Blender 5 Basics: Part 3 - Camera & Daylight

1. Blender Camera & Daylight: Welcome back to this series on Blender Basics, the bouncing ball. My name is Aaron F Ross, and this is P three. It's all about camera and daylighting. We'll set up a camera in order to frame the shot, and we'll also apply some basic lighting using a built in daylight system called the sky texture. So let's get started with camera and daylighting for the bouncing ball. 2. Using the exercise files: Let's take a look at the exercise files that I provided with this series of courses. You should have already downloaded the exercise files and extracted them in the first course in this series. Let's take a look once again at what's in the exercise files. This is a project folder, and primarily we're concerned with the three D scenes. Go in there, and we'll see a series of scene files. Generally, there's a scene file for every movie, but in some cases, there isn't. Also, these scene files generally flow from one to the next. So a scene file, such as oh 302 oh two camera dot len is the begin state for that movie, but it is also the end state for the previous movie. And once again, to refresh your memory, I provided all the scene files for the entire course series of six courses. They're all in this single package, and the scene files are numbered with first the course number, then the chapter number, and then finally the movie number within that chapter. Going back up. We also have input textures. These are required in order to open the scenes. You don't want to do anything like change the input textures folder name. Otherwise, Blender won't be able to find the textures it needs to load the scenes. But you can change the root folder name or move it to a different drive or whatever, and Blender should be able to find the input textures. 3. Configuring the Quad View: Up to this point, in our course sequence, we've only needed one viewport at a time, either the perspective view or the top view or whatever. We're setting up a camera, we probably want to see our scene from multiple perspectives or vantage points at the same time to get a really clear sense of exactly where things are in our world. For that, we need a Quad View. We can invoke that with a keyboard shortcut, which is Control Alt. And on the Mc, that's going to be Control Option C. Now we have a Quad View. We've got a single perspective view, top view, front view, and right view. This is going to make it a lot easier for us to understand where things are in relation to one another. We also want to be in wireframe view to see things most clearly, so we're not distracted by textures and backgrounds. We know, of course, we can go into wireframe view from the viewport header. You'll want to memorize the keyboard shortcut for toggle wireframe, as well, and that is Shift Z. We'll toggle wireframe on and off. I do want wireframe to be on right now. We can hide our toolbar over here. We don't need that right now. The keyboard shortcut for that is we also want to frame all the objects in all the views. And we previously saw that we can use a keyboard shortcut to frame an individual object or selected objects in a particular view. For example, if I select this ball, I can press the dot key on the number pad, and that's going to frame that object in that view. We could hover the mouse over a different view and press that dot key and frame the selected object. If you don't have a number pad, you can use the menus to do this. And you will want to go up to the View menu and you can choose frames selected. That's going to work in the current viewports. You need to actually activate that view and then go to view frame selected. If you want to frame everything in all views, there are a couple of keyboard shortcuts you'll need to learn for that. First, to select everything is the A key on your keyboard. A is just select all, and we can see that in our outliner. There's a menu item to frame in all views. We've got frame selected Quad View. Or we could use the keyboard shortcut, which is listed here, control number pad dot. So going back to our viewport, pull down control, press the number pad dot key, and now all four views are framed to enclose all of the geometry in the scene. And one last thing I recommend is changing the transparency of the viewport header. It is slightly transparent, and we can see that if we go back to a solid view or material preview with Shift C we can see there's transparency up here. Well, that actually changes the composition of this window, and it actually makes it harder to read menus, for example, I can navigate with Control Middle Mouse. And we can see that header transparency is actually causing the menus to fight with the scene geometry. I want to disable that transparency. Go into the edit menu to preferences. Go to the themes tab, open up the three D viewport section, scroll down to the bottom of that, open up the theme space, and we have the header on the left is the color. On the right is the transparency or the Alpha channel. Click on that, set the Alpha to a value of one. And now we have an opaque header. That's not causing visual confusion. And notice, if the Alpha is not set to one, it does shift all the geometry in the view here. So that's just another reason why we want to set that Alpha to a value of one. Okay, and I'll go back to Wireframe Shift Z, and we're ready now to create a camera in the next movie. 4. Creating a Camera: The default Blender scene has a camera in it already. But at the beginning of this course sequence, I erased everything that was in the default Blender scene. And so this scene doesn't have a camera in it. Let's add it. Go into the viewport menu and choose Add camera. And that camera is created at the origin. If we go to the object properties, we can see its location is 000. It's got some non zero rotations, though. Let's investigate. We can zoom in on that camera using control and number pad dot. It's oriented so that it's kind of tilted down. And we can get in closer here in this right view with Control Middle mouse. I want this to start out being level, not tilted, and also not panned in any way. So I'm going to set the rotation X to 90 degrees and then rotation Y and Z, drag my mouse across those, set those to zero degrees. And now that camera is kind of neutralized. Now, if we back out again, control middle mouse, it is very, very small. We don't want to scale it. We want to change its viewport properties. And to do that, we'll need to go into the camera object data properties. Our data properties are found in this tab that has a green icon that icon changes depending upon what's currently selected. Right now, it has the look of an old school film camera. Select that. Go into viewport display, and increase the size, set that up to maybe 1 meter. Okay, so we created our camera and we can move it around like any other object. We can choose the move tool, which is G and position that camera, and we can look in all the various views, get oriented. Maybe orbit in that perspective. So we can now sort of get our camera set up in an initial starting position. 5. Working in a Camera View: We've created a camera and positioned it roughly where we want it to be. Now we need to look through that camera and see what it sees. To do that, first let's go back to our material preview mode, pull down shift and press the Z key. That'll toggle off our wireframe mode. Then we want to go to our perspective view, and we have on the right, an icon that looks like a camera. Click on that. Now we're looking through that camera's view. And depending upon which kind of navigation we do, we're either going to be manipulating this viewport in a two D mode, or we're going to be dropped back into an ordinary perspective view. And this is a little bit strange behavior of Blender, but that's what happens. So depending upon which navigation you try to do, you'll get different results. First, let's talk about Zoom or dolly. And that's the control middle mouse keyboard shortcut. We're just doing a two D Zoom. Notice that our camera is not moving. Okay, so we can set that so that the camera frame line as indicated by this rectangle is taking up most of that viewport. We can also position the camera frame within the viewport using shift and middle mouse. So we can position that maybe so that none of the viewport elements are crashing into our frame I'm going to just set that to be in the center. So if we move using shift middle mouse or if we zoom using Control middle mouse, we're operating on the viewport and not on the camera. But if we just orbit using the middle mouse, suddenly we're dropped back into a perspective view, and we're not seeing the camera view anymore. And this is a bit strange behavior, but you'll hopefully get used to it. We can click on the camera icon, and now we're back to our camera view. If we enable this little padlock, now, when we do anything to navigate in this camera view, we'll actually be changing the camera's transforms, not just changing our view. Okay? So if I do shift middle mouse, I'm changing that camera's position, left and right and up and down. We can see that in the other viewports. If I do control middle mouse, I'm dollying that camera forward and back. If I do an orbit with just middle mouse, I'll be orbiting around the center of the world or around the selected object. If I select my ball and middle mouse, I can orbit my camera around that selected ball. Okay, so I can just frame this up. I can control middle mouse and dolly back a little bit. So now I'm taking in more of my scene. Another thing you need to know about camera and perspective views in Blender is that in this standard three D viewport, you're not allowed to see through the camera lens and also see a third person perspective, like a director's point of view, at the same time. You can either look through the camera or your perspective can be an objective director's point of view. And you actually cannot do both of those things at the same time. You would think you'd be able to load a perspective view into one of these ortho viewports, but you actually cannot do that. If you need to see through the camera lens and have a third person perspective, you'll need to actually open up another window. So I can open up a new window, and I can make that my camera view, and I can switch that to material preview mode. I can zoom in two D with Control Middle Mouse. Once I have that established, I can padlock that and I can change whatever other options I want. Like, for example, in a camera view, I might want to turn off all overlays and all Gizmos. Turn off all that stuff. So now what I'm seeing in this floating perspective window is just the camera view and nothing else. So it's really, really clear now what I'm dealing with. So I can go back to my main viewport and I can turn off Quad View with Control Alt or control option C, and I can turn off the padlock and middle mouse drag and I'm now orbiting. Now I've got the situation where I can see a third person perspective of my entire layout and also see through the camera lens at the same time. Just to make sure that my lighting is exactly the same, I'm going to go into this three D viewport, material preview options and turn on scene lights and Scene world. Now I've got a really clear what you see is what you get representation of what I would get when I rendered this shot. 6. Adjusting Camera Data Properties: Let's take a deeper look at the camera properties. First, I just want to hide the header in this floating three D viewport. I'm not going to be using that. It's just cluttering up my view. I can right click on an empty spot and choose header, Show Header. I can always get that back if I need it by clicking on this tiny little down facing arrow in the upper right corner. That'll restore the header. If I want to hide it, I can right click and choose header and disable Show Header. Okay, so now let's talk about the camera properties. We previously saw that we could change the size of the camera, and we do that by selecting the camera and going into the camera's object data properties. And in the viewport display section, we have the size of the camera. That's just how big it renders in the viewport. It doesn't actually affect anything except the visual appearance of the camera here in the view. And by the way, this big opaque arrow pointing up indicates the top of the camera frame. Also in viewport display, we see pass part two. That is a French word. That means a frame that has a mat. Okay? So this is the transparency of the area outside of the camera frame. This rectangle is the camera frame or what's inside the renderable area of the camera, and anything outside of that doesn't get rendered. We do want to see slightly beyond the edges of the frame so we can have a clear sense of exactly what's in the frame and what's not in the frame. But we don't really need to see this very well. We can dim that way down by increasing the opacity of the passpart. So if I drag that upward to give that a larger number, we see that we have a darker area here outside the frame. I'm going to turn that up to a value of 0.9. The main thing that we do want to concern ourselves with in the camera properties is the focal length or the angle of view or field of view. And that's right up at the top. We see focal length. And that's how far the camera lens is zoomed in or out. And it's set up to use millimeters. On a real camera, there's an effective focal length, which is the distance from the center of the lens to the camera sensor. This focal length is not usually the literal distance from the lens to the sensor. For a composite lens, that focal length is a value that will give a crop factor if it were a simple lens. Okay, so maybe that's a little bit too technical because, in fact, in CGI, we don't often care about this, at least for straight up animation. If we're trying to match to some existing footage that was shot with a real camera, then this would all be very important. And the focal length and millimeters would interact with the camera sensor size down here in order to produce a particular crop factor or a Zoom factor. So we could set this up so that it corresponded to some actual camera measurement, the sensor size of a real camera. By default, it's set up to act like a DSLR or a digital single lens reflex still camera, not a motion picture camera. Okay, so for the purposes of straight up animation and not trying to match to any kind of existing background image or plate or something shot with a real camera, then we don't actually really care about millimeters, and we don't actually care about focal length, real or virtual. We want to set the lens unit to field of view. Now this field of view value reads out in degrees, and that is the degree value of the horizontal arc of the camera. The camera is represented in the viewport by a pyramid that indicates the area that is visible to the camera. That's called the frustum. And the angular value from the left side to the right side is this field of view metric in degrees. If we increase the field of view, drag that to the right. We'll be zooming out, and the shape of the camera frustum changes accordingly. If we have a really extreme field of view, then we get some serious distortion in our final camera output. And by the way, this is an issue because the way a CGI camera works is it's a flat projection, but the way a real camera works is it's a spheroid projection or a barrel distortion lens. And in a real lens, when we have a wide angle shot, the objects in the center become larger relative to objects on the edges. But for a CGI camera that doesn't have barrel distortion, it does the exact opposite. So that just complicates things if you're trying to match to an existing camera footage. Okay, so for our purposes, though, we only really care about this field of view, and we can zoom in or zoom out. A naturalistic value for this might be something like 45 degrees, and that might correspond to more or less the amount of perspective distortion that we would see with our naked eye. Let's set this to whatever looks good in your particular shot. This Zoom factor here is going to interact with the dolly or the distance from the subject. So we would need to also do a control middle mouse here in the view in order to position our camera to take in our entire layout according to whatever field of view. So those two are going to interact with one another. A 7. The aesthetics of framing: To conclude our brief discussion of cameras, I want to talk a little bit about the theory of the aesthetics of cameras. Why would we choose a particular camera framing over another? That's going to be determined by our artistic intentions. What are we trying to convey with this shot? And it fundamentally breaks down into whether we want the shot to be an objective framing or a subjective framing. An objective shot is one that's just trying to tell the story of whatever's in front of the camera, but without any particular bias, without any particular point of view, whereas a subjective framing is taking on some point of view or some vantage point. And a subjective framing could be from the point of view of a particular character within the narrative or not. It could just be a shot that draws attention to itself as being from some unusual perspective. Maybe it's from an extremely low angle that a human viewer would not generally be able to assume. So let's do that. We'll do a subjective shot from a low angle. I'll select the ball. And then in the floating three D viewport, middle mouse drag, and I'm orbiting around that ball, and I can bring that camera down really low so that it's pointed upward. And that's a low angle shot. This is a subjective point of view. It's not necessarily from the point of view of any one or anything, but it's a subjective framing. Likewise, if we orbited upward and looked from an extreme high angle, this also has a very subjective character to it. It's really giving the impression that we're looking through the eyes of some thing. Maybe it's a bird. So those are subjective framings. Another example would be if we did a really extreme close up, I'll middle mouse drag and go back to the framing I had before. With that ball still selected, I can zoom in on that with the number pad dot key. And now I've got a close up on that ball. And again, this is a kind of subjective shot because especially for a bouncing ball, this is not an angle that we would normally be able to see. So it's kind of the magic of cinema. If we're doing a shot like this, then we also want to think about focus. And I mean focus both figuratively and literally. We want to draw attention to a particular subject, so we're focusing our attention on that object. But we also want to control the blurriness of the shot. We want our subject to be in sharp focus and things that we're not drawing attention to, like the background, be in soft focus or even totally blurry. And we can do that through the camera properties. I'll select the camera in the outliner view. Go back to those camera data properties. And there's a section labeled depth of field. Open that up. In cinema terms or photography terms, the so called field is a range of distances away from the camera at which a subject will be in focus. And that field itself has fuzzy boundaries. But basically, if we have a deep field, that means things at a wide range of distances are going to be in focus. They're going to be sharp. If it's a shallowed field, that means only a narrow range of distances are going to be in focus. It's a naval depth of field. And we can set our focus distance numerically and bring the ball into focus. Right now, the focus distance is at 10 meters away, so everything's blurry. We can set that value numerically, but we can also just choose to focus on a particular object. And to do that, we can go to this field here that's labeled focus on object. Click on the eyedropper and then click on some object either in the viewport or in the outliner. The outliner is a bit more predictable. I'll click on ball. And now the ball is the focus object. We see the ball is in focus and the fence in the background is out of focus. We can make that more extreme by changing up these aperture values down here. For example, I can bring the F stop down to 1.4, which is doing a really extreme amount of blur. Okay, so that's another subjective shot. I'm going to turn depth of field back off. I'm going to dolly back out, control middle mouse. And frame this up as a more or less objective shot, making sure that I'm not seeing any of the background here. So I want to zoom in or dolly in, and I want to frame this shot more or less objectively. I'm not trying to give the impression of looking from some special point of view. I also probably want to preview the motion of my animation. I can select the ball and go to the move tool with a G key and kind of preview that motion in my floating three D viewport and figure out, Okay, is my framing right? If the ball is at this distance away, it looks like in this case, it's going to be a little bit too close to the bottom of the frame. So I can either move the ball closer to the background object. Or I can change my framing. Let's say I do want the ball to be more or less centered over the ground plane. That would mean I need to change my framing here. So, again, I can middle mouse drag, and I can shift middle mouse, change the position. So this is a sort of quasi subjective view. It's kind of a low angle, but it's not really super drawing attention to itself. So that feels about right. And if I really want to be precise about this, I can select the camera and just make sure that it's not panned left to right, that it's looking at the shot straight on, go into the object properties. I've got rotation X, set that to a value of 90 degrees and that'll level the camera out. Okay, so I don't want that. I'm going to undo that control Z. I can set Y to a value of zero, and it didn't make a lot of difference there, but it rotated the camera, so it's more square to the world. And the same with the Z axis. I'm going to set that to zero also. Okay, now I've framed up my shot, and we're ready to proceed with lighting and rendering. 8. Interactive production rendering with Cycles: We're now ready to add natural daylight to our scene. We could render that in EV, but it's not going to be a very high quality rendering. It's not going to have bouts light and so on. For that, we need Cycles, which is the production renderer. At the beginning of this course sequence, we set our preferences to use whatever graphics processing unit is installed, and now we're going to put that to work. I've got a floating three D viewport here, and I've currently hidden the header, but I need to bring that back. So I'll click on the little arrow here at the very top right, and I'll need to expand this window a bit to get at all of the icons, especially those on the far right. There's just not enough room. We can't hide any of this other stuff. We can minimize the menu, though. We can right click on an empty area and disable show menus. And now we've got a hamburger menu there. That helps a little bit. So we need to be able to get at these viewport shading modes on the far right. Currently I'm in viewport preview mode, and that uses EV, which is the viewport renderer. If we click on the far right button rendered viewport shading, currently, nothing will happen because the default production renderer in Blender is not Cycles, which is a bit unexpected because Cycles is a far superior renderer, and one would expect that that would be the production renderer, but it's not set up to be the production renderer. EV is set up to be the production renderer. So we need to change that. I'm going to switch this back to material preview. Go over to the property panel and go to the render properties. And at the very top, we see render engine EV, and a lot of the settings that we see down here are going to be specific to EV. If we want to change any of the EV settings, the current render engine must be EV. As soon as we switch this over to anything else, we cannot access those EV properties. In order to change the properties of a renderer, it needs to be the active render engine. I'm going to set this to Cycles, and we see the device is set to CPU by default. I'm going to set it to GPU compute. And as I mentioned, at the beginning of the core sequence, that's going to drastically accelerate the render process by a factor of ten. It's going to be ten times faster or even more. If you have any issues, if, for example, Blender crashes, then you'll want to go back in here and set the device back to CPU. That's always safer. Okay? The render engine is Cycles, and the device is the GPU. I can go back to my camera view and enable rendered viewport shading. Click on that, and it may take a moment, but now we're seeing a very high quality rendering. Now, it's not that great aesthetically right now because we haven't added any real lighting. We're just using the HDRI image that we used for material testing. So we will want to add a Sky Texture in a later movie. But now we have Cycles in effect, and we can turn that back off again by going to the material preview. 9. Assigning Hotkeys: I want to detour for a moment into customizing the Blender interface because the Blender UI is a really mixed bag. There's some great things about it, like the node editor or the shader editor. That's really well designed. But there's some really big problems with the Blender user interface, especially around viewports. As I mentioned, we can't have a camera view and a perspective view at the same time, so I had to open up this separate window, and that actually creates problems as well, because now I've got this clutter from this header up here. And if I hide the header by right clicking and choosing header, show header, now I can't get at the buttons in order to change the viewport shading mode, and I need to be able to do that, especially if, for example, it's a heavy render and I need to stop the process. I'm going to re enable that header by clicking on this down facing arrow. And I'm going to assign hot keys or keyboard shortcuts, and that's really easy. We can just right click on something and assign a keyboard shortcut. I can go up here and right click on any one of these shading modes and choose assign shortcut and then press a key. I've already researched this. The F five function key is not used for anything, so I'm going to press F five. And now I can get at that menu by pressing F five. Position my cursor anywhere in this view, press F five, and I can switch to one of the other modes such as solid or F five, again, go into material preview mode. If I press F five and go to rendered mode, I invoke Cycles, and we may want to pause the rendering without going into a different render mode. And there's going to be a pause button on the extreme far right. But again, if I have a small window, I can't get at that button. So I'm going to assign a keyboard shortcut to the pause button, as well. I'll right click on that pause button and choose assigned shortcut, and I'm going to use the pause key on my keyboard. Press pause. And if you're on the Mc, you may not have a pause key, use something else. But I've assigned the pause button to the pause key. So if I press the pause key on my keyboard, I'm not rendering anymore. Okay? Press the pause key again, and rendering is re enabled. Hover on my mouse over that view, press F five, and I can switch back to material preview. And having done all that, now I don't need the header. I can hide that permanently and still get at these super important controls. I'll right click and disable Show Header. And that's how to assign keyboard shortcuts or hot keys. 10. Wrangling Workspaces and Areas: I want to talk a little bit more about customizing the Blender interface because especially if you're going to use the program a lot, you really need to optimize that interface because it's not set up like it would be for a normal three D program. Again, because Blender developed in total isolation, it's got a lot of really weird quirks around it. I had to create this floating three D viewport in order to have the ability to see through the camera lens and see a perspective view at the same time. That is a solution, and it would work fine if I had a different monitor. I could put this three D viewport on that other monitor, and that would be fine. But if I have a single monitor, it's kind of a problem because it's covering other stuff up. Now, luckily, it does float above the main interface. But that also means that wherever it is, it's going to be obscuring something, even the main menus. If I go into the menu to File Save, for example, that menu is hiding behind this three D viewport. So I'd have to move the viewport, and it gets very annoying very quickly. What we want is just a conventional setup where we have a four viewport layout that allows us to load whatever we want into each one of those panels. Blender doesn't want us to do that. It's not set up for that. We can do it, but we have to jump through a lot of hoops just to get to kind of a standardized interface. We actually have to rejigger all of the Blender interface in order to optimize it for usability. So I'm going to get rid of this three D viewport, close it. And by the way, when you close a window in Blender, it's a permanent operation, and you cannot get back to those settings unless you went through and did them all again manually. Okay, now I've got a four viewport layout, the Quad View. But again, I can't have a perspective view and a camera view at the same time. I also can't have different shading modes in different quadrants. If we choose, for example, the material preview mode, it's going to apply to all the viewports. If we choose the production rendered mode, that again is going to apply to all of those quadrants, and that is really never, ever what you want. That's going to put a huge load on your computer, even with the GPU. So that's a terrible idea, as well. So what we need to do is actually build our own interface in Blender. So I'm going to start by going back to the non Quad View control Alt Q. And I'm going to now create a new workspace. I'm in the layout workspace. I'm going to make a new one. Click on the plus sign on the far right of the workspace tabs and choose duplicate current. Now I've got layout OA one. I'll double click on that and rename it. I'm going to call this 14 views. Now I'm going to modify this panel layout. And that's done by hovering the mouse near the corner of one of the areas or panels. When you bring your mouse close to that, it turns into a plus sign. And this can actually bite you. This is a really common thing with new users is they accidentally click and drag on this and they don't understand what they did. They actually create another area. But that's by design. If you click and drag on that little corner, and you drag on top of something else, you have the option to replace that other thing or move this area to that other thing or split the other area. Okay. So now I've got an up and down layout, top and bottom. I can divide this one as well and divide this other one. Now I've got four areas, and I can load different things into each one of those. I'll leave this user perspective as it is, go to this user perspective and load the camera in there. And then over here, I'm going to go to my top view. Click on Z, and I want to go over to the negative Y view in this fourth quadrant, which is going to be the front view. Click on negative Y. Now I've got what I want. I've got those four viewports, and I have independent control of each one of those. I can set their shading mode, or I can set whatever display properties I want for each one of those. I'm going to first minimize the menu so I can get at more of those icons. Right click on the header and disable show menus. I can get at those menus from the Hamburger menu here. I want to go into the Gizmos and disable Navigate. Because I'm going to use the keyboard shortcuts. I'm also going to go into the header and just hide it entirely, right click Show Header. Now I've got my perspective view. I can navigate with the middle mouse button or shift middle mouse or Control middle mouse, and that's all good. Same thing with this top orthographic view. I'm going to right click, go to header and disable Show menus, go into the Gizmos and disable the navigation Gizmos and then right click and disable the header. Same thing over here, I'm just going to repeat that process. Right click, hide the menus, go into the Gizmos and hide the navigation Gizmos. And again, finally, right click and hide the header. I previously set up a keyboard shortcut for the shading modes, which is F five. So I can press F five here and go into solid mode, for example. You go over here and press F five, go into solid mode there. Or I could do ShiftZ which will go into wireframe and I can do that over here too, Shift Z. Now, actually, I can go back up to the header for this camera view and once again, hide the menus. And on the header, I'm going to hide all the gizmos and all of the overlays. So now this is just a camera view and then finally right click and hide the header. So this is now a very conventional layout, and there's no weirdness. I have the ability now to choose shading modes for each particular viewport. So I can hover my mouse over this camera view, press F five, and go into rendered mode in that camera view. And I've got a Cycles render here, and I can go into these other views and navigate around and not do anything weird with my Cycles view. I could, for example, select the ball and then press the G key and move that around, and we can see that it's updating in our camera view for Cycles. Okay, so I've got a new layout. And if I want to go back to a single view, I can go back to the other layout. And this is the traditional Blender layout. If I go back to my four views, I can illustrate that I can actually maximize a particular view. There is a keyboard shortcut for that, which is control and space bar. But that's going to maximize that view and take over all of the interface. So I lose my properties panel, I lose my outliner, et cetera. Now, we can work around that if we want to. We can load up a new window and put our outliner and or properties panel in that floating window and that way preserve that. So there's always a workaround. If you really want to go that deeply into customizing the UI, I will leave that as an exercise for you. I can go back over here and go back to a material preview mode, F five, and go back to material preview. Now this layout is going to get saved in my scene, and that's a thing in Blender. Blender actually stores the entire user interface inside each scene file. And again, this is extremely unorthodox. No other program has this. Blender stores the user interface in the scene file. So if I want to use this four viewport layout in the future, I will need to use this scene. Okay. Well, right, well, that's kind of weird. What you would do is you would delete everything out of your scene. You just have a blank scene, and at that point, you could say file defaults, save startup file. And that would finally save this new custom layout for all future scenes. 11. Optimizing Cycles in the viewport: We've assigned the Cycles renderer as our current render engine over here in the render properties. Let's invoke that in this perspective you where I've zoomed in on the ball so we can analyze the quality in the viewport. And I set my keyboard shortcut F five to enable the viewport shading menu. I'll choose rendered. And that's going to take a sec. When it's finished rendering, it'll say rendering done up here in the upper left. And it looks kind of grainy. The Cycles render engine is of a type called a Monte Carlo renderer, and it fires out rays from the camera in a random pattern. Additionally, it's testing each pixel multiple times in order to determine the color of that pixel. And because of the random firing of those rays, different pixels get tested different numbers of times, and so we get this grain effect. We can control that in the sampling sections for Cycles. There's a viewport sampling section and a render sampling section, and render is your final output. The viewport is the interactive production renderer, which is what we're seeing currently. We have MAX samples, Min samples and the noise threshold amount. Max and Min samples are how many samples are going to be taken on each pixel, and the noise threshold determines when the sampling will stop. If we turn noise threshold off and navigate in the view just to make sure that we've updated what we'll see here is our sample count goes up and up and up, and it's still rendering, even though it looks finished, all the grain is gone. It was forced to go all the way up to 1024 samples because we disabled the noise threshold. If we set samples to one and then navigate on our view with middle mouse, we'll see it super grainy. This is only one sample per pixel. Okay? So we want to find just the right number of samples, and that's what noise threshold is supposed to do. I'll re enable noise threshold, and I'll set the max samples back to 1024. And we can adjust the noise threshold to control the number of samples and therefore control the amount of graininess. If we set the noise threshold to its maximum of one, then we're not going to get the 1024 samples ever. We've got a really high noise threshold. We're allowing a lot of noise in our final rendering. If we bring the noise threshold down to, for example, 0.01, now it's going to calculate a lot more samples, and in fact, it went all the way up to 1024. So the noise threshold of 0.01 was essentially the same as setting it to zero, meaning there is no threshold and go ahead and calculate all of the samples. Well, we're going to trick this out. We're going to set this to optimize it so that we have very few samples for extremely fast interaction, but it'll still look pretty good in our viewport. Right now, if we navigate, it's going to take a long time to update. Now, by current standards, this is a long time. By historic standards, this is incredibly super fast. Okay, so I'm going to set my MAX samples down. Set it to the minimum number that's going to give me a nice clean look, and that might be something like 256. And again, I need to refresh my view. So that's pretty good. It's not perfect, but it's better than what we had. We could increase the number of minimum samples, but that's not actually going to do very much. It just means that we're never going to have fewer than, in this case, eight samples. But if we combine all this with denoising, we can get really fast interactivity with extremely clean renderings in the viewport. I'll open up the denoise section and turn it on and we see the denoiser is set to automatic. The options you see here will depend upon what hardware you have installed. I'm going to choose optics because I have Invidia hardware. Okay, so now we also have the start sample, which means, at which point is denoising going to begin? So if we set, for example, our MAX samples to 16 and our minimum number samples to eight and our start samples to eight, what that's going to do is it's always going to calculate eight samples then it's going to denoise at that stage. It's going to start denoising after it's done eight samples per pixel. And it's only going to go up to 16. It's never going to go any higher than that. And we have really fast interactivity in here. As soon as I release the mouse, boom, it's done. There's no waiting. So this is really highly optimized. It's going to depend upon each particular scene, whatever lighting you have in your scene. But these are pretty good settings for viewport interactivity or interactive production rendering. 12. Adding a Sky Texture: For natural day lighting, there's a system built into Blender called the Sky Texture, and that's a procedural texture that generates a high dynamic range panorama that can illuminate the scene and also optionally provide a backdrop. We're not going to use the backdrop. We just want to provide lighting. So we're going to swap out this environment texture with this high dynamic range image for the procedural Sky Texture. I'll go to the world properties, and we see the Sky Texture parameters listed in the color section. We want to replace this environment texture with the Sky Texture. Click on the yellow dot, and that's going to allow us to replace this node in the shading network. Click on that yellow dot. We get a menu, and in the texture section, you want to choose Sky Texture. And now that replaces the image based environment texture node. And if we navigate around in this user perspective view with the middle mouse button, we can see that there is actually an environment now. But there's no sun in the sky here, and that's because EV doesn't actually support the Sky Texture fully. If we wanted to use this in EV, we would need to create a light object. We can set the light object to be of the type sun, which is going to be a distant light source. So we could get a similar effect as the Sky Texture would give us. But we don't actually want to use EV as our production renderer, in this case. We want to use Cycles because it has bounce light and retracing, and it just has much higher quality than EV. I'll go over to my camera view, and I want to go into viewport Shading mode rendered. And because I've hidden the header, I can't get to that, but I created a keyboard shortcut, which is F five, which allows me to access the viewport Shading menu, and I'll choose rendered. And now we see a cycle's rendering in this view. And it looks very similar. And the reason is that although there is a sun in this cycle's rendering, the sun is actually coming from the positive Y quadrant of the world. In other words, the sun is actually behind the fence here. And if I hide the backdrop object by toggling its visibility off in the outliner, now we see the sun shining directly into the camera lens. Okay, I'll re enable that backdrop. The most basic thing we can do with the Sky Texture is move the sun's position in the sky, and we can do that with the sun rotation parameter. We and drag on that slider and we'll be rotating the sun around, and we can start to see the sun creep out from behind that backdrop object and just keep increasing that value. And if we bring it all the way up to a value of 180 degrees, then the sun is coming from behind the camera. So there it is at about 180 degrees, and we've got a shadow cast from the ball onto that backdrop. Well, obviously, this is super, super intense. It's way too bright. We need to art direct the look of our Sky Texture. We'll do that in the following movie. 13. Art directing a Sky Texture: To artict the natural daylighting coming from the Sky Texture, we want to work with the Sky Texture properties. I want to first enable Cycles in this camera view. We got a keyboard shortcut for the viewport shading menu F five. I want to be in rendered mode. Now we've got the interactive production rendering. And we can see that the sun and sky is way too bright. We can reduce the strength in the world properties here. Now, we could deal with this other ways. We could go into our render settings and change the exposure, but that might complicate our lives. So just to keep things simple, I'm going to reduce the strength of the Sky Texture. I'm going to bring that way down to a value of 0.05. And now I don't have any overexposed areas in my shot. I'll probably need to change that strength again later once I finally settle on where I want the sun to be in the sky and the color of the light coming from the Sky Texture and so on. We've seen the sun rotation parameter. That's going to control the angle of the sun around the horizon. There's also the sun elevation, which controls the angle of the sun above the horizon. With a high sun elevation, the sun is coming from high in the sky. With a low sun elevation, the sun is coming from low in the sky, nearer to the horizon. And notice that the color of the light actually changes. So this is actually a physical simulation of the sun and sky. And as the sun goes lower closer to the horizon, the rays of light are passing through more layers of atmosphere, which causes that light to become orange or gold. And the hour just after dawn or just before dusk is known as golden hour in photography and cinematography, because the light is gold. So we just want to artert this, set this to something that we like. I'm going to set my sun rotation to nearly 180 degrees. And the reason is I'm concerned about the shadow of the ball. We'll see this in more depth in the next movie. But I want to be able to make sure that the first frame and the last frame of my animation will actually precisely match. So it'll be a looping animation. And I want not just the ball to leave the frame, but I want that shadow to leave the frame as well. So I'm going to set my sun rotation to approximately 180 degrees. Additionally, we can account for different weather conditions. What we're seeing here is sunlight on a very bright sunny day. In this shot, there are no clouds between the sun and our set or layout. We can't actually add clouds in our Sky Texture. Some renderers will actually generate clouds, but blenders cycle Sky Texture is not one of them. But we can at least change the quality of the light that reaches our set. And one way we can do that is by changing the size of the sun. This is an angular value that represents the size of the sun in the sky, and it's set to a naturalistic value. In other words, this is the actual size of the sun in the sky as seen from the surface of the Earth. If we increase the sun size, we'll effectively give the effect of the sun being behind a cloud. And if the sun size is up high enough, we actually won't get any shadows. Let me bring that down to let's say maybe ten degrees or 11 degrees. And now we see there is a shadow cast upon that backdrop object, but it's a very blurry shadow, and that's because the sun is very, very large. And this is giving the effect of the sun being behind a cloud, and the cloud acts as a piece of giant diffusion gel that's scattering all the light. Okay, well, if the sun were behind a cloud, then the intensity of the sun would be reduced accordingly. I can bring that sun intensity value down, and that's going to cause the overall color of light to become more blue because now with a sun intensity of 0.1, the sun itself is very dim relative to the light coming from the sky. That would mean I would need to increase the strength accordingly in order to get a good exposure, maybe I could bring that strength up to like 0.2 or something like that. So now, this is the effect of a very cloudy day. So we can adjust these parameters in order to get different looks to our daylight. I'm actually going to set this back to a sunny day. Set the strength back to 0.05, my sun size to 0.5 degrees and the sun intensity of one. 14. Lighting and shadows in compositions: The last detail I want to take care of in directing the Sky Texture is taking into account the fact that this is not a still image, but rather an animation. And additionally, it's an animation that I want to cycle endlessly. And that means the first frame and the last frame need to match. Otherwise, there will be a discontinuity or jump cut. That means that I want the ball to be completely out of frame on the first frame of the animation. And I want the ball to be completely out of frame on the last frame of animation. But the ball is not the only compositional element that's moving. The shadow of the ball is also important here. So to see that, we need to enable Cycles in this camera view, hover my mouse over that camera view and set the viewport shading mode to rendered. And I've created a keyboard shortcut for that F Bive set that to rendered. And now we're rendering in Cycles in that view. And I want to move my ball around and sort of preview my animation. And I can also preview the shadow of the ball. In the front orthographic view down here, I'll select the ball. I'll invoke the move tool with the G key and preview my animation. So here's where the ball might be on frame one. And we can see the ball is out of frame, but the shadow is in the frame. And then in the middle of my animation, ball's going to be on the ground here. And at the end, it's going to be somewhere off camera. Pretty close to the ground because it'll have bounced, and it'll land over here somewhere. Okay, well, I'll need to bring that ball pretty far over to the right in order to make that shadow fully leave the frame. But I have freedom to do that, I think, at the end of the animation. I can just extend my animation a little bit farther. Whereas at the beginning, I don't want there to be a long gap between the first frame of the animation and the first frame of the actual action when something actually takes place. So just artistically, aesthetically, I think I want this shadow to be out of frame at the beginning of the animation, but the ball to be just barely out of frame. So that means I probably want to adjust the sun rotation parameter or property. So I'm going to reduce that sun rotation and set that to be something less than 180 degrees, maybe something around 160 degrees. And I'll check the result. So here, at the beginning of my animation, both the ball and the shadow are out of frame. In the middle, the ball and the shadow are clearly visible. And this is good, too, because now the shadow is more interesting. It's not just, you know, the sun coming from directly behind the camera, Sun's coming from somewhere off to the side, a traditional key light setup. And then at the end of my animation, we can see that the shadow is still visible, even though the ball is out of frame. That just means I'll need to extend my animation a little bit. So the ball and its shadow can completely leave the frame. Okay. So I think I've art directed this pretty well. I've got good daylighting here, and I've taken into account the fact that this is not a still image, but an animation. And that concludes our course on camera and daylighting. 15. Next Steps: We've seen how to create a camera and frame a shot and also how to apply basic daylighting with the Sky Texture, and I snuck a couple other things in there, such as how to create hot keys and wrangle a workspace. So now we've got our shot set up and we're ready to animate. And that's going to take two courses. The first hour of that is Part four. That'll be the basics of keyframe animation and how to set up a basic rig. And then P five will be the Finesse of animation, talking about interpolation between keyframes and also animating deformations for squash and stretch. I'll see you in part four.