Easy-to-follow video tutorials help you learn software, creative, and business skills.Become a member
When we're constructing an interior rendering, anything we can do to save time and preserve quality is very valuable. I've fine tuned my lighting, materials and final gather, my bounce lighting quality. And I'm almost ready to render, however, I know I'm going to take a hit in time in rendering that animated camera. So what I'll do is look at caching that final gather and seeing if I can reuse it in here. I'll switch over to the animated camera, pressing C for camera and choosing camera ani. Now I'll go into my render setup, and in here I'll scroll down or roll up final gather in GI and look in the Reuse section.
What I'm going to do in here is to look at different ways of caching that final gather. There's a couple of different modes available when we choose to cache this, a single file and once per frame. The difference is fairly straightforward. If we're only animating a camera or running still images roughly in the same direction, we can use the single file option and cash the final gather once, and then recall it for all those frames. As long as the camera animation isn't too drastic, like flipping around a hundred and eighty degrees, this will work.
So this animated camera doing a slow dolly in and nodal pan to look up the stairs is ideal. If we have animated objects, we can cash this final gather once per frame. What this lets us do if we need, is to cash the final gather at a smaller size, for example, and project it onto a larger file, maybe working at 960 by 540 for a 1280 by 720 frame. I'll choose the single file only option, and what I'm going to do is in the final gather map, incrementally add those points and go in and name that file.
Right now it's called temp.fgm and I'll delete it by pressing X to delete, and then go in and name it. It'll browse out to the render assets, and I'll name this one lobby. It'll call it an FGM and it's not necessarily a map like an image you can read. Rather it's a final gather cache, and so it's simply a file to hold that lighting data. I'll click Save and check generate final gather map file now. My final gather cached. It didn't actually go through the rendering stage, because it was just for final gather.
The advantage in doing this is, is we can run our final gather with a much higher accuracy, because we're only doing it once and then save it and load that in. Loading in the same high accuracy final gather, time and time again. Now what I'll do, once I've run it, is drop down here under the final gather map section, and read the final gather points from the existing file. This is now reading that FGM I've cached, so when I go to render this particular frame, it'll jump straight into rendering. I'll show what this looks like.
I'll click Render Production, and re-render that daylight scene. As you can see, I didn't cache the global illumination, so it jumps into photon emission, and then will freeze for a sec, load the final gather, and jump straight into rendering. It's a much faster render. And even if it takes an extra hour in front to render the final gather, we get a time savings, per frame, of minutes possibly. And this really translates into a lot of time. Think of it this way. This animation has 72 frames. If I take that render time from eight minutes down to four minutes, times 72, I'm saving 288 minutes of rendering time.
Remember, when you are doing this the camera can't flip around drastically. And if your running just a single file, there can't be moving objects. If you need, you can cache that final gather once per frame and do it at a lower resolution and then open that or reload it when you are rendering the hi-res frames, and this is another way to same time. My render is done, and I'm looking at a much lower render time. This is 2:31 a frame, compared to nearly eight minutes previously. Caching the final gather has its advantages, where we can run it once very very high and then save it and reuse it time and time again to speed up our render time.
Just make sure you're very careful about how you're caching and what's animated or not in the scene. When you're ready to render out your passes then and render out your scene make sure you go into the render setup. And in the render setup, in the Common tab, bring that image size back up. Here it is at 1280 by 720. If I was going to render this full, I should make sure to cache that final gather full size, take the time hit once, and then render out those full frames with the correct lighting. In here, because I'm going to render an animation, I'll put it in as the active segment, one to 72, and scroll down and make sure that frame is named and has the correct extension chosen in the render output section.
I'm using OpenEXR files because I want the additional G-buffer channels of my material ID, object ID, z depth, velocity, transparency, and render node ID. Lastly, before you hit render, go into the Renderer tab and check on the sampling. Right now, we're dealing in a decent, but moderately low quality anti-aliasing. I'll boost this up to a quality of 0.75. It'll take a little longer per frame, but it'll get a better quality overall. You also may want to consider the blur you're using.
If you are applying depth of field in Post, you may not need the softness that a Gaussian blur gives, and you may be able to choose Triangle, for example, or a Box blur, for a quicker render. This may require some tests and is a great reason to run a test, get those G-buffers out, check how it looks all the way through Post and come back and use that to calibrate how much quality you're putting into your rendering. When you're all ready, you can hit render. Then relax, or render over night, and come back to a sequence of well constructed interior images.
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.