Easy-to-follow video tutorials help you learn software, creative, and business skills.Become a member
We come now then to the final choice that V-Ray offers in terms of image sampling engines which is of course the adaptive subdivision system. Again, we need to come up to our Render Settings window, if we want to enable that, so let's open that dialog. Come to the V-Ray tab. Inside our Image sampler rollout, we want to use our drop-down and enable the Adaptive subdivision system. In here we get access to of course our control parameters. We have the maybe familiar Min and Max rate settings, we have a Threshold setting as well.
But there are one or two options in here that I just worth making note off. For instance we have this Edges option that when checked, will allow us to essentially supersample the edges of geometry in the scene. We have this Normals option, which along with this Threshold value will allow us to supersample any surfaces that have widely varying normal directions. One option that we definitely will want to leave checked is this Jitter option. It basically allows the samples in the system to be slightly offset to be displaced a little bit so as to produce better antialiasing of horizontal and vertical lines.
Our Min and Max rate values do work again a little bit differently from the straightforward subdivisions used by the other two image sampling engines. We do need to be careful when setting values in here. To demonstrate what I mean, if we use a Max rate value of 0 in this system, then we are essentially using a single sample per pixel. If we set a value of 1 in there, then we will be using four samples per pixel. But if we just jump up to a Maximum rate value of 2, we are all of a sudden using 64 samples per pixel.
And that kind of exponential increase continues as we step up the Max rate values. If we are using a Max rate of 5 or 6 in our scene, we could potentially be using huge numbers of samples per pixel that would definitely slowdown the system, so just be careful when entering values in here. As of course we have our other image sampling engines, we want to work with some parameters in here and see how they affect the rendering of our scene. Throughout this exercise we will be working with an Adaptive subdivision Minimum rate of -2.
This means that the system can undersample wherever it feels it is appropriate. A Minimum rate setting of -2 means that the system can go as low as using a single sample for four pixels in our rendered image as we say the Adaptive subdivision system is able to perform this kind of undersampling very nicely. We also of course want to make change to our Max rate. We are going to set this to a value of 0, this means we are using a single sample per pixel and we are going to make our Threshold value 0.1.
This means the system is not really that sensitive to adding extra samples into the render. Now of course we can jump into Adobe Bridge and navigate to our pre-rendered images. So come into our assets folder, into the Ch06 folder. Let's grab our Adaptive subdivision (Adv_ SubDiv) renders and select the first one. And if we just maximize this by using the Spacebar and then the left mouse clicking, you can see perhaps not unsurprisingly that we have a very low quality image, we have lots of breakup along the edges of our geometry, lots of noise in the scene.
Although the lines in our materials not faring too badly, our noisy material in terms of our vast object and our surface here, definitely missing some detail, but still not looking too bad. And we have a reasonably fast render time of a minute and 18 seconds. Clearly though we need to do work here. So let's go back into Maya. Let's set our Max rate up to a value of 2 and with a no other changes, we'll go and examine how that will affect the rendering of our scene. Let's select our second image and maximize that.
And you can see we have improved things, the edges of our geometry looking much cleaner, our lines, definitely they are working very nicely. We still have lots of noise in the scene of course and we are still at three minutes and twelve seconds, so pretty fast we could definitely use this as a test render for sure, but if we wanted to clean this up for our final output, we still have a little bit more work to do. So let's jump back into Maya and come to our rate settings. Of course as with the Adaptive DMC system, increasing our Maximum rate value is not always the best thing to do.
Sometimes we need to make the system a little more sensitive to using the samples available in the Max rate. In this instance, if we add zero into our Threshold setting, so 0.01 and if we were to go back to Bridge and pull up our third render, you would see that just making that quick tweak to our Threshold value has really helped clean up the noise in our image. Still a few little areas that we may want to work on, may want to clean it, we can see there is some small noise detail in the shadow, small noise detail on this diffused brick and of course we are all the way up to 21 minutes now.
But we can tell that what we would need to push things a little bit further if we really wanted to clean up this scene. So let's come and set a Max rate setting of 3, we are going to leave our Threshold value where it is and if we examine our final Adaptive subdivision render, you can see that we really do clean things up very nicely indeed. If we were being really picky, we would still say that some of our shadow areas have a little bit of noise. We do need to remember that the DMC sampler threshold will still be affecting noisy render effects in the scene.
So we may need to tweak those a little bit even though we are working with the Adaptive subdivision system to clean those types of effects up. Of course we are all the way up to an hour and five minutes, so we have really increased our render times. And really that is because the Adaptive subdivision system is not generally speaking good at working with scenes such as the one we've set up here, one that has lots of noisy or blurry effects in it. The Adaptive subdivision system works well with scenes with lots of black or plane color in them, scenes that have very few noisy effects, it can give you very fast and clean renders from such a shadow.
Of course as we have mentioned its weakness is with noisy effects in the scene and another big one is the fact that it is extremely memory intensive. The Adaptive subdivision system holds all of its sampling information in memory so we can find ourselves running into out of memory errors with big scenes. That having been said, if we choose the scenes to use this system on wisely, if we play to its strengths, then the Adaptive subdivision engine can be a powerful tool at our disposal. What we want to do next then is really have a look at how the final renders from each of our image sampling engines compare.
Let's see how they stack up against one another in terms of the speed and quality of output.
Get unlimited access to all courses for just $25/month.Become a member
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.