The final 3 podcasts have just been released!
Podcast #021: Loose Ends includes several tips and tricks that didn’t fit into the other podcasts.
Podcast #022: Music is an audio-only interview with composer Christopher Reyman.
Podcast #023: Sound Design is an audio-only interview with sound designer David Bewley.
One of the 3D artists who worked on PI is currently looking for a job. He’s been in Germany for the last two years and is willing to relocate anywhere within Deutschland. My website statistics show that there’s a good number of visitors from Germany so hopefully this will find its way to someone looking for a good generalist.
3d Demonstration Reel from eburke on Vimeo.
I just have to say I’m so proud of Eric. When he worked on PI (mostly helping out with the cars) he only had a couple months of 3D experience, but he’s come a very long way in a short amount of time. If he’s got a reel like that after less than 2 years in the industry, I can’t wait to see what his next reel looks like!
Wow. This is a Herculean effort by Alex Roman. Twelve and a half minutes… entirely CG except for a couple of very small composited elements. A lot of people have commented on the obvious awesomeness of the textures and lighting, but the thing I love most is his restraint with the camera. A lot of people get a nice looking CG set and their first instinct is to spin the camera all over it without regards to real-world photography. In this, every shot feels real because he’s using a virtual track/dolly/jib. He even uses the crane’s shadow as an artistic element in a couple of shots.
Plus, he really went the extra mile to imitate the way light strikes a piece of film. In the breakdown you can see the subtle glares and vignette’s he added, the DOF follows real-world lenses, but that little touch of awesomeness that makes it look organic is that he actually used separate colored vignettes to simulate the way light gets diffused as it passes through the three emulsion layers on a piece of older color film. Its sometimes those little details that can breathe life into an image.
The Third & The Seventh from Alex Roman on Vimeo.
Compositing Breakdown (T&S) from Alex Roman on Vimeo.
I missed the SIGGRAPH session that this was in, and I am still kicking myself for it.
Painting with Polygons from Outside Hollywood on Vimeo.
There’s a more complete explanation and some really cool examples here.
Podcast #019: Rending is online and shows some tricks for optimizing your renders and dealing with expansive environment shots.
First off, I’m no expert on the high level tech/math stuff so feel free to correct me if I’m wrong on anything, but I wanted to jot down some thoughts on the different renderers out there. Hopefully it’ll help point you in the right direction if you’re trying to make a decision whether to invest in a 3rd party renderer or use a built in engine.
Mental Ray – This is the one I used for PI. Chances are you’ve all run across it because it comes standard with XSI, Maya and Max. The plus side is that its one of the widest range of feature sets you’ll see in a renderer: radiosity, area lights, accurate reflections and refractions and that’s not even scratching the surface. The downside is that its slow and you’ve got to really break down large scenes to get them to render.
Renderman – This is a bit confusing because Renderman isn’t actually a renderer, its a standard. Think of it like a file format except instead of describing what the data is, it’s describing how to pass data in and out of a renderman compliant renderer. The most well known Renderman compliant renderer is…
PhotoRealistic Renderman (aka PRMan or Pixar’s Renderman) – This is Pixar’s internal rendering solution based on the Renderman standard which they also developed. While the standard is open source and fully-documented, the renderer itself is most definitely not. In fact, the only drawback to PRMan seems to be the price. It’s widely viewed as the most robust and stable renderer out there, but does require a fair bit of technical know-how to put that power to good use. As a result PRMan is found mostly in larger facilities needing to push an insane amount of data through the pipeline as fast as possible.
3Delight – This is kind of like PRMan’s close cousin. Its a renderman compliant renderer so uses the same architecture to describe its input and shader language, and its also a REYES renderer just like PRMan. REYES stands for “Renders Everything You Ever Saw” and its big advantage seems to be that its essentially a scanline technique, but the scanline works by dividing the mesh into micropolygons at rendertime. Basically, it splits the mesh to give an appropriate amount of detail for every pixel so subdivision surfaces are completely smooth to the eye, no matter how close they get to camera. Another big advantage is that REYES renderers are designed for motion blur and depth of field, so you get real 3D blur for free (more or less.) There’s a more in depth article on REYES here.
Brazil – I honestly don’t know too much about it other than its fast, cheap, and a lot of people use it for specific types of shots. As far as I know it only works with 3dsmax?
Arnold – This is another one that I don’t know too much about, other than I’ve heard that it gets a lot of use on big projects when they need to do global illumination.
POSTED BY JAVIER BRAVO
Vray – It’s often used in architectural visualizations but you can use it in all aspects as well. It’s fast and produces a very realistic render. It’s compatible with 3dsmax, Cinema and I guess the XSI version is in Beta.
As you can see, I’m trying to do this a bit like wikipedia. If you’ve got something to add, post it in the comments section and I’ll add it to the main post. Please refrain from any sort of flame wars over preferences. I’m most interested in what something does well, and what its weaknesses are as that will help people make informed decisions on what fits their needs.
Podcast #018: Single Pass Lighting is online and talks about a few shots we lit “all in the render” and how that compares to a multipass approach.
Just wanted to let everyone know that they’re re-running my “Lighting in XSI” course at FXPHD. The complete description of each class is here and the material is 99% different from the “XSI In Production” course that I just finished. All of the videos have been recorded but I’ll still be on the fxphd forums to answer questions directly.
As I was thumbing through the specs for XSI 7, I noticed that there is a new method of implementing BSP trees. I haven’t seen it in action, but supposedly it auto-configures all of the settings for you. If you haven’t messed with them before, BSP trees can DRASTICALLY reduce render times. Most of my render times on PI were cut in half, often more once I tweaked the settings. However, it takes a really long time to tweak every pass for every shot by hand, so I wrote a little auto-tuner for it. There is an auto-tuner in the netview, but it is wicked slow because it renders everything at full res and I need something that can tune a pass in a couple of minutes. This script may or may not work once version 7 comes out, but it saved my butt several times over and should work on 6.5 and below. Just load it up in the script editor and hit run.
http://www.pigeonimpossible.com/Downloads/LM_BSP_Tuner.zip
Oh, and let me know how it works for you. I’ve only tested it myself on a couple of systems and its a little janky, but seems to be pretty stable.
I just found a great blog post by Jason over at CGPipeline that brings up a couple of great issues that I forgot to mention in the lighting episodes:
First some specs:
I’m doing everything in HD res (1920×1080) @23.98 fps
The bit depth of the renders and composites is 16-bit linear.
The final master project will be 10-bit HDCam-SR (4:4:4 RGB) at 23.98PsF
I haven’t decided whether it’ll also be going out to film or not yet.
However, more directly addressing Jason’s post, the bit depth issue is CRUCIAL. If you’re doing any sort of compositing, 16-bit is essential. I’ve personally never found much cause to go to full-float, but I could see that being useful in some situations.
Jason also brings up the alpha issue which is something I’m fighting. I’ve never had much luck rendering straight alphas out of mental ray so I’m using premultiplied. In some cases, this has lead to edge issues. However, I’d also like to mention one other possibility that’s slightly more “out there” but works really well in certain cases.
The idea behind it is this: if you’re setting your antialiasing levels to anything above 0 in mental ray, then you are effectively “baking” data into pixel. That pixel now contains an average of a couple of different rays that each took a long time to render. So, what I’ll sometimes do instead is take the sampling down and increase the frame size. The rule is that for every step you decrease your sampling, you need to double the frame size. So…
If your settings are 0,2 and your frame size is 720×540
That is the same as -1,1 at 1440×1080
and also identical to -2,0 at 2880×2160
(Now, before anyone yells insanity and ships me off to the looney bin, this is a trick that I use very rarely and only when working with SD. However it has gotten me out of a pinch a couple of times.)
So essentially, this technique takes the exact same amount of time to render because the rays its rendering are identical in all three sizes, but it gives you every single ray as its own pixel (if you go as far as -2,0.) The only downside is the file size which gets EXTREMELY bloated. However, it opens up some great possibilities:
You can generate proxies and set your compositor to quarter res. It’s just like working with regular files, but you’ve got your super high-res master.
There are NO edge issues when you render out the full res. None, ever, period. Transparent objects still have their quirks though.
2D post effects such as Depth of field, motion blur, etc. work much better.
You can use the gaussian filtering on the pixels after compositing which solves a couple of other geeky technical issues.
The object IDs work flawlessly. Even if you don’t have object IDs, you can now use every single color rather than just RGB like I mentioned in Podcast #007
Anyways, that’s my crazy nerdy compositor side coming out again, but its just one more trick that can be handy to have in your bag. Of course, I didn’t use that trick on PI because the renders would have come out to 24TB (yes, terabytes) but as technology advances, it starts to become an interesting and more viable way of working. A big thanks again to Jason at CGPipeline.com for inspiring this rant and also the plug.