Cloud Computing Democratizes Digital Animation 167
kenekaplan writes "John McNeil is the chief creative officer and founder of a digital arts and communication company based in Berkeley, CA. After turning to Amazon's Elastic Cloud Computing service for the first time to finish animation under tight deadline, he was impressed by how it would let him compete with bigger studios. He said, 'Cloud computing is the first truly democratic, accessible technology that potentially gives everyone a supercomputer...it's a game changer. I could never compete or be able to deliver something at the level of a Pixar or a Disney, given what I have at my disposal inside the walls of the studio,' McNeil said. 'But if I factor in the cloud, all of a sudden I can go there. And then the limitations of whether or not I can deliver something great will be on my own talent and the talent of the people that are part of the studio.'"
gives everyone a supercomputer... right (Score:5, Insightful)
"And then the limitations of whether or not I can deliver something great will be on my own talent and the talent of the people that are part of the studio." ... and also how much money I can put. Using a massive computing power on a cloud requires a lot of money.
Re:gives everyone a supercomputer... right (Score:4, Interesting)
Queue the "but it's cheaper than owning a render farm" comments!
But, hey! It actually is cheaper when you can't utilize a render farm as efficiently as big studios can.
I think this is a prime example of rent-a-hpc done right.
Re:gives everyone a supercomputer... right (Score:5, Interesting)
Re: (Score:3, Informative)
This is a different, and more generalized, version of the renderfarms-for-rent that were available in any area where there was a strong animation industry. Some of us were even experimenting with internet delivery of job info, though initial loading of images and models required shipping a hard drive or three.
SGI had something called (iirc) Drums or something like that back in 90s to attempt this very thing. A bit a head of the market, but still a neat project.
The major problem with building your own re
Re: (Score:3)
Absolutely. The problem is, people make it sound like it's going to let talented Joe Blow in his basement create the next Toy Story or cure cancer. It's not. "Cloud computing" is just servers for rent instead of purchase. For a lot of the suggested projects (not rendering, but anything scientific) you could have gotten access to a large cluster by submitting a worthy proposal. Now you can just fork over cash for your project, worthy or not.
Re: (Score:2)
No, I think queue makes more sense, unless of course you really want to hear those comments to which the GP is referring.
Re:gives everyone a supercomputer... right (Score:4, Insightful)
Really? You can do everything that say a studio can do? I am going to sound very cynical, but this is a typical techy answer to why they are not as popular as Pixar or Disney.
1) So you want to purchase computing time? To get something like Pixar you are going to digging deep into your pockets. Not just a little deep, but very deep. I am giving a talk at a developer conference on the merits and usability of the Amazon cloud. Granted it is an enabler, not going to debate that. But to say that it will put you on equal computing power as Pixar is a little navie.
2) Yes Pixar and Disney have oodles of techy's all running around using the computer. Lest it be known that Pixar and Disney employ's a whole bunch of writers, producers, musicians, and so on. The reason why CGI is so cool and neat is because the people (read actors, writers, producers) are professionals and they know how to BUILD a STORY! I have seen short movies done with Blender and let me tell you while it looks good, for the most part the movies are pure and utter rubbish!
I am not saying it can't be done, I am saying that just because you have the cloud does not mean you actually have a movie worthy to be compared to Pixar or Disney.
Re: (Score:2)
Blender movies are more of tech demos. I agree that their stort often stinks, on the other hand it's very intresting to see how the whole creative process works. They document their progress very thoroughly. Also, considering that their team usually works with alpha* quality software, I'm impressed that they can get anything done.
* Sintel for example was done with Blender 2.5 Alpha 0-2. That thing crashed me every three minutes.
Re:gives everyone a supercomputer... right (Score:5, Informative)
Your point 2 is exactly what the guy in the article said, quoted by GP. Except he said it more succintly.
Let me re-quote: "And then the limitations of whether or not I can deliver something great will be on my own talent and the talent of the people that are part of the studio."
Re:gives everyone a supercomputer... right (Score:4, Insightful)
Let me clarify my point... Let me put it in the context from my world which is trading:
If I only had the computers like Goldman Sachs I could make money like Goldman Sachs. With cloud computing I am now able with a couple of good traders to make money like Goldman Sachs.
The fallacy of the argument is that Goldman Sachs needs the computers to make money. The reality is that Goldman Sachs needs the traders not the computers. For Goldman Sachs the computers could be hamsters running in a wheel, as the important key are the traders.
Thus by him saying that now with cloud computing he is on equal plane with Pixar is completely missing the point that it has never been about tech. It has been about how story writers use tech! Thus cloud computing will neither enable or disable you. Anybody can create a Pixar type story given enough computing time. Remember Pixar needs to pump out a movie in a year. Who is to say that you cannot create a short story in a year using plain vanilla computers.
My case in point: http://www.youtube.com/watch?v=xOgYOD5S8gk [youtube.com] This guy was a sensation because what mattered is that he had good enough animation with an incredibly good story. Or how about the numa numa guy? http://www.youtube.com/watch?v=KmtzQCSh6xk [youtube.com] It does not get any cheaper and simpler than that! Yet this guy has 18 million views and featured on South Park. And while I am on this thread, how much real computing power is needed for South Park?
Understand my point? Understand why the GP is incorrect?
Re: (Score:3)
how much real computing power is needed for South Park?
Actually, South Park is made with Maya (except for the pilot short which they did in Aftereffects). They've likened the process to using a bulldozer to build a sandcastle.
Re: (Score:2)
Unfortunately the guy in the article conveniently left out the GPs point #1 (which is the important one). You still need money.
Re: (Score:3, Insightful)
I agree completely. Your first point illustrates a common error among amateur artists: Photographers always say that they'd take better pictures with a better camera, or put another more illustrative way: "If I only had Hemingway's typewriter I could write like him."
Your second point is the more important. Any production is about the story, not the technical expertise. To mirror your comment about Blender, I've seen crap done with it too, but on the other hand, I've seen great work done with it. And w
Re: (Score:2)
Wow hiding behind anonymous, but spoken like somebody who like me has gone through it. I learned by the market. I thought, "hey I am a geek and can write algos thus I should be able to make money regardless..." HA right... The tech is secondary!
Re: (Score:3)
Wow hiding behind anonymous, but spoken like somebody who like me has gone through it. I learned by the market. I thought, "hey I am a geek and can write algos thus I should be able to make money regardless..." HA right... The tech is secondary!
With movies, you're making entertainment. You don't need flash-bang effects to provide entertainment. You do want a good story, told well. (Now, if someone could just persuade the blockbuster-pushers in Hollywood that this was deeply true, we'd be better off overall.)
Re: (Score:2)
The overwhelming popularity of low budget, grainy, handheld independent films testifies to this effect.
Re: (Score:2)
Flamebait? Really?
Now, THIS is flamebait you whiny little protein deprived pond scum progeny of a mutant Monsanto labs reject.
See the difference?
Let's say I work for Disney or Pixar (Score:4, Insightful)
If I'm afraid you're going to "compete or be able to deliver something at the level of a Pixar or a Disney", couldn't I just get your ISP to block you off from the cloud because I don't feel you're doing enough to prevent Pixar/Disney intellectual property from being incorporated into your work?
Re: (Score:2)
That's right, they could create something stealing valuable Disney intellectual property like Cinderalla, or Snow White, or the _Hunchback of Notre Dame_, or something resembling the Buster Keaton short film _Steamboat Bill Jr._.
Re: (Score:2)
>and also how much money I can put
That banal thing you said normally goes without saying.
You are missing the point. Historically
1: fixed grade scale of computer power avaliable with huge gaps in performance that one cannot break, you either work with your desktop or massive mainframe.
2: linux clusters: you can add power as you get money, but you also have to be a computer geek to maintain a cluster (not only animator and a user of your rendering program)
3: clouds: you can order as much power as you have
Re: (Score:3)
The critical thing missing for people to decide is numbers. Depending on the numbers, there may be particular windows where it's cheaper to use the cloud than it is to use your own hardware. Or, it may be the case that it's always cheaper (although that's incredibly unlikely). It also may be the case that it's never cheaper to use the cloud (also not very likely because of the special cases where you need a truly massive amount of computation done in a very short time (like rendering a two hour video to mee
Re: (Score:2)
It also requires you to upload a massive amount of art data and downloading a movie in full resolution, with minimum compression.
Or you can 'physically' send them (I guess Amazon has a service like that)
Re: (Score:2)
But you dont need an investment. This can make the difference between doing and financing a first few projects yourself, as you need these instead of investing a lot on money, which sits in the corner no matter if you need it or not.
It wasn't that expensive (Score:2)
Re: (Score:2)
Its not that expensive.
I have a demo that I run in Windows Azure where I spin up 16 servers to render a 1500 frame animation in HD quality. I demo it an conferances and user group meetings, it fits nicely into a one-hour session. It takes about 10 minutes to provision the environment and around 20 minutes to render the animation.
The total cost for compute time, storage and data transfer is less than $2.
potentially gives everyone a supercomputer (Score:1)
No, it doesn't.
It potentially gives everyone who has enough money a supercomputer, meaning nothing really changes.
I doubt there is much difference between running your own computer or sharing a supercomputer that is roughly ten thousand times as powerful with at least ten thousand people.
Not to mention the bottlenecks caused by the internet connection you need to tell it what to do, and get what it has done back from it.
Re: (Score:2)
If you share a computer ten thousand times as powerful with ten thousand other people, *they will not be using it all the time.* Most processing power is unused most of the time. But the capital investment in a supercomputer is pretty substantial.
Of course, most supercomputers involved time sharing long before cloud computing came along.
Re: (Score:2)
Yes, but then you have to do some math to figure out if you're getting a good deal. It's a similar calculation to the ones done to decide how high the jackpot needs to rise to make buying a ticket a worthwhile endeavor. The problem is ultimately coordinating usage.
Re:potentially gives everyone a supercomputer (Score:4, Insightful)
Rendering projects tend to be quite small in their models form. When utilizing a render farm you really don't have that much I/O after the initial project upload. granted a big project (a movie) might have gigabytes of texture data but still nothing that can't be uploaded in half a day or so. After that it's all internal I/O (which in ec2 is very fast), the control server telling the nodes what to render and providing render assets and the nodes just returning rendered products and requesting new tasks. After the task is complete you just have to download the rendered products and you are good for post processing :-)
Also, I don't think your argument stands. It is very affordable if you are in that line of busyness. Hell if you are a hobbyist and have a couple hundred € to burn, you can run a mini jaguar for the weekend, just to get a feel for it.
Re: (Score:3)
Granted a big project (a movie) might have gigabytes of texture data but still nothing that can't be uploaded in half a day or so.
I measured this a couple of years ago. At the time, you could ship data to S3 or EC2 at about 10GB/hour provided you're not saturating your own network or doing it at a very busy time. For a truly large amount of data (1TB up) I suppose you'd just FedEx a hard disk...
Re: (Score:2)
I doubt there is much difference between running your own computer or sharing a supercomputer that is roughly ten thousand times as powerful with at least ten thousand people. Not to mention the bottlenecks caused by the internet connection you need to tell it what to do, and get what it has done back from it.
Err. Not sure you grasp this.
Computer A and Computer B
Computer B is 10,000 times more powerful but is shared with 10,000 people
The key difference is timing and time. If all 10,000 people want to do the same thing that they do on their A computers on computer B at the same time then I agree with you. However that is not real life. If the distribution in time of tasks from the 10,000 users is spread out then my task will run up to 10,000 times faster on computer B than computer A.
If I could use compute
Re: (Score:2)
Is there an security involved in the Amazon cloud?
As much as you want to put in the VM image. If you're not diligent with protecting your nuclear simulation code, it's your fault, not mine or Amazon's.
Re: (Score:2)
That's true of pretty much everything. If you're in town for a day you take cabs. If you're there for two weeks it's probably be cheaper to rent a car. If you're there for a year you might buy one.
Re: (Score:2)
Rendering animations is a pretty ideal case. Both the input and the output are relatively small, and the processing power required to get good results is immense.
You could not be farther from truth. The actual problem in all serious CGI production efforts (Final Fantasy, LoTR, Pixar movies etc.) is not the raw processing power, but the sheer amount of data processed. In 2001, the numbers were something like 1-15 hours per frame, with an average of 2-3 hours or so, and with 1-2 GB of scene data (RIB files per animation frame) and typically several GB of texture data. (I have to look up the original post - a response from a pro render wrangler to an announcement by a
Re: (Score:3)
These days it's even bigger. In 2007, in Transformers, rendering a single frame with one bot in it too
Re: (Score:3)
These days it's even bigger. In 2007, in Transformers, rendering a single frame with one bot in it took hours. Rendering the whole gathering at the observatory took around 25 hours per frame. The total models only took around 2TB or so of data.
These numbers would mean that even if you leased just one rendering node (capable of rendering a frame per day - not very useful), you would still need to be pushing 200 Mb/s of scene data into it on average - over the Internet. Now perhaps your animation studio has such an external backbone access, but multiply it by two thousands if you're doing a movie and not just a couple of stills and we're talking about +100Gb/s lines from your company to the Amazon service. Yummy...
Qa'pla!
Re: (Score:2)
Qa'pla!
(That should have been Qapla', of course...)
Re: (Score:2)
Now perhaps your animation studio has such an external backbone access, but multiply it by two thousands if you're doing a movie and not just a couple of stills and we're talking about +100Gb/s lines from your company to the Amazon service.
The workaround is to do everything in Amazon, including the streaming/distribution of the "final product" to your viewers.
Of course there's this too: http://aws.amazon.com/importexport/ [amazon.com]
So you could still ship TBs of stuff to/fro your studio whether for further processing or for backups.
Re: (Score:2)
The workaround is to do everything in Amazon, including the streaming/distribution of the "final product" to your viewers.
How will, say, Maya running on EC2 display its viewports on your remote (wr/t Amazon) machine? Does EC2 support X11+OpenGL tunnelling well?
Re: (Score:2)
I have to look up the original post - a response from a pro render wrangler to an announcement by a wannabe-enterpreneur who naively thought that the idea of an online rendering shop with scene transfer over the Internet is technicallty viable, but was beaten by the facts
Yeah, that naive wannabe-enterpreneur was me, funny to see this comment today. Yes, it was hopeless project. Beaten by the facts - not really. Before that question I just didn't have enough informations.
I only wonder what pricing does EC2 have for dozens of terabits of data in total between all the nodes you just hired as a rendering farm...
Inside one availability zone - $0. Inside one region - $0.1/GB. If you don't mind that your farm will be unavailable for about one hour a year, your transfer costs can be $0.
If you need a render farm only for a month, ec2 will be cost effective. If you need render farm all the time - probably your own hardw
This project does NOT (Score:2)
Re: (Score:2)
Well, as I said in a comment above, it is doable. It probably will get expensive (but when you involve render farms what isn't) but you can certainly do it, ec2 is not a toy infrastructure anymore. What I really want to see though is how it would compare to one of those comercial rent-a-farms like www.renderrocket.com
Re: (Score:2)
It probably will get expensive (but when you involve render farms what isn't) but you can certainly do it...
Things don't get more expensive. Cost reduces as usage increases.
Re: (Score:2)
That is very, relative. Even inaccurate..
Unit costs usually decrease as unit production increases and even that is not always true (Oil industry).
Re: (Score:2)
Clearly they're going to run these super computers on electricity they produce by burning oil that they themselves drill for. Quite self sufficient.
Re: (Score:2)
Had this been an eponymous poster I would have been inclined to answer, even though the relation is quite apparent.
Democracy for those who can pay for it... (Score:3)
Re: (Score:3)
Of if you use something like Kickstarter to get some money upfront.
Hmph. (Score:2)
And then the limitations of whether or not I can deliver something great will be on my own talent and the talent of the people that are part of the studio.
Oh yeah, I suppose, there'll be some cash needed to pay for all that compute time to render it like the big boys. Great big stinking wads of cash. But yeah, it's totally levelled the playing field now.
*rolls eyes*
Re:Hmph. (Score:5, Insightful)
Yeah, but they don't need to maintain the systems for all the time they don't use them, which the big boys can afford to do (usually because, with multiple projects, they'll be able to have much less downtime).
Ex: You need about "100 units" of CPU time per year. A computer that does this costs $500. Now, lets say, you have about a month from sufficient data to start, to the deadline. Now, you need a computer that can provide "1200 units" per year. This will probably be closer to $6000. And part of that money goes towards having it for 11 months where you don't need it. You might pay more ($1000 maybe?) to get the job done in a month, than with a computer that, given a year, could do it, but you get it done on time, which is probably worth that extra $500.
It may not be ideal for everyone (or even most), but for a smaller player to move up in the world, when they can't afford to have enough projects simultaneously running, that would make maintaining their own desired system financially viable, then this kind of CPU timesharing, is not a bad idea.
Re: (Score:2)
Dear lord, I have a sentence in there that needs to be shot. It looks like a run-on that just WON'T DIE.
Sorry :-(
Re: (Score:2)
Don't worry, we're geeks. Content is more important to us than presentation, as long as it's understandable, and your content is on the spot.
Re: (Score:2)
Don't worry, we're geeks. Content is more important to us than presentation, as long as it's understandable
Ahem! You misspelled "we're geeks. We constantly correct people for unimportant spelling and grammar errors even when it would be impossible for us to misunderstand the text".
Re: (Score:2)
Eye konw waht yuo maen.
What a load... (Score:2, Interesting)
You could hire computing cycles for a long time, there have been companies hiring out temporary server hosting, for however short or long as you want for decades... and of course for the really old, hiring a certain amount of performance on a larger system is exactly what mainframes were about.
Of course, it has become easier but that is because computing has come down into the general market of the last few decades. More people can now afford computing in general including buying access time on mainframes,
Re:What a load... (Score:5, Insightful)
Let's not forget, you can render a really nice smooth teapot for peanuts, but you can't tell a good story, choreograph a chaotic rescue scene, or draw the expression on the protagonist's face when a little old lady kicks him in the nuts. You also have no chance of getting Jack Black or Angelina Jolie as voice actors.
Pixar is not just a render farm. It's also a studio. The South Park guys showed you can still beat a studio if you are willing to target a new demographic (people who liked The Simpsons, and want sex jokes as well as fart jokes) and have talented actors and writers, but it's not easy.
Re: (Score:2)
http://worldfilm.about.com/od/indieactors/American_Independent_Actors.htm [about.com]
Having a project good enough they'd want to join in would be the hard part. Also the opportunity to be able to tell them about it would take effort. Not impossible. i'd say the Blender project have enough eye candy to be able to get a "star" on board if they loved the story and character enough.
http://www.blender.org/features-gallery/blender-open-projects/
Re: (Score:2)
You don't even need actual actors, just voice actors, which charge a lot less, at least around here.
Re: (Score:2)
That said, I wonder how you would you go about getting a decent music score made. I guess an orchestra costs a fair bit. -I've no idea about anything to do with music recording. Crowdsourcing all those instruments and balancing recording quality would be a project and then some.
Re: (Score:2)
I guess an orchestra costs a fair bit. -I've no idea about anything to do with music recording.
I'd say going for a full out orchestra would _not_ be the first option. In fact, there were some interviews talking about "Warhammer 40K : Space Marine" and how they were really excited to actually be using a real orchestra at all for recording the soundtrack. Of course good audio gear to make a synthetic orchestra also looks expensive, but I would assume quite a bit less investment than the real thing.
Re: (Score:2)
It's not just the actors. A Pixar/Sony Pix Animation/Dreamworks film can have 500 people in the credits list; Pixar itself, for example, has about 700 full-time employees, many of the artists who do the shows are freelance.
In the credits, 10-20 of the 500 are actors, and maybe a dozen are sysadmins and maintenance of the farm. If you want to make a film, you've gotta hire the rest. A server farm won't do character design for you, or animate, or do sound effects, or record the voices, or write the music,
Re: (Score:2)
Let's not forget, you can render a really nice smooth teapot for peanuts, but you can't tell a good story
Wait.. what?! What does rendering have to do with telling a good story?
Re: (Score:2)
I think you greatly underestimate the value those actors bring. They are not just voices added after the animation is done. The personalities of the actors is used to make the characters themselves. Woody from Toy Story IS Tom Hanks. Buzz IS Tim Allen. Would Alladin be the same movie without Robin Williams as the genie?
Re: (Score:2)
No, it would probably be better. I'm sick of seeing and hearing Robin Williams every time I turn around. All he does is "Robin Williams". Every role. Minor variations.
The man's become a distraction.
Re: (Score:2)
All he does is "Robin Williams". Every role. Minor variations.
Not true. Some exceptions based only on movies I've seen:
And those are just the ones I've seen and recognize from IMDB. Sure, he's best known for his slapstick, but he's a good actor with a wide range. Even a movie like "The Fisher King ", where's he's doing the "Robin Williams" thing, is a well-developed character made to fit the movie.
Re: (Score:2)
They do it because it's really easy work. They don't need make-up, they don't need many shots, they don't need to learn the lines, the sound guys can fix everything up, and they just sit in a studio with a script ham acting into a microphone.
Good voice actors won't need as much post-production, and can do more to develop a character, but that's more of an issue in long-running cartoons.
Re: (Score:2)
But he can, if he has the talent, do a pretty good rough draft. Good enough to get some attention, and funding for polishing and render farm time for a full quality 'print'.
If he has the creative bits down, there is plenty of entertainment value in watching the rough draft.
As long as this isn't all shut down because you need to go back to the 19th century for music that isn't copyrighted out of anyone-but-big-content's reach, and YouTube hasn't been shut down.
If your definition of democritize is to encheapen, (Score:1)
Then I agree with you?
SOPA render farms? (Score:1)
Re: (Score:2)
And here is the real reason for SOPA/PIPA (Score:2)
This democratising of computing ability makes content creation within the ability of anyone with the (extreme) talent. This explosion of content and weakening of the grip of big media is what is being fought.
And it will only get cheaper. No wonder they are fighting it. Ironic that they are using a provision that was originally designed to encourage creation!
I am wondering more and more whether the concept of copyright is suitable these days.
Bandwidth ? (Score:5, Informative)
Does the CEO realizes that he's trading CPU's limitation against bandwidth's limitation ?
Generating a picture in full HD requires 1920 x 1080 = 2,073,600 pixels.
But for a movie, the resolution is 4096 x 3112 = 12,746,752 pixels
http://en.wikipedia.org/wiki/4K_resolution [wikipedia.org]
This gives 36 megabytes per picture.
Now, you have to create 25 pictures for one second.
You get 5400 megabytes for every minute of movie.
It may be faster to compute digital animation, but you still have a large IO problem, both in storage and in preserving the data (you may lose pixels when downloading the files) !
It's similar to outsourcing tasks: it's a short-term solution for larger problems.
Re: (Score:2)
Does the CEO realizes that he's trading CPU's limitation against bandwidth's limitation ?
Does it matter? Disks are cheaper in many ways than computation, and digitally rendering a movie will use a lot of that. Not having to find a building to hold a large datacenter which you're not using all of the time anyway... that will surely save masses even with all the additional networking.
Large downloads could also be handled by shipping physical disks. Major cloud vendors will do that.
Re:Bandwidth ? (Score:5, Insightful)
Re: (Score:2)
100mbps connection isn't that expensive if you're in the right area. On that connection (assuming you have the computing power on the other end) you could have near real time processing even at the bit rates you're talking about. This is a really good and interesting solution for their problem.
Re: (Score:2)
Re: (Score:2)
First of all, many assumptions you make are or may be incorrect. It is not certain that the image is 8-bit per channel (say, jpeg) as it can go up at least to 32-bit per channel and there can be more channels than 4 (RGB + alpha), such as depth passes. Which can also be of different bit-sizes or even different files. There are also lossy and lossless compressions schemes in almost all practical file formats (PNG & EXR to name 2). But yes, it's a lot of data.
Assume additionally that an output for a serious effect shot or animation shot has, say, a dozen layers (separated by scene depth), with a single layer having multiple passes for fast adjustment during composition (beauty pass, shadow pass, specular pass, mirror pass...). Basically, increase your number by 1.5-2 orders of magnitude. You're finally approaching the right numbers...
Re: (Score:2)
Yes it really is a game changer (Score:5, Interesting)
Studio Pyxis (www.studiopyxis.com) is a Burbank based production company that is taking full advantage of cloud computing, both for our own productions and for sharing with clients. The company is a new model of production company leveraging years of development in real time technology such as the Virtual Production process used on Avatar which involve real time graphics and visualization so that directors can shoot visual effects interactively as if they are really happening in front of you, and cloud computing that can complete the photorealistic renders on the back end in record time.
One feature of cloud computing that is often over looked though in these production discussions is the breadth vs depth computing model. It's obvious that it's a value to have a massive cold room that you don't have to buy up front, but the real advantage comes when it costs exactly the same thing to run 1000 cpus for 1 hour as it does to run 10 cpus for 100 hours.
Visual effects and animation production is all about revisions. It's a huge win to have your full renders back sooner. Being able to run every frame of a shot at once regardless of how many frames you have means that you have the entire shot in the time it takes the longest frame to render. This has never been possible before. Production has always wanted a dynamically scalable solution but as always had to contend with some fixed capacity. Granted EC2 has a fixed capacity as well, but it is so much more massive then a typical production facility as to be a non issue.
As for what some commenters are saying about bandwidth issues it is true it's a factor, and this is why it's not a turn key solution for the average small company. We've spend a fair amount of time creating an infrastructure that mirrors assets in the cloud, renders and composite locally to the cloud, then generates compressed images and movie files for download at review. Only when we approach the completion of a shot do we download actual exr, or dpx data. But we do make our infrastructure available to other companies to help them be more turn key.
Another aspect that more then democratizing cg production actually gives an advantage to the smaller facility are the limitations that larger facilities working on mainstream studio pictures have such as MPAA rules about keeping film assets off the internet and/or on physically disconnected machines. Whereas small facilities like ours can be satisfied with a VPN connection to Amazon, larger facilities are often legally obliged not to.
The one area that still needs to be solved to truly make this work for everyone is for the software companies to start offering the same type of pay as you go licensing so that we can more easily use the professional tools. It would be relatively easy for a company like Pixar to offer a RenderMan license server that one could connect to over the internet or even EC2 based that would monitor your hourly usage. Are you listening Pixar?
Re: (Score:2)
have you actually looked at the numbers? AWS, for instance, has a phenomenally high profit margin (and PhB's have the mistaken impression that running your own farm is expensive.) it's not some kind of close thing (oh, they get power 10% cheaper, or their machinerooms run at a PUE of 1.15 whereas ours are 1.4) - cloud providers are charging O(10x) more per compute-unit than cost.
Re: (Score:3)
We've been living with the numbers for the last three years. In other industries I'm sure the cost of cloud computing quickly eclipses the cost of owned computing capacity, but in visual effects and animation production you have to remember one important factor; your computers are idle most of the time.
Re: (Score:2)
We've been living with the numbers for the last three years. In other industries I'm sure the cost of cloud computing quickly eclipses the cost of owned computing capacity, but in visual effects and animation production you have to remember one important factor; your computers are idle most of the time.
It's the same in small engineering firms. Most of the time they're physically building their current product, not doing simulations to design the next one. Idle computers are a total waste in such a situation. Yes, they could sell on the spare cycles but that would also require them to have a full time sysadmin, someone spending time on sales of compute time, and so on; the tail would wag the dog. Far better to buy it in when needed. (What's more, I know that at least one of the software vendors in the area
Re: (Score:2)
how can that be? you really only have one project going at a time, and the whole project becomes renderable at once?
Re: (Score:2)
CG production is all about burst computing. Even with multiple productions going you don't end up with even capacity utilization. You're constantly going from near zero need for the render farm, to wishing you had 20 times the capacity. Just look at an individual artists work flow:
1) Playing video games because the job hasn't started yet.
2) Designing and planning a shot.
3) Render testing (local machine).
4) Production of a version of the shot.
5) Render (artist back to the video games).
6) Review with super
Re: (Score:2)
I'm sorry I don't have specific details, but having worked in major studios for years I can say they have many strict rules outlined by the MPAA on how assets can be managed. For example final editorial boxes are not supped to be connected to the internet at all. VFX production is usually a little more lax, but the facilities do often have to sign lengthy agreements making them liable for the studio assets. If something where to make it into the wiled they have have huge liabilities to the studio. Small
EC2 is great as a renderfarm!! (Score:4, Informative)
EC2 let me render in 5 hours what would've taken my own computer about 110 hours to render. The cost came out at about $.06 per core hour. The commercial cloud renderfarms charge about $0.75 core hour for comparison.
Re: (Score:3)
Forget the cost per core hour. How much did it cost you to render your project in 5 hours instead of 5 days? Was it worth it? Was there a reason why 5 days would have been too long?
Render farms predate "cloud" computing (Score:3)
Commercial render farms have been around for years, long before "cloud" computing. Search for "render farm" to find some of them. They compete on price, so the pricing is good. The concurrency generally consists of running one frame on each machine, so the intercommunication during rendering is zero. Pixar was doing this by 1995, using a set of shell scripts called "Ringmaster" to push the data around. If the CEO of an animation house just discovered this, he's way behind.
Rendering isn't the bottleneck on cinema projects. People are. Look at the full credits at the end of any modern animated production, and watch as a thousand names scroll by. There's an army of people drawing background objects, landscapes, crowds, and fine detail. That's where the cost goes.
It hasn't improved much in the last decade, either. A decade ago I knew a director who'd done some feature films with mixed animation/live action, and he was hoping the technology would get the cost down, so he could do a feature for $20 million instead of $80 million. A decade later, budgets for A pictures are up, not down. "Tangled" came in at $260 million. Which is why what gets green-lighted is usually a known franchise.
If you're willing to drop to video game levels of quality, animation can be really cheap. See Next Media Animation [www.nma.tv]. Fastest production house in the world.
Re: (Score:2)
Actually using the cloud is significantly different from using render farms. The overhead of using a render farm is very high. You have to package your data specifically for the farm, which includes methodically fixing all your resource paths, you have to setup an agreement with the provider, and the rates are typically exceptionally high in many cases more then 100x the price of equivalent EC2 instances.
Whereas on Amazon you control the boot image, have exactly the software you need installed and can cre
Re: (Score:2)
Based on your posts I see I am not an idiot for salivating about EC2. I am actively trying to figure out jobs I can run on it, just because its so cool. Thanks for your perspective. Sometimes the slashdot comments really, really rock.
andy
Redefining "democratic" (Score:2)
So I guess "democratic" means "relatively cheap and easy to use" now?
Duly noted.
Re: (Score:1)
Also, CPUs are so fast that Pixar is mostly IO-bound, which is something the cloud won't solve for you.
Re: (Score:2)
CA is Canada (http://countrycode.org)
I didn't know Berkeley was located in Canada, but apparently it's somewhere in Ontario (http://maps.google.nl/maps?q=berkeley+canada&hl=nl&sll=44.366667,-80.716667&sspn=0.05283,0.093126&vpsrc=0&gl=nl&hnear=Berkeley,+Grey+County,+Ontario,+Canada&t=h&z=15).
Re: (Score:2)
Not only are Berkeley in Ontario in Canada (Berkeley in CA) and Berkeley in California (Berkeley, CA) both places but not only is there an Ontario in Canada (Ontario, CA) but there's one in California, too (Ontario, CA).
Re:Why is Slashdot not blacked out? (Score:5, Informative)
I think the SOPA blackout is useful because it helps inform the general public who is unaware of the problems and ramifications of SOPA and it's three vile siblings.
I doubt there is anyone on slashdot who is unaware of the ramifications, even those of use who support some copyright stuff, and who are against piracy, are aware that SOPA is a dangerous piece of legislation that will only harm society, and not help any artist.
Re: (Score:2)
Posting to revert bad mod.
Re: (Score:2)
Don't worry, they'll post the MPAA and RIAA sites on their homepage
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
That's the thing where dollars vote, right?