an extreme distortion on that first instance of the boil, I can dial that up or I can turn it back to be a little bit more subtle. And with that boil FPS, I could change this to any number of frames per second that I want to completely customize the look. Just like that I've added in some animated texture that is still completely controllable from this one, global control null object. One thing I do want to point out though, if I go back up to Edit, there are two options here, copy with property links and copy with relative property links. So let me go into any one of these Boyle effects and point out how any of the expressions on that effect is written. It starts with comp, sushi, which is the name of this comp. So it's looking specifically for this comp. And then the layer named global controls and then the effect boil one and then the property amount. What this means is that if I duplicated this comp and I now have sushi too, and I go into that comp and adjust the global controls for one of the boils. So let's say I turn the size way up. That is not going to affect anything in this comp because these layers expressions are looking for it. This comps global controls null. And the properties that we have set here, if you wanted to make something that was a little bit more universal. So it wasn't controlled by one specific comp. That's when you would use copy with relative property links. So if I grab boil wanting boil to go to edit, copy was relative property links. Select boil 1.2 here and paste. Now all of those expressions are going to be written a little differently. Instead of saying sushi, It's saying this comp, meaning whatever comp, this expression is written in. Now if I copy all these effects to all the other ones, I'll just remove all the effects and paste them again, and then just copy that boil over to the tree layer as well. Now all of my expressions are relative to the comp that they belong to. I can delete this sushi to duplicate sushi one more time. And now all of these global controls will affect what's in this comp. So that's just something to keep in mind when you're working with copy with property links. 12. Layer Styles: I've now pretty much recreated my animation for this entire class project. But there are some more things that I wanted to show you that you can take advantage of, especially with copy, with property links. Like I said, a null object is just like any other layer. It just doesn't render, but you can actually apply more than effects to it. You can even apply layer styles. So if I again go down to the tray and I right-click and go to the Layer Styles. Let's say I want to add a white stroke around the entire thing. So I'll add that stroke, change its color to white and then maybe make it larger. Now it's kinda going to look like it's almost a sticker without white stroke outline. I want to copy that stroke layer style over to the null object. So Control C, Control V. Again, nothing visual changes, but now that I have it here, I can select it. Go to edit, copy with a relative property links this time, and then paste it on all of the other artwork including the tray. So I'll just select everything and paste and everything gets that layer style. It all looks like stickers now. And I can affect the color and the size right from the global controls null. So maybe that size is a little too big. I'll drop it down to say seven. Everything gets a little bit thinner. Maybe I don't want it to be pure white. Maybe I want to drop it off of pure white just a little bit. So it's not as intense. I actually don't think that looks as good, so I'll undo that. But the point is you can control a lot more than just affects with copy, with property links. Again, freeing up the need to have to dive into all of these individual artwork layers just to change something visual about all of it. It's all being driven by this global controls. No, freeing you up from a lot of tedious work. 13. Thanks!: That's it. Congratulations on finishing this class. I hope that expressions are no longer as intimidating to you as they were before you started and that you really feel like you have a good understanding of how to move forward and using them in your own projects. If you liked this class, I would love it if you gave me a review, let me know what you liked about the class, what you didn't. So I can always improve it. The next one, I always want to improve my classes and the way that I teach it. So don't hold back. If you use anything that you learned in this class, be sure to tag me on Instagram or Twitter at Jake in motion so I can see your work and show it off. Also check out my YouTube channel, youtube.com slash Jake and motion of for regular content that's being posted there all the time. And definitely be sure to follow me here on Skillshare. So you're always up-to-date with my latest classes and you can start watching them as soon as they're posted. Thank you so much for taking this class and I'll see you in the next one.