Easy-to-follow video tutorials help you learn software, creative, and business skills.Become a member
In contrast to the Fixed Rate Image Sampling Engine, the Adaptive DMC System, as the name suggests, does indeed add adaptivity into the Image Sampling process. This means that based on user input parameters, the engine can make some very deliberate choices about where and how it chooses to place and use samples in the scene. Again, let's set the system up. If we come up to our Render Settings window, come into the V-Ray tab, we see we have the Image sampler roll out open by default. Let's go to our Sampler type dropdown and set now the Adaptive DMC system to be active.
Now, in here we have a number of user control parameters. We have our Min and Max Subdivisions. We also have this Lock threshold to DMC sampler Threshold setting, this means that at this moment in time the system is going to be working along with this Adaptive Threshold value to determine how to clean up noise effects in the scene and how to use Image Samples. One thing that is worth making note of here is the fact that we never really want to set the same value in our Min and Max subdivs. Essentially when we do that we kill adaptivity in the system and any render we take would be identical to using the Fixed Rate Image Sampling Engine with a subdivision value of 1 set.
So let's work with some parameters here and see what kind of effect it has on rendering our scene. So let's set Min subdivisions at 1, we'll set the Max subdivisions value to 3, and I'm just going to come across to our Adaptive Threshold value here and I'm going to set it to be 0.1. That means it's not really sensitive to noise in the scene at all. Now, if we jump across to Adobe Bridge and just navigate to our pre-rendered files, so just come into Exercise_Files, assets, Ch06, and of course we're working with the DMC system now, so let's choose our first image, Spacebar to maximize, then Left Mouse click to zoom in.
And as you'd perhaps expect, we have quite a low quality image, although apart from the noise, lots of areas of the scene aren't fairing too badly. Our lined material is looking pretty good. Our noise material, not looking too bad, there are some gaps in it that we can see, but hopefully they will clean up as we increase our quality settings. And of course we're getting a very fast render time of just 50 seconds from this scene. Obviously, we need to push our quality settings though. So let's go back into the Maya Application. So come to our Image Sampling controls and let's now set a Minimum subdivision value of 3 and we'll set Maximum to 7.
We won't touch the DMC Sampler controls, we'll just see what kind of an effect that makes as we look at our second render inside of Bridge. We can see we have cleaned up the noise somewhat. Our scene is definitely improving. Still lots of problem in there, lots and lots of noise that we can see. We have increased our render time now to 3 minutes and 7 seconds. Clearly then we need to push the quality. But if we jump back into Maya, oftentimes these Min and Max subdivs are not necessarily the best way to approach adding extra quality to the scene.
See, at this moment in time our DMC Sampler controls are not really sensitive to any noise that can be found. They're not really allowing the DMC System to use the full amount of samples available to it. So let's set this to a very sensitive value of 0.005. I'm going to keep our Min and Max at 3 and 7, and if we go back into Bridge and select our third render, you can see we really do clean up a lot of that noise. So things are starting to look very nice indeed.
We can see noise problems in the shadow areas. We can see fine grain noise on our diffused block, and of course we have bumped up our render time now to just over 14.5 minutes. Clearly, we need one more push on the Render Settings. So let's go back into Maya. This time we're going to set Minimum subdivisions of 5 an Maximum of 16 and, again, if we jump over to Bridge and take a look at our final render. You can see the quality is quite nice indeed.
Line detail looking nice, shadow detail looking nice, not really any discernable noise in the scene, picking out lots of detail in our counter texture. We are though at 27 minutes, which of course is significantly higher than our previous render, but still a long way short of our final Fixed Rate Render. Now, the difference in these two renders is due to the fact that this system is able to be adaptive, whereas the Fixed Rate engine of course is not. Now the DMC System can make intelligent use of the number of samples available to it.
It can focus them on areas of the scene where they're really needed. This means the DMC engine is extremely good at working with noisy render effects; so Motion Blur, Depth of Field, and others, this system can work very nicely with them. Another good thing about the DMC engine is that it is extremely memory efficient. It doesn't hold its sampling information in memory. In terms of weaknesses, there really aren't as such that I would ascribe to the DMC engine, other than the fact that while being able to work adaptively, it is not able to perform any kind of undersampling.
Well, let's move on then and take a look at our final Image Sampling Engine, one that can perform undersampling, and this is the Adaptive Subdivision Engine.
Get unlimited access to all courses for just $25/month.Become a member
107 Video lessons · 35531 Viewers
100 Video lessons · 6058 Viewers
94 Video lessons · 24835 Viewers
83 Video lessons · 10461 Viewers
Access exercise files from a button right under the course name.
Search within course videos and transcripts, and jump right to the results.
Remove icons showing you already watched videos if you want to start over.
Make the video wide, narrow, full-screen, or pop the player out of the page into its own window.
Click on text in the transcript to jump to that spot in the video. As the video plays, the relevant spot in the transcript will be highlighted.
Your file was successfully uploaded.