Architectural Visualization with C4D and Octane: Compositing in Blackmagic Design Fusion, Part 2

Photo of Brandon Clements

Instructor Brandon Clements

Share this video
  • Duration: 18:19
  • Views: 1612
  • Made with Release: 18
  • Works with Release: 18 and greater

This video will go more in-depth about Blackmagic Design Fusion. Here we will learn about basic nodes to tweak our image to create a final look and feel for final presentation.

show less

Transcript

Welcome back, everyone, to part two of our Fusion 8 video. So, what we're going to do is pick up right where we left off, and I'm going to drop down another merge node in here. So, I'm going to say that this is going to be the background, and what I want to do is take my reflection indirect, and I'm going to just copy this and paste it again, so I have another copy of the indirect illumination. If we want to look at that, we can just hit 1 on the keyboard. So, you can see in this pass, we have kind of all of the metal showing up, all this nice kind of shiny look to it. What I'm going to do with this is, I'm actually going to drop down a new node. So, we're going to go up here to blur, and just drop down a simple blur, and I'm going to change the blur size to around 20. And changing that to 20 is just going to make it blur a lot. So, let's just put that into input, and I'm going to view this in our first viewer. So, you can see that this is just going to kind of add a subtle bloom to everything. So, I'm going to put this onto the foreground, and let's view this in our first viewer. So, number one, this new merge node. And I need to actually drop the alpha gain now so that it adds, okay. So right now, this is just a little too strong, so I'm going to come up here to the blend on the blur and I'm going to put that to 0.2. The cool thing that what you can do inside of Fusion 8 is you have this S1 and S2. So for S2, you could actually blur this even more, and then you can just click S1 and S2, and see the difference and figure out which one that you really like. So, let's just go ahead and say S2 looks best, and we'll continue on. Now, we're going to start to utilize our ZDepth pass. So, I'm going to bring this just over here, and I'm going to bring down a channel Booleans node, okay? So, the channel Booleans has a lot of different operations we can do to it. We can use this as a and/or, we can also use it as a negative, which would be an inverse node. There is so much that we can do to this. So, we're going to bring this into the background, and we'll bring our ZDepth into the foreground. And that operation really matters here, because you can see we can change between background and foreground layers, and what we are actually going to load into the red, green, and blue in alpha slots, okay? So, let's go ahead and hit 1 to view what we have right now. So, the channel Boolean basically can take channels and swap them into different channels. So, kind of think of it as a switcher almost. You can put in an input, and then reroute it to a different output. And I know that sounds kind of confusing right now, but let's go ahead and just say everything is going to be set to do nothing. And as I do so, you can see the changes on the viewer one. So, everything is set to do nothing, and the operation is just going to copy the information over for us. So, what we need to do is load in a new Z buffer, and we're going to take the Z buffer from our ZDepth. So, in our auxiliary channels, we need to enable our extra channels. Don't forget this step. It's going to enable a slew of different types of inputs. So, our ZBuffer is going to be the luminance value of our ZDepth that we had from Octane. So, we're going to actually use the lightness foreground. So, we'll take the lightness foreground, and you can see that nothing has changed. But what we have essentially done is just enabled to have a Z buffer into the channel Boolean. So now, we can utilize this as a Z Operation. So we have a Z Channel now. We're going to go to Add Tool, and we're going to add a, let's see, deep pixel, we're going to add a depth blur, but we're also going to utilize this as a fog, too. So, let's go to deep pixel, and let's add a fog node. So now, we can start to branch off and utilize this Z buffer channel for the fog and the depth blur. So, let's first just do the depth blur. We'll put that as the input. Let's go ahead and view that on number one. And we're telling it to use the Z Channel from the Boolean, and we're going to adjust the blur size, the depth of field, the Z scale, and we're going to pick a focal point. So, to pick a focal point, just left-click and drag and drop, and just pick something here in the environment. So, let's say like, the lamp right here, and as we adjust the blur size, let's go ahead and bring that up even more. And we're going to take the scale of the Z down. And as we do so, you're going to see a dramatic change, especially once we get down to these lower values. So, let's try something like 10, and then let's zoom in again to the lamp. And you can see, the front part of the lamp is in focus. The Z Scale is almost going to compress everything. So, you're going to get a very shallow depth of field, and our blur size is only set to around 10, if we rounded that up. So let's say, let's say something crazy like 60, and you can start to see we notice the box in the filter show up very evident. Now, the good thing about the box is that it's super-fast to preview, as we go to Soften, it's going to take a little bit more time. So, you can see our nodes are kind of working here on this overall graph. And you can see that's a little bit better. And as we go to super-soft, you can see everything coming through, and that's going to give us a better result. But it's going to be slower. So as you work, it's probably best just to have this on the box, and then we can adjust parameters. So, I want this to be a just subtle effect, so I'm going to set the blur size down to 20, let's set the Z scale to around 20, and the depth of field is going to kind of act as a threshold. So, if you want less, you can increase this. So, I'm going to just put this to one just to kind of knock that back a little bit. And then, let's pick a focal point, maybe something like, you can see that it's pretty interactive, too, as we left-click and hold overtop of these objects. So, let's just pick the croissants. And we're going to set it to super-soft, and our blur is still a little bit too evident. So, let's go ahead and set it to two. We would just like this to be a very subtle effect, make it look a lot softer. So, this allows us to kind of tweak in post our depth of field. So, let's utilize the channel Boolean. Again, we're loading the ZDepth luminance into the channel Boolean to act as a Z pass. Right here you can see, set to Z. So, we're going to do the same thing with our fog. We're just going to plug this into our image and let's hit 1 on the keyboard to just start focusing on this fog. And to zoom out, you hold Ctrl+mouse wheel. If I hadn't mentioned that before. So, let's go ahead and pick our near plane, which is going to be somewhere close to the camera, and our far plane, which will be the opposite end. And immediately, you can start to see something happening. As we adjust the Z Scale, you can see that this is starting to envelop the entire room. There's a lot of different things that you can do with this fog, but for us, we're just going to utilize it as a way to kind of brighten the overall appearance of the room. And of course, we have this fog opacity that we can adjust, and the color. So, let's go ahead and pick a brighter kind of color. We can use this or we can just twirl down and get kind of a little color wheel here. So, we'll kind of increase the blue just a little bit more. And just something to kind of play around with and mess with. So, I'm going to blur this fog, so we're going to come up to blur, and let's add just another blur. And let's view this on viewer one. But first, let's kind of bring our Z Scale up, so that we can encompass the whole room, and then we'll blur. And bring this down just a little bit, just so we can see the effect. So, in this whole node tree, we are basically accepting all of this data, and it's going to blur. So, it's not only going to blur the fog, it's going to blur everything previous from this node, too. So, everything flows into the blur, so everything's going to get blurred. So, I only want to change this to about five on the blur, and let's come to the fog, and let's put the Z Scale somewhere around five, so it almost envelops the whole room, and we're going to put the fog opacity around, like, 0.2, something kind of low. And this is just going to be a subtle effect, so we're going to add our merge node. Again, like I said, we're going to be using that merge node a lot. Let's go ahead and put this overtop and we're going to put the depth blur on the background. So, let's go ahead and hit 1, and we're going to use our alpha gain. And let's take our blend slider and start to pull this back a litlte bit. So, like I said before, we only want this to be subtle, so let's try, like, 0.05, and that's going to brighten our room up. If we look at the merge node, so let's look at merge three on the second monitor, and then our merge four is on the first monitor, so utilizing the channel Booleans node to get the ZDepth information to create a fog and a depth blur, we've kind of added a little bit more atmosphere. So now, it kind of feels like a photograph, and we have all the settings here that we can go back and tweak if we want to create a different type of look. So, we can always go back and tweak later, since we have the node set up if we want to increase the depth of field, or we want to have that fog a little bit more present, we can do so. Okay, so let's move along. And we're going to start doing some more color correction. So, I'm going to show you a few nodes that I really love to use, so we're going to take the output and we're going to use this color curves, and let's view this in the second viewer. And we're going to kind of create just a subtle increase in our, a little bit of shadow, midtone, and we're going to kind of bring up the highlights just a little bit more, and let's kind of decrease the shadow to add a little bit more contrast. So already, we have a pretty good-looking image, just from slightly adjusting these knots. So, I think that's going to be pretty good as a base. Let's continue. So with this color curves node, we're adding a little bit more contrast to the shadows, we're boosting the midtones into the highlights. So, how can we have just more control? Let's just drop a color correction node down, and I'm also going to add a merge node. So, I'm going to put our color curves into the background, and we're going to add the color corrector onto the foreground, and then I'm going to take all this information from the color curves and pipe it into the color corrector. So, let's view this on viewer one, and let's take merge five, and we're going to view it on the second monitor. Let's go ahead and bring the alpha gain down to zero. So, think of it in Photoshop, when you duplicate a layer and you just set that second layer to add, we doubled our data. Okay. So, this is going to be a lot of fun. Let's go ahead in the master, and we're just going to add 1.2 to this value, just to boost it a little bit more. Let's come into the shadows and let's boost those shadows a little bit more, and we're going to add some contrast with the gamma, and we're also going to decrease the brightness. So, we're only working on the shadows here, and you can see this is creating some more contrast in those dark areas. And I am looking at the additive version, not just the color correction version over here on this monitor. So, let's go ahead and just put the color curves to one, so we can kind of see a before and after. And we can go into the midtones, and we can just basically tint the overall look of this image by sliding around this color wheel. So, if we wanted to cool it off just a little bit more, we could do so. And then in the highlights, I am just going to drop down the gain to around, let's say 0.8 or so. So, let's kind of step backwards. This is the power of having these nodes. We're going to go to our reflection indirect. We can view that. Let's view it here on this first monitor, and you can see that the area that is really blowing out is this part of the couch. So, what we can do is drop a color correction node, and if we drop that on top of our line, we can essentially just connect that very quickly. Kind of move these nodes around just a little bit. And we're going to come into the highlights, and we're just going to drop that down significantly. Let's put this on the first viewer again, and let's kind of get that to around 0.3 or so. We can see in our second viewer from our very last node, that overall it's still too bright. But we're going to continue on and fix that. Roping it off down the line is going to help us fix it further as we continue. So, let's keep going and I'm going to do kind of the same technique again. I'm going to add a merge node and this is going to be additive. And what I want to do is add a LUT from my file. So, let's go to LUT and file LUT. And I'm going to take this information, what we've got so far in this viewer, and I'm going to add it to this file LUT node. Let's view this on the second viewer by hitting 2, and I am going to navigate to a LUT file. So, there are a ton of free LUTs online. Adobe has a lot of free LUTs that you can download. What I am going to use and what I would suggest is Greyscalegorilla has a great LUT pack. I have basically just those files. I'm going to go to the everyday. I'm going to use the 3DL file type, and I'm going to use the Let There Be Light, so let's go ahead and open that. So, this is giving a really nice result to the overall image. It's taking those midtones and highlights, and just shifting them in a different direction with color. So you can see the difference between where we were and adding that LUT file on top of it. So, I really like what this is doing, but I'm not going to use this 100%. We're going to just drag this into our merge node, and let's take the output and put it on top, so this is the foreground and this is the background. So, we're going to take this, let's view it in number two, and again, we're just going kind of to blend this result together to around 0.4 or so. Okay. So, I'm really happy with the progress we've made so far. What we're going to do now is add another color correction node. We're going to bring this into the input, so let's go ahead and drop that into the input. Let's view it on viewer number two. We're going to zoom in here. What I'd like to do is, overall, just kind of take the shadows and let's make those just a little bit brighter, so we're going to push those shadows up a little bit in the gain. Let's come to the master and let's kind of push up the gamma a little bit, too, overall, just to make it a little bit more brighter. Let's try, like, 1.1. Let's come to the highlights again. We're going to kind of rope those down, just so we can see a little bit of detail here on the couch. So, the very last thing I'd like to do is just add a soft glow and I have it at zero, and I kind of played with these settings a little bit. Let's try, like, 0.15 and 1.5. Somewhere around there, just to give it a very subtle glow to the overall image. Okay. So, we have pretty much reached the end of our composition, so let's go ahead and drop down a saver. And make sure you navigate to where you want to save it, and name it .png. This is very important that you add an extension. So, let's go ahead and hit Save. We've set it to the frame range of one. Let's connect this up and let's hit 2, and we're going to save it as PNG, so this looks good. Let's go into our format. Make sure our compression depth is set to 8-bit, and we'll just go ahead and increase that. And then, we need to come to Export. Okay. So, this is very important right here. So, we have a LUT on our viewer and we've been working with the LUT on the viewer. So, I'm going to take this off because right now, we need to go ahead and change our output gamma space. Okay? So, we're going to say this is going to be SRGB, just like what we had here. And once I hit Apply Curve, you can see that it's just the same as what we had it before. Okay? So, this is very important, and this is with no LUTs. This is going to be baked into the actual file. So, let's go ahead and hit Render, and you can see that, at 4K, it took about eight seconds for a frame. So, let's go ahead and hit OK. Let's navigate to where that saved, and we can open this and just view it real quick. So, let's go ahead and view it in Photos. And you can see that everything looks pretty good. So, let's change some of the settings real quick. I just want to show you guys if we come into the 8-bit depth and let's go ahead and take the minimum compression down, let's go ahead and drop that down to zero, and you can see that it's thinking right now, because we can actually view what that result's going to look like. Let's go to File and let's name this one 02. And let's render this out. So, it actually saved over the last one, but if we go and look at the properties now, you can see that this is 31 megs, whereas the other one was 13 megs. Okay? So, the compression really does make a huge difference, especially if this was an animation. So, if we view this again with the photos, you can see that everything is much, much sharper, especially as we zoom in. So, there is absolutely no compression on this whatsoever. So, that will conclude this video on compositing in Fusion 8. I hope you guys picked up a lot of new tricks. We'll see you guys in the next one. Thanks a lot.
Resume Auto-Scroll?