Is This a Lie?


Recently, a hobbyist has posted the following video and claimed that he has made it after a serious study of 3D for about two and a half years. If his claim is not true, he would be guilty of robbing art/film schools of potential enrollees and possibly of robbing VFX/CG firms of potential clients as well. Here’s a recent interview with him. Do you think he’s lying? What do you think art/film schools should do about such claims?

the short:


What reason would he have to lie? I think if a person works hard they can do anything they put their mind to. I personally don’t think art school can teach you how to push yourself in becoming better. They can only show you how to work the software and teach you theory,the doing is all up to the student.


Wait till UDK4 premiere. It will mark beginning of the end of animated movies made by big players. Next Shreck/Happy Feet/Cars/Whatever will be made in the basement by 14 year old kids that have nothing better to do in spare time.


I think he is genuine. He did mention that he has a few years experience with After Effects, which would have helped.


I think this quote from the short’s creator explains everything

I modeled 95 % of what’s in the video. But It wouldn’t look as it is without Furthermore, my true weakness are characters (That’s why in the animation, all the character scenes are dark, nearly hiding them from being shown, or fully lighted). To create them, I used the MakeHuman software. Thanks to this software, a naked human character pops up in the 3D view, and you can start from this helpful basis, like suiting a mannequin, re-work the body. Plus, clothes can be generated as well. So no, the character are not entirely my creations. I’m very limited when it comes to characters, I don’t focus on topology either, getting it right is really difficult to me, and you’ll have to wait long before seeing a simple walk and dialog in a daylight environment from me. I prefer to use the “Not showing the shark” tips from Jaws (Thanks Steven Spielberg). For the eyes scene, the head used the same process, I didn’t modeled it myself, it has been generated, mixed with other skin (material) textures, then received make up. The futuristic “suit”, is an old model of mine with a few additions (like lights, oxygen , etc…), you may have already seen him in previous animations. I dream to be able to create anything, anytime. Obviously, i’m far from there yet, and I can’t just take what’s in my head and transfer it into a screen.

So yeah… you can make “Space Oddity II”, by yourself, with the tools he used… sure.
But people should not take that to mean you can make anything just the way he did.

I can understand what he’s saying and some of the methods he used are similar to what we did on our Blender film.


Does UDK4 have a “make animated movie” plugin?


That was probably a joke directing at the VFX thread xD

But yea the environment alone looks entirely possible to be genuine.


No. But is does have built in realtime renderfarm that can fit in one workstation with good graphic card. Thats more than enough to shorten production time and amount of people needed.


It’s a bunch of suspenseful sequences “inspired” by various movies and games with carefully chosen camera work to need almost no assets or deformations, put together in a textbook manner (as in safely played cinematography and direction fundamentals), with nothing technically or narratively challenging in it.

Yes, I fully believe he made it on his own, and that he got there in a couple years as a hobbyist in his own time. His strength clearly lies in not overdoing it, and having good taste in borrowing.
He took on a bite sized project and did it well. Kudos to him, he’s smarter than most newcomers.

Why would people doubt it is beyond me.
Without wanting to take away from his accomplishments, an experienced generalist given a storyboard and a full set of paid for production tools could get it done within a month of working days very, very easily, and to a higher technical standard.

Schools focusing on CG, for the most part, aren’t gonna be robbed of any students, they are already largely preying on the naive or uninformed, and anybody saving their money on one and going instead for more affirmed and traditional courses, or the self-thaught route and investing that money and time in professional workshops and cheap training material, would only be doing a smart thing.

VFX/CG shops are unlikely to get robbed of work because there’s one person out there who managed to stitch together a bunch of cliches as a learning experiment and retain some good taste in the process.

I commend the guy for his dedication, drive, simple plan, abundance of smarts in choosing what to do and how, “borrowing” tastefully, and a number of other skills that define him more as a potential director and producer than as a CG craftsman IMO.

I don’t feel the jobs of millions and the poor CG schools are at any risk. If common sense and good taste, the two main qualities of the creator, were so easy to come by and so abundant, the former wouldn’t even exist anyway :stuck_out_tongue:


You really think a render farm is in any part the distinction between the large shops working on full feature content and the ones who don’t? Or that the farm is in any way what makes the production times as long as they are?

You are sorely mistaken.


Yes, I believe and I’m not mistaken. Visualization takes a lot of time when you have just 4-16 cores with standard render engines. Now, when you have it realtime, that changes a lot of things. Another thing is realtime collision and destruction. There is a CryEngine plugin for car destruction that gives better results than 80% people here that don’t have experience with destroying things would get if they do it with normal simulation that also takes time. Instead they can tweak it just on eye and don’t have to wait hours for result.

Look at Final Fantasy Spirits Within and compare it to Playstation 4 Deep Down presentation. Square Enix needed 900 Pentium 3 processors to make it. Now you have this in quality in realtime with probably toon of realtime simualtion possibilities that they couldn’t do 10 years ago.

How many people worked on Happy Feet ? How many would work if they had already possibility to work in realtime engine that could deliver the same quality with both, graphics and simulations ?


I am not sure if you are joking or being clinically obtuse. Do you believe the biggest roadblock in making a good film is a plugin for car destruction and a realtime renderfarm (whatever that means)?


Can’t tell if serious…


For people that have talent? Yes. Thanks to this they will have possibility to focus on things they should, don’t waste time on rendering/simulation and don’t cut on their vision.


Your underestimating the scope of making a feature film. These things still take time, people or money as well as hardware and software. We already have game engines quite capable of creating images for animated films - so if all it required was a fancy game engine we would have seen a lot more films made the way your describing.

For what it’s worth there was a French film made using a game engine a few years ago. It didn’t do very well…

I do think that someone will probably make their own one-man-band feature film using a game engine, but I think it’ll be like this short - it’ll limit itself to certain stylistic choices or animation in order to make the whole project feasible.


OK… but think of it this way… with UDK4… say you want to make a set… and it’s the interior of the Sistine Chapel… do you really think it makes a difference if you have 900 node render farm or UDK5 or 6?

Do you think replicating the interior of the Sistine Chapel takes any fewer artists today than it did when Square was making Final Fantasy Spirits Within?

Think about it.


This is so far from realistic it makes me cringe.
I was going to type a long response, but it’s so wrong on so many levels I don’t even know where to start…


What he did isn’t all that hard. There’s no lip syncing, no facial animation. Very simple walk cycles. Just camera movements and lighting. Even the robot is a simple animation.
The modelling and texturing might take a while.
He just made a plan, stuck to it and finished it.
That’s the hardest part.


Let me chip in.

UDK came with a lot of tools. Texture generator, landscape and tree generation. And a lot of real time editing. This, combined with current character animation tools, and motion recording with IPI.

For those with skills, a lot of thing can be done. And there will be markets for such movie. Red vs Blue and the rest shows that there are market for this "machinima’ things. Its not anything new. But with UDK, and CryEngine, and now even Unity demo their ‘Butterfly Effect’ that means we have move past Quake + FRAPS days.

What I don’t get it is that why Blender didn’t focus on machinima with their blender game engine, specially considering the licensing problem in creating games with it.


Im really glad the the more level headed responses have come out of CGTalk. Something that really irks me is when artists post up their work, and people call them liars.
I have had comments on my YouTube channel, and even emails where people have actually taken time out of their days to tell me that Im a liar, and swear at me about my work.

This trailer is very impressive from an editing and visual point of view. I really love the camera work and lighting.
He clearly borrows from films like Alien, Sunshine, and 2001-and rather unashamedly so. Which to be honest, all of those films heavily borrowed from previous ones. There is a certain language that ‘Space’ movies have-certain shots that are shared among them, and this trailer uses them all very effectively.

I really dont see how a statement like “If his claim is not true, he would be guilty of robbing art/film schools of potential enrollees and possibly of robbing VFX/CG firms of potential clients as well. Here’s a recent interview with him. Do you think he’s lying? What do you think art/film schools should do about such claims?” (apart from the double negative) is even remotely in the correct ballpark as fair.

2 years is a HELL of a long time. I started my project 2 years ago with NO experience in game development, music, or sound design-and in that time have taught myself how to compose music, do a decent sound mix, animate for games, and program and script a game that has been given press along side some of the larger indies, and commercial games. People that don’t think that something like this could be studied and recreated in 2 years are clearly terrible at time management.

A statement that someone made over on Reddit, when someone was asking about ‘where to start’ for learning an artistic endevour (I believe it was getting into VFX) is that “In 2013, you have the collective information of mankind at your fingertips. USE IT”.