Hello, everyone, welcome back. And in this video we're going to be talking about how we can improve our render settings, our speed, and rid our actual full scale render of noise. So our target is going to be 4K. We're going to do this in 3840 by 2160. The current frame right now is just set to 53 just because I kind of have this hair simulated on the pillows. And I just hit play and that's where we just kind of stopped it at so that it would be resting. We'll save this render settings window for the end portion of the video. Right now we're just going to concentrate on the samples for pixel, diffuse depth, specular depth, and talk about how we can justify our settings. First thing I'm going to do is just grab a render region. And then I'm actually going to use the Lock. So what this is going to do, it's going to lock our resolution at 4K. And once I do that, it's just needing to reload. Okay, so with the render region I can just kind of drag out areas that I want to get a better look at, maybe this breakfast set and the candles area. And you can see now that we're seeing much more detail and this is going to really help us magnify and see our shaders and see how long it's going to take for this to actually clear up. So I'm assuming at this point that 4000 is going to be on the high end. I do not want to go over 4000, definitely not. I try to keep most of my renders around 2000 to 3000. And you can see, it's looking pretty good at 3800 but there's still just a little bit of noise here. So I'm going to play with the GI clamp right now. So the GI clamp is the overall contribution of indirect lighting to the scene. So that's going to tie in to the diffuse depth as well. So these numbers are going to affect each other. So when I usually do test scenes I can set that to one and play around with the slider just to see how that's really affecting it. Of course, we bring this down, so closer to zero. We're going to lose that overall indirect illumination in the scene. Okay, so that's definitely going to be a threshold. So if I set this back to one, and then I come up to the Diffuse Depth, and I set that to one. You can see that we're kind of getting the same type of look from the GI clamp when we had that down very low, close to the zero. So you may be asking yourself, "What is the proper diffuse depth?" Well, the best way that I can put it to you is to adjust the diffuse depth accordingly with your scene. So, we're at six right now. So I pull this down to five, do I see a large difference between five and six? It doesn't seem like there's a noticeable difference, at least not yet. So, from five to four, put that back to four. From five to four, not quite a huge hit, maybe here on this left candle. So from four to three, definitely seeing a noticeable difference, especially here on the auto min. So I believe I'm going to leave this at five, how it was, and that's kind of a simple way that you can arrive at your diffuse depth count. Now your specular depth count, if we were to just sample the translucent and refractive materials, as I move this down from 11 to 10, you can see a noticeable difference here on this glass. You can see that the two panels here from the geometry become black. So this is tracing the ray depth going through specular materials. So as we increase this to 11, you can see that that becomes better. As we go to 12, you can see it becomes even clearer. And let's keep increasing this. So I think it's safe to say that 14 will do this pretty well. Now, the caustic blur setting, I believe by default, is set to 0.02. I don't usually worry about caustic blur inside of path tracing unless I need to really focus on something that is going to show caustics. I'll switch it to path tracing Montecarlo, the PMC colonel. I feel like that is way better for something like caustics. So I usually leave this around 0.2 just in case if there is any kind of caustics that are showing up in the scene, we can blur it by a pretty good amount. Okay, so we're back again at this GI clamp. So another way that the GI clamp can really help us is cleaning some different parts of noise. Okay, so we're at one right now. I'm going to go ahead and we're going to use this surrounding area to kind of see the difference of this value. So let's go to 0.8. And you can see the difference from 0.8 and 1 from in and outside of this boundary. And it's not a huge difference. So let's keep driving this down. Let's say 0.6, and that we're starting to see a little noticeable difference, this is appearing a little darker now. Again let's go down by 0.2 again, so to 0.4. Even more appearing dark. So let's go to 0.2, and now you can really tell a difference between these two images here. So the GI clamp will help rid some noise. There have been many cases where I have used this GI clamp to kind of choke up some sampling from the indirect light that was not needed. So I'm going to leave this around 0.45. Now this is a value that is going to change definitely from scene to scene. And if you're going to use this as a way to choke out some noise, then you're going to have to play around this just like I did in this scene. By no means are any of these diffuse depth, specular depth, or GI clamp a number that's going to be reliable through every single scene. Okay, so we're going to move on to another area of the image over here. And you can definitely tell in this area that it's appearing a lot darker from where we had started before. I believe this part of the process is really all about balance. So adjusting it to 0.6, I belive, is a pretty fair compromise. Especially as we get closer to the higher samples for pixel count, this is looking very, very good. So there's one more setting I want to talk to you about, and that is going to be the Coherent Ratio. So let's sample this side of the image. And I'm going to start to move this up. So as I increase this value you can see there's a lot of weird color artifacts that are happening at the very beginning. But as the sample count rises up, you can see that it's getting to be a lot clearer and the image is starting to look really nice and sharp, especially around the 3000 pixel count range. So let's go a little bit higher to...let's try 0.45, and now you can see that that darker area is rendering pretty quickly. Let's try 0.6. I find with the coherent ratio that it is a speed boost. But from an animation standpoint when you're going frame by frame, you're going to get a little bit different results with this coherent ratio, and that's where the static noise comes in. The static noise is going to try to keep that difference between frames very close so that you don't see a noticeable flicker in your animation. So just out of habit I always check on the static noise in case I do have to fly camera through this. Okay, so let's grab a larger area and see how that is looking. Okay, so as our render is nearing 3000 samples for pixel, you can start to see that at 4K everything is starting to look very clear, starting to look very nice. There's not really any real visual noise. So going from 3000 to 4000 I don't believe is going to be worth the time that it would take for it to increase 1000 pixels. So I'm going to go ahead and just decrease this 3000, and that pretty much will do for the path tracing colonel settings. The next thing that I would like to look at is our actual render passes. So let's go ahead and go in order of these tabs here and talk about some of these things. Our geometry control and the engine and everything is going to be okay. This is basically for animations, this top part up here. So it's really not going to come into effect in our scene. The render buffer type, I always set to float tone mapped. So that means that there's going to be floating point information in the render. If we're going to use all of the GPUs, I just check this out of habit. I'll be using network rendering for the final rendering just to speed it up. So, basically, if you have other computers or work stations on your network, maybe at your home office or at your actual office, you can launch the Octane Render Slave, and this is where you're going to be seeing a list of those chosen computers. So over right colonel settings, I'm not going to enable this. What this will allow us to do, change this settings that we have on the Octane render settings panel. Our render passes, I'm going to go ahead and enable. And I'm going to set this to a directory. And if we were going to render EXR, we're going to have a ZIP format compression. We're going to have full 32-bit floating EXR. So if I turn off of the render region setting and let's just go ahead and send everything right back to Octane. Once that's refreshed I'm going to right click and I'm going to say Toggle Info, and that's going to allow us to see our different render passes here at the bottom and visualize those. So let's talk about the passes. The diffuse pass is actually the direct and indirect combined into one pass. I like to have those separate because sometimes I like to adjust the intensity of the indirect or vice-versa. So the reflection is the direct and indirect combined. But again, we'll split those out. We're going to have refraction. We're going to enable transmission. As we enable transmission, we have it appear here at the bottom. If we hover, it says, "Contains all samples where the camera ray is transmitted by diffuse material on the first bounce. So, if we actually click on the transmission, you can see the overall contribution of that transmission. Now this will get clearer and clearer as it renders, and we get more samples. And right now our coherent ratio is kind of affecting the overall color. It's making it look a little bit odd. But anyways, we can go back to the main. The render layer, the lighting passes and the render layer mask sections, we're not going to be using right now. But if we come down to the info passes, we are going to be using the Z depth and the ambient inclusion. I also wanted to point out that the wire frame is pretty cool, just to show a simple wire frame of the scene. We can look at the ambient inclusion of the scene, and then adjust the AO distance if we need to. Right now it's a little spread, so if we pull this distance back a little bit, you can start to see we're going to introduce a lot more white values into the scene. And with AO, we really don't want this kind of middle gray so much. We just want this to kind of give us a little bit more contact shadows and make our scene feel like it's grounded. So I'm going to pull this back pretty far. So this is going to be scene dependent. But for me, when this is actually multiplied, I think this is going to be the result that I'm going to like. I want to be able to see those nice contact shadows here around the trays, around the floor boards, and around the seams. I don't really want this to be spread so much and I would like to keep the detail how it is on this fireplace, for instance. Okay, let's go ahead and visualize the Z depth. So right now with the Z depth, it's a little weird. And we can adjust the depth max value to get kind of what we want. So for me it looks like a value of 10 is going to be a pretty good compromise of collecting everything that we want in the scene. Imagine this kind of being a fog. The areas that are going to be close to the camera are going to be dark values, and then as we go further back into the scene it's going to become lighter values. So this represents all the objects in our scene pretty well. All right, I think that will do it for our render passes. I'm going to reset our directory real quick. And once that is named accordingly, we do not have to use the Cinema 4D multipass or regular image save outputs. So we're going to be able to just render this from the picture viewer to this directory, and we'll still be able to monitor everything while it renders. The last thing I forgot to mention is I'm going to go ahead and increase the max samples on our info passes from 128 to 256. That's just going to give us a little bit more clear Z depth, AO pass. Let's go ahead and render this to the picture viewer. Okay, so while that is rendering, that will conclude this video. In the next video we're actually going to take a look at how to take all these passes into a compositing package and be able to tweak the image to our liking. The compositing package I'm going to use is Fusion, but you guys can follow along in any compositing package, Aftereffects, Photoshop, Nuke, anything that you like to use. I'm going to go ahead and use Fusion. Fusion is available from blackmagicdesign.com under their products page in Fusion. And you can download it for OSX, Windows, and Linux. So we will see you guys in Black Magic Design's Fusion 8. Thank you so much.