 Okay, we're back live at NAB. This is theCUBE, siliconangle.com, siliconangle.tv's telecast where we go out to the events and extract the signal from the noise and obviously the theme has been software and we're going to continue that with this segment. I'm John Furrier, the founder of siliconangle.com and I'm joined by my co-host. I'm Dave Vellante of wikibond.org and John, we're going to talk about special effects and the tech behind special effects. Mark Schoenegel is a tech evangelist at Autodesk which is a very interesting topic, right? We all know the special effects. We know them, we love them, can connect to them. Yeah, this continues our theme of software, really taking advantage of all the laser grays and hardware and Mark, studios have had put a lot of demands on hardware and have actually bought a lot of big systems in the past because they got to power their work. Sure, absolutely. So what's going on in the studios? You go in there, you knock on the doors every day, talking to them about their challenges, their opportunities and providing solutions and obviously Autodesk leader in software over the years going back to the PC now through to IN. What's the update on studios and what you guys are doing with them? Oh man, things are exploding. I think a few years ago we came up with 64-bit technology. I remember getting my first dual core with eight gigs of RAM thinking what am I going to do with all this stuff, right? Now I've got 32 cores and 16, 32 gigs of RAM, whatever, and I'm still wanting more. It's a data center and a laptop, right? Yeah, pretty much, it's ridiculous. You go to these big studios and their render farms are getting bigger and bigger and bigger. The demands are getting bigger and bigger and a lot of it's just the effects we're doing. We're doing fluid simulations that just five years ago I would have thought were 20 years away. It's really the multi-core, the multi-core angle that Intel's been able to provide has just been amazing for our industry. Yeah, let's talk about the multi-core angle. That's a good one. The studios we've been talking with, some practitioners like show reel, that if you give them the power, the Ferrari, keep on upgrading the cars, they're going to ride it and so they're going to push it. So what have you seen? You mentioned five years you couldn't do these. What are the things we're doing right now that we couldn't do just a few years ago? And what do you think's coming around the corner? You know what, if you, later this afternoon I'm going to be doing a presentation up here where I'm going to have about 10,000 individual little characters running together, camera pans out and they form the Intel logo. I couldn't have, I mean, three years ago I had 200 characters on my screen and I was blown away. I mean, I had a whole audience, thousands of people at our user group, who in an eye, that I had a couple hundred characters on the screen. Now I've got 10,000 on a laptop. It's just, it's mind-boggling how much we're able to push the envelope just with RAM and core speed and CPUs. And in the studios, what are they doing? That's really impressing you these days. You know, bigger and better. The, I'm really keen on the simulation side lately so just seeing things like fluid simulations where instead of simulating 30,000, 40,000 particles you're seeing tens of millions of particles get simulated and those give you just waves that just look incredible where as before it was a lot of artwork in there, a lot of, you know, hand touching up every frame. Now let the simulation just take effect and you get, you know, tiny little bubbles, all these little, little kind of nuances that you just couldn't, couldn't have before. There just wasn't really the RAM to hold all that. Tell me about how soft image is being used to make the special effects in some of the studios. Can you give some examples? Sure. Well, about five years ago we came up with a technology called ICE and ICE is a fully multi-threaded visual programming language. So, absolutely takes advantage of all the cores that we have in the machines and what that allows you to do is through a very easy node-based interface create crazy special effects, whether it be fire, rain, wind, I guess you don't really, well, you don't really see wind, but snow, how about snow? Smoke? Smoke, smoke, there you go, smoke. You know, all of that, you know, the complexity of that, you know, we look at it in the films and a lot of times we're looking at practical elements, you know, going out in the back lot, I've been at ILM before and I've heard, you know, the lights go off on all the phones that tell you they're about to blow up stuff in the back lot, they don't have to do that anymore. You know, everything is just mostly done now, just through the computers and ICE is being used all over the place, which is great to see. So compare that to, sorry, like ray tracing, what's the difference between? You know, you're, at the end of the day, you are ray tracing those particles, those effects to achieve the look, but what you wind up with is, you know, thousands of little particles describe the motion of the fire and then you put your shaders on top of that to give it the transparency, the look, the believability that those little particles are now actually forming something that looks like fire. So that gets sent off to the ray tracer. And what's really unique about Softimage is that not only do we have a fully multi-threaded programming language to create the particle flow, once we pass it off to the rendering engine, whether it be mental ray, Arnold, render man, whatever, those also take advantage of all those cores. So we're really now in this, finally this nice era where the software, not only a render time is taking advantage of our hardware, which is great. So what about other IO architecture? Is Flash coming into the equation yet or what do you see going on there? Flash as the, like. As SSD, for example. Oh, I'm sorry, I thought you meant flashes. Adobe Flash. I know, that's not relevant. Yes, no, SSD, oh my gosh. Tremendous, you know, I build monstrous cache files. I mentioned that scene with the little guys making the Intel logo. It's about 60 gigs of data, 4,000 frames. Every one of those frames is I scrub the timeline coming off a conventional, especially on my laptop, a little slow, two and a half inch drive. Yeah, that's a big bottleneck. I've got this, you know, crazy cores, lots of RAM, but I'm just chugging as I'm trying to scrub through my timeline with the SSDs, you know, it's real time. I've got one back there that was, I just copied a file onto a 600 megabytes a second. Crazy. Yeah, John, you remember a Polydisk? We were at Oracle Open World and she referred to the spinning disk as the horrible storage stack. So it was a software company, I mean. I mean, you just said, you just basically did the commercial for SSD. I mean, it absolutely changed the game. I mean, there's like Moore's Law supposed to be double. You know, that just like, see ya, right? Like one year, I went from, you know, 30, 40 megabytes a second to 400, 500. You know, that doesn't happen very often. So what's the trend for the studios now? How are you guys impacting their business model? Because obviously with speed, all kinds of disruption is happening. What are the business model changes that you guys are a part of? And in general, what are their business models? How are they changing? You know, it's a mature market now. It's not, you know, you don't go to SIGGRAPH and NAB anymore and, you know, see the latest hair technology, which didn't exist before. You know, pretty much all the tools are out there to deliver effects that we've come to expect, which are, you know, absolutely believable special effects, character animation. So really it's, for us, it's just, it's increasing workflows. It's a lot of training still. The software is getting so complex. You know, I go to a lot of studios and just have training sessions with people who've been using Softimage or Maya or Max for 10 years. They're still hungry for that knowledge. So our job now is to keep developing tools, listen to our customers, obviously. You know, that's a big part of my job is going to the studios, listening to the artist, writing all that down and sending it off to Montreal to get it incorporated into features. So what are some of the special effects that connect with the audience? I mean, some of the movie hunger games and Lord of the Rings. I mean, that's, is that all you guys? I mean. You know what, Autodesk, strangely enough, now that Max, Maya and Softimage are all under one roof, it's pretty much 90% of the effects you see out there. You know, and whether it be a character animation, whether it be, you know, I consider effects hair, you know, hair simulation. That's a crazy effect. What are some of your favorites? Oh man. You know, Pixar always blows me away. They're water stuff. Pixar always has those shorts at the beginning of their films, you know? And it's always like, for me, that's the best part of the film because that's them just showing off, you know? So load up any Pixar movie and those shorts at the beginning just kill me, just, they're always pushing the envelope with those shorts. So yeah, I think that's my favorite. And then, jeez, anything ILM does still is, I'm just having strange thoughts in my head of this cool stuff blowing up, you know? I like explosions. You like explosions, yeah. The phones don't go off but the self tower might get blown up. No, but seriously, now let's talk about democracy and media. So the film industry obviously pushing the envelope with animation. Do you see the special effects in animation world getting more integrated into just traditional broadcasting and corporations? Because they're kind of coming late to the party but yet with the price points and the ability talent wise in the marketplace for developers, you know, there's a whole new renaissance going on here. Well, we need to make it a little bit easier. You know, in the past 3D was very complex. You didn't quite need a PhD but you needed something to get some good effects done. And a lot of the broadcasters, you know, they don't have time to go learn how to make a crazy particle effect. They want presets. They want to be able to load up a simple piece of geometry. You know, Autodesk has delivered some really cool technologies like Shutterfly where you can take pictures of people or objects. And people used to always ask me, can I just take a picture of that computer and have it be in 3D? No, that's ridiculous. I mean, you gotta model it. Of course you gotta model it. But now through some of the technologies we have, I can take a picture of that from, you know, a few different angles, build myself a cool model of it and literally drag and drop some presets to as I like to blow things up, blow that thing up. You know, the geometry will fracture. Another preset can cause the geometry to look like it's on fire. Then another preset makes it smoke. So this has been the big innovation on ICE. So ICE really enables this visual programming. Sure. And that's really the key value for these developers, right? Yeah, for sure. And it allows you to easily prototype. It is a true programming language. So you're literally programming the effects, but we've really simplified it in a way to where, you know, something that used to take 10,000 lines of code, I can do it in 20 nodes. And that's, you know, we're going to see ICE technology in some of the other Autodesk products. It is such an innovative, cool development that we're going to see that in my other sale there. So talk about the demo you're going to do. Is this going to be the Intel characters that you're going to do the logo? That's what's going to come in the afternoon? Yeah, so when I knew I was going to be up here, I decided to muck around the Intel logo. So first I made it explode and I thought, you know, they probably don't want to see the logo explode. All right. So then I dropped it and made it. Branding's going, that's not branding. That's pretty much what happened. So like, all right, well, maybe I'll just make a bunch of little characters all come running together and they all stop to form the Intel logo. And that actually brought some fairly unique challenges. So, you know, I have 10,000 characters all running from different angles and then eventually they stop and then the guy behind them stop. Well, the ones that need to get over here, they have to run around them. So all that technology of pathfinding and everything is in there. But when you're dealing with 100 characters, it's easy for everybody to nicely get to a point. But when you've got 10,000 guys and you want to pull the camera back and see them kind of take the shape of those letters, they're not humans. They need to have some intelligence. You know, you don't want them stopping and guys bumping into them. So just spend a little bit of time using some of the new crowd technology we have in SOF 2013. And really, I mean, just in a couple hours I had something that, you know, before would have taken, I don't want to say weeks or months because you probably could have found a shortcut. But- Yeah, it's expensive too. I mean, all the development time. Absolutely. So you had to essentially re-architect to take advantage of all these new cores, the SSD, talk about that a little bit. And how far will this take you? When we started developing SOF Dimage, the next generation of SOF Dimage about 12 years ago, there were no multi-core machines. You know, we were on a single core CPUs. So, you know, we developed, we created the software, awesome stuff. And then a few years later, I went to WinHack and here's Intel showing me this Prescott chip that had two cores in it. Oh geez, this is gonna go somewhere. And then they're telling me, oh, we're gonna have eight cores in a few years. I don't believe it. Well, here we are. So we actually had to go back and literally re-architect the entire core of SOF Dimage to take advantage of those extra CPUs. And now it scales almost linearly, which is awesome. I've got the new E5 over there, eight cores per socket, 16 threaded out to 32. And man, the stuff is just flying on there. And it's really because we made the investment a while ago to re-architect the entire core of the software. And that's not trivial, right? From a software developer standpoint. No, no, that's a big decision. That was five years that, you know, I mean, it's paid off in spades. The speed at which I'm able to work. And I keep, my home machine, I keep all the old versions loaded up. And it's really funny to load up something that my old demo scenes that I used to think were just crazy fast and like, geez, even on a new machine that's slow because it's all on the old single core. Load the same scene up on a later build and stuff is just flying. So do you feel like that's a core competence of Autodesk or are you somewhat because you've been around for a while that handcuffs you a bit, the innovator's dilemma? Talk about that a little bit. You know, that's a good question. Not being so much on the development side. You know, I don't think it's really a handcuff. It's been, you know, I mean, that team, the team that built that core, they once they gained all that knowledge of, you know, programming 3D multi-threaded is not trivial. Took a lot of iterations of writing the core. I think at one point we just scrapped it and started all over. And now those guys are working on those, you know, the same team I've been moving on to other places to implement that in other Autodesk packages. So definitely, you know, good lessons learned. And then once you get on the curve, like you said, scales linearly. You're excited and you don't want to have to do that every couple of years. Exactly. The uptake has been fantastic. Everyone we talked to uses your stuff. So congratulations. It's out of their positions. Great innovation, Mark. Thanks for coming on theCUBE. My pleasure. Thanks guys. Good to see you. Great interview, technology innovation enabled by great hardware and amazing software. Great bet paid off. Come back for the demo this afternoon. Yes, for sure. Demo this afternoon. I'm blowing stuff up. Blowing stuff up. I love the Intel logo blowing up, but I'd love to see that one. Okay, we'll be right back from theCUBE here at NAB day two. We'll be right back.