Now, push the movie's technical and creative design to the edge of any cinematic limit. Editorially collage human actors with immersive photography and CGI together to create a new format. Add tons of colors out of any palette of your wish. Overlay a multitude of expressive non realistic effects throughout the background, midground and foreground.

Forget the feeling of film texture.

Here you have the real first "poptimistic photo-anime": Speed Racer, whose world captures the eyes with techno colors, ever bending perspectives, car-fu and flashing lights, in what is the latest accomplishment by the brother Wachowski, John Gaeta (Award winning Matrix trilogy's visual effects designer), Dan Glass(VFX Supervisor for V for Vendetta) and their team's techno-magic spells.

If, within the Matrix, he had creatively enabled several kinds of emergent effects techniques to visualize a new type of movie, whose innovation was awarded with the recognition of his abilities by the Academy Award for Visual Effects, BAFTA Award for Best Achievement in Special Effects, and a pair of VES Awards (to name but a few); in Speed Racer John Gaeta and his long time collaborators move forward again with the incubation of a completely new genre and film format.

A step ahead in every project he enters into, Gaeta is a man with a vision, a man capable of inventing new ways to let you see the world as you've never seen it before, treading within the realms of dreams, virtual reality, computer generated content and excesses to come out with the ingredients necessary to create an explosive product...and - oh yes! if it weren't stressed enough already - Speed Racer is one.
Speed Racer's backbone philosophy was to deconstruct from what had been accomplished in Matrix but, like a serendipitous discovery, by searching for less, an unexplored land was found, from where the masterminds came back with yet another load of otherwise unconceivable special effects.

From its conception, to its first ideas to the proof of concept; from its composition to its post production, VRMag invites you to discover the tricks and the creative and inspired people who were the minds and hearts behind the scenes, in their words: from John Gaeta to Euisung Lee, from Dennis Martin to Lubo Hristow and Jake Morrison, you will walk through the scenes of Speed Racer, with this exclusive material, courtesy of Warner Bros. Entertainment.

OBJECT MOVIES OF THE CARS

MACH 5

Spin horizontally | Spin vertically
© 2008 Warner Bros. Ent. All rights reserved

MACH 6

Spin horizontally | Spin vertically
© 2008 Warner Bros. Ent. All rights reserved

THORAZINE

Spin horizontally
© 2008 Warner Bros. Ent. All rights reserved

How did you and your collaborators decide to use photographic VR images so intensely for the backgrounds of the movie Speed Racer?
Spherically constructed location photography has been used extensively for visual effects for years. For us, many of the fundamental shooting and post manipulation techniques being pushed en mass in Speed Racer were first advanced during the Matrix Trilogy by our remarkable R&D teams, whom became masters at an emergent new form of "simulation cinema".

HD resolution, QTVR-like photography and it's inverse cousin, Universal Capture, were a base from which we photogrametrically extrapolated 3 Dimentional form(static and temporally dynamic) and texture in one heavy acquisition. Meaning: photography of real world people, places and things can be transformed into hyper real virtual content components from which one could blend with other elements(cgi, live action photography, etc.) to make movies. For example, Neo and his surroundings.

Since the first Matrix, The Wachowski brothers, myself and a spectrum of talented people tried pushing shot design through an emerging new paradigm dubbed Virtual Cinematography.
In Speed Racer we are still pushing Virtual Cinematography, yet refocusing earlier techniques through a retro-modern style filter we call PHOTO ANIME.
In a way, Photo Anime is a bit of a creative deconstruction of the past, a technical devolution in pursuit of getting at a freer and ever fun new Universe. We became much more interested in how a moment made you feel then what was visually real. Qualities that Manga and Anime have in their DNA.

The Wachowskis challenged us to consider the form and we realized that the purest way to invent "live action anime" would be to create many compositional, color and visual analogies between the very expressive structure and attitude of the best of Japanese cell animation and state of the art photography and visual effects. Therefore, a cell animated matte painting would essentially be replaced by a pano and an old down shooting rostrum camera would now be replaced with 3D compositing software and so forth.

The idea to replace anime images from the original cartoon with live action and photographic images re arranged in a "2.5 D" way rather then a "3D" way moved us away from Sim Cinema more toward " Sample Cinema" because we still felt strongly about the possibilities of enhancing real world things into hyper real and pop art things. "Sim" inferring a " 3 dimensional, computer generated and image based render" toward the virtual reproduction of people, places and things into effects, shots or scenes in a movie. "Sample" meaning a composite enabled shot or scene, constructed, collaged or layered 2D images based on real world people, places and things. Similar to the idea of creatively mixing sampled sound and music into new stand alone compositions.

So, Speed Racer is perhaps the most expansive use in history of high resolution or HD QTVRs (which we call bubbles) toward a cinema application. They are applied to many, many scenes and are often beautifully combined with other elements, people and CG objects, like cars, and other phenomenon.

SPEEDRACER VR BUBBLES

click here to view Liquid Room, Moeckernstrasse 10, Berlin Kreuzberg, panorama
© 2008 Warner Bros. Ent. All rights reserved

click here to view Casa Cristo - Cruncher Block's Room panorama
© 2008 Warner Bros. Ent. All rights reserved

click here to view Casa Cristo - Mt. Doom
© 2008 Warner Bros. Ent. All rights reserved

You can find more original Speed Racer location bubbles here.

What were you doing or looking at before Speed Racer, did this have an influence on your interests toward a new format?
After we had worked many years on the Matrix trilogy, we had given our minds a chance to rest and ourselves a chance to essentially come back into the real world and see what was happening around us. You know, when you work on a project as long as we did on the Matrix trilogy (it took us 4 years of sustained work to make these projects), a lot happens in the outside world; while you're inside of an enclosed studio, creating your own reality and often you lose touch a little bit with what's happening in other areas of art and entertainment.
After The Matrix trilogy was over, we had all decided to take a little bit of time and become "re infused" with contemporary happenings in all areas, whether that be media or technology, entertainment...everything from online experimentation, art experimentation and digital experimentation, to movie making as well.

At the time I was consumed with the idea of enslaving video game engines toward cinema-interactive hybrids...I was looking at the format potentials available for passive stories networked with embedded play portals and the challenges with mashing real time virtual cinematography with real time graphics. I want to make film imagery in real time and I want to port the real world into games. As the landmark game Myst showed there could be a marriage made with limited real time three dimensional compositing of images of photographic, QTVR-like or computer generated environments with actor video clips placed on cards within the explorable mise en scene. All that was very intriguing to me, I just wanted in in High Definition.

Moreover, my interest in traditional effects movies was drifting, ...
I was looking into the shadows for examples of frontier type content that could satisfy my cravings for visually stimulating, psychological tweeked entertainment. Within the previous couple years pre Speed Racer, Sin City came pretty close for me. I thought it's hyper-graphic design and structure was starting to break some old school vfx genre molds. It was attractively closer to dark and mental comic book art then I had seen before. Robert Rodriguez pushed a non integrationist design approach and an end to end composite methodology which produced a different visual experience and thusly a different viewing experience. I also had some real appreciation for "Eternal Sunshine for the spotless Mind" by Michel Gondry: another hero taking a wholly different approach to sensory cinematography. In that film, cinematography and innovative editing was the visual effect.

I spent some time in Japan after the Matrix Trilogy and directed an HD short there called Homeland. My shooting, post and visual effects teams were all obsessed provocateurs of Manga. I started turning my color dials to 11. When Larry and Andy called to tell me to let me know where they wanted to head with Speed Racer, creative synchronicity seemed instantaneous.

Which were the conclusions you reached?
Speed Racer had to be high definition Digital Cinema in order to achieve the array of anime design analogies and faux cinematographic nuances we were hoping for. We didn't want the "film" to look like film with grain or noise texture; we didn't want it to look like video, we wanted very high fidelity, super clean, pure pop art color, not cinematic yet not toon (ie. A Scanner Darkly). We wanted people to have the same sensations they get when you see a beautiful animated work on the big screen, but we wanted it to be based on photography.
We cared about the realities that film distribution was a multi week big bang but yet legacy media is where so many eventually see the work. We thought it was time to consider the ideal process to acheive a spectacular widescreen AND HD Home Theater/Blu Ray destination.
When we were able to get past that fundamental film vs HD decision, we felt liberated.
So - essentially - what we wanted to make was this photo-anime.

We felt that the pivotal barrier in moving toward more expressive, less real, visuals was the perception that today's audiences require photorealism from their effects in order to prevent a drastic disconnection from the photographed drama ...but that notion is really connected to the reality of the story universe itself. The Wachowski's were aiming at a whole different wavelength of narrative which they were attempting to port from their childhood imaginations. Besides, we wanted to have fun, and we wanted to do it across an entire picture.
Thus, we attempted to devolve away from the techniques of precision integration of all live and fx elements and evolve toward a more emotional-graphic underscoring of moments. Like things were done in older days of animation, before computers.

To help us with this, we created an analogy between the idea of a classic down-shooting camera(ie. Acme, Mitchell) shooting drawn cell art as planar layers while doing rostrum camera moves upon them and the idea of a virtual camera as enabled within the software of a 3D compositing platform aimed at depth layered 2D elements(like actors, set pieces, etc.) in front of a 360° spherically photographed location. The illusion of a camera moving through space in cell animation was often done by allowing fore, mid and distant layers to slide against one another.. Parallax, was a nuance that we thought we could manipulate to capture the feeling of inexactness and often exaggerated depth in anime. So in addition to our foreground actors, and production design midground pieces we decided to slice up our bubbles in multiple layers as well.. like layers on onion.

The HD QTVR or Bubble was essentially our navigatable surrounding world art. Depending on the actual location and the spatial proximity of elements surrounding the camera, one could actually do certain types of camera moves even though the perspective was frozen. Add to that a near limitless ability to customize your lens. Being virtual(also referred to as "Faux Lensing"), there was really nothing preventing us from subliminally mixing within the same frame varying focal lengths changing between fore, mid and distant layers.Flatten or near orthographic characters mashed against bizarro fisheye backdrops happen a lot in Manga, so why not in our movie? Designer defocus and motion blur become a matter of choice, not a factor of mathematically grounded depth of field parameters. A super telephoto lens could have infinite depth of field at sunset, cars could have one shutter angle blur approach to the nose of car and a different angle for back. At anytime during a shot, all the properties of the lens could invert, if expressively justifiable within the context of the story moment.
This was "Virtual Cinematography" aimed at kids and therefore we tried to adopt kid logic when adding cartoon attributes to our images. This, no doubt, may inevitably lead to some disconnect in older viewers unable to detach them selves from traditional film.

Once we had determined that our PHOTO can be ANIME analogy seemed sound enough, we began a lot of color and light design experimentation and previsualization work to simulate the possible looks of the movie.
So, there were phases in which Dan Glass and I worked alongside some very talented people...Euisung Lee, Steven Lawes, Lubo Hristov, and more and we started to plot out options of all sorts for approaching the shoot and post cinematography and "Editography" phases. These studies covered every topic from drama to blending actors and "Car Fu" moments together. Lubo and Steven would experiment with collage, colors, detail enhancement, cutting depth layers and juxtaposing character/actors inside these virtual spaces. Euisung Lee's studies involved many things including grafting and gluing bubbles together, then attempting to cinematically navigate the merged location. A location could have majestic rooms and doorways...and if we wanted to, we could take the virtual camera and fly through the doorway and then, on the other side of the door, there would be another bubble, in which you could go and navigate and continue the scene. If we wanted, we could connect multiple bubbles together and create non-ending steady cam type camera moves from bubble to bubble to bubble.

Can you tell us about the proof of concept?
Euisung Lee: The tests I made were various attempts to utilize VR image to create shots, for stylistic and practical purpose. VR is not 3D but it has pseudo depth, which allows you to travel within it to certain extend. That is, as long as you don't travel too far and the subject of your scene doesn't make contact with the VR background. (i.e., never show the feet touching the ground) As you can see in many examples VR background worked quite well, and the slight distortion caused by camera being off the center of the VR sphere, which would have been an anomaly in conventional films, had a potential of being part of the stylistic vocabulary for the Speed Racer.

Another example of conveyer belt ground plane shows emulation of 2D animation trick. Often 2D animation would create an illusion of characters moving by sliding the backgound and the foreground in different directions. Ground plane is animated to have perspective and it loops with the character's feet, and the BG would just slide and scale slightly. What we did was the 3D or 2.5D version of it.
Character is walking on a looping conveyer belt, and is moving through a VR environment, so you are not quite in a 2D nor 3D world yet gets the sense of moving through space.

click here to view LHSN conveyerbelt test birdeye
© 2008 Warner Bros. Ent. All rights reserved

click here to view LHSN Vrdistortion test birdeye
© 2008 Warner Bros. Ent. All rights reserved

click here to view LHSN Vrdistortion test
© 2008 Warner Bros. Ent. All rights reserved

click here to view IHSN 3D cut v01
© 2008 Warner Bros. Ent. All rights reserved

click here to view IHSN 3D cut v02
© 2008 Warner Bros. Ent. All rights reserved

At this point of conceptual development, we now needed to demonstrate that the formula for this approach (HD Actors + Bubbles x Hyper Enhancement) would actually look engaging at high resolution. To do this, we got together with the brothers and David Tattersol and designed a proof of concept test consisting of test shots which we would act as a canvas for look demonstration and mini scene. We would test the visual fidelity and attributes of various HD camera systems while shooting variations in lensing, light and camera perspectives.
We pre designed a slate of camera, color, blur and beauty modification type ideas that we referenced with Dave T on stage. We also decided that at this point we needed to power up some innovative post processing strategies as well as on stage compositing approaches to experiment and review with Larry and Andy as we shot. For this special expertize we turned to our associates at Digital Domain. In particular, Kim Libreri, Mohen Leo and especially Darren Poe whom crafted certain essential faux lensing looks in the final presentation of the test. It was also in the making of this Wachowski directed mini scene(Racer X based), that the brothers employed some new editorial stylizations as well as found their initial interest in the potential format approach.

How did the movie making process work?
As soon as we and the Wachowski s were convinced of this new process, we realized that it required a massive amount of element acquisition considerations and pre-thought.
What were the right locations to shoot?
How many perspectives had to be taken within these locations?
How much should we pre factor the actors blocking within a given location and the potential editorial outcome(ie. perspective changes) before we shoot it?
How do we interlock cinematography strategies with the D.P David Tattersol(whom shot the the last 3 Star Wars pictures for George Lucas)?
Would we ever need to interchange between multi planar 2D bubbles and a more hardcore 3D virtual background approach?
Should we prepare for this?
How should we approach virtual car racing across moving landscapes?
How do we visualize in 3D, then deconstruct to make 2.5D optical illusion based, infinitely endless roads cannibalizing bubbles as a source?
...The questions came fast and furious and we soon realized that our "simple" 2D-like, anime-like style was actually getting a bit exponential with depth compression and perception riddles.

We realized that we would need to create a department that had never existed inside a standard film production before, and we called this department "the world unit": its job was to basically capture thousands of these bubbles around the world. So...we set this up.
There were teams of photographers within it, and it was led by Dennis Martin. Dennis is a number one photographer and one of the world's leading expert on using spherical photography in virtual locations for movies. Dennis has worked with me on the Matrix trilogy, creating image based virtual environments using similar methodologies and for the Speed Racer movie we sent him across the globe hunting for our exotic pieces.

The preproduction phase involved a lot of discussions with the Wachowski Brothers, the locations department, the art department, the visual effects department: we would determine where and when and how we would shoot all of these locations...It was a giant effort that all decided need to happen ahead of the live action photography: our goal was to shoot all of these bubbles and bring them back to Berlin where our base was. We would routinely review the quick stitched bubbles like film dailies and make selections on key positions. Those selects would then be immediately rest itched at full resolution.

What we would do at that point was to take the HD QTVR selects and some strange universe design concepts and hand them over to our World Art Director, Lubo Hristov.

Lubo Hristov, is a master matte painter, with whom I worked for years on the Matrix trilogy paintings. Lubo has some very, very special skills: he is able to paint and modify specialized photography, he can paint a photo realistic matte-painting in a traditional set, but he is also able to work on strange formats. Chief matte painter, and similar to MC Escher, he found himself painting in fisheye format in a way that transformed the QTVRs into 360 degree concept art and fantasy backgrounds for our film. Lubo transformed - through extensive matte painting techniques - the bubbles into "Techno Color", hyper real, enhanced locations, essentially making them locations within the story universe of Speed Racer.

What we would ask Lubo to do, is to get half way to the final product...but while and where we were shooting, hence considering the amount of discovery process which happens on stage. We would talk to David Tattersall - David is a very progressive director of photography and was very unafraid either of high def, or of the HDR stills approach for the backgrounds. David could study the color and light, then light and compose a complementary actor/character to into this format.

Moreover, we used a software called Sparky, derived in part from a 3D composite technology called Nuke, created by Digital Domain. Sparky could take the HD resolution painted bubbles and project these bubble onto a three dimensional model, shaped spheres - which - in turn - could be viewed with a simple virtual camera live, real time.
We could live comp the actors over green screen and compose them inside the bubble as an immediate on stage reference. I could faux lens and re aim the camera at will, to complement what the director of photography was doing on stage or to experiment with far out framing concepts. These live comps would play on high definition monitors to show the actors and the brothers their virtual backgrounds as they performed for camera.

Thanks to the software designed by Digital domain, Kim Libreri, Mohen Leo and talented others, we could have critical discussions on stage about how we wanted the scene to editographically progress but once the components are captured, we essentially just begining to make our Photo Anime.

Jake Morrison, our digital effects supervisor and head of our in house visual effects group (called "Exhaust"), did the most extensive compositing of actors into pano bubbles and pioneered an array of "Faux Lensing" and anime spun virtual cinematography approaches. It was once our Berlin shoot was complete that the theory moved to application and the design and techniques deployed went exponential.

MAKING OF EXAMPLES:

1. MEMORIAL SCENE WITH SPEED RACER AND TRIXIE

1.1 original location photo stitch:

© 2008 Warner Bros. Ent. All rights reserved

1.2 multi-layered matte painting:

© 2008 Warner Bros. Ent. All rights reserved

1.3 final composite:
(in the center a set piece of the Memorial torch to be added)

© 2008 Warner Bros. Ent. All rights reserved

1.4 view the final interactive panorama:

click here to view Memorial panorama
© 2008 Warner Bros. Ent. All rights reserved

1.5 View the final composite scene:

click here to view Memorial scene compositing
© 2008 Warner Bros. Ent. All rights reserved

2. DRIVERS CLUB SCENE

2.1 original location photo stitch:

© 2008 Warner Bros. Ent. All rights reserved

2.2 color corrected plate, matte painting elements and removed elements for multi-plane:

© 2008 Warner Bros. Ent. All rights reserved

2.3 separate chandeliers for multi-plane move:

© 2008 Warner Bros. Ent. All rights reserved

2.4 final composite:
(the water seen thru the windows will be replaced with live action moving underwater element)

© 2008 Warner Bros. Ent. All rights reserved

2.5 View the final interactive panorama:

click here to view Driver's Club panorama
© 2008 Warner Bros. Ent. All rights reserved

2.6 View the final composite scene:

click here to view Driver's club scene compositing
© 2008 Warner Bros. Ent. All rights reserved

3. RENDEZVOUS FIGHT SCENE

3.1 original location photo stitch:(the complexity of the sequence required multiple locations and extensive matte painting)

© 2008 Warner Bros. Ent. All rights reserved

3.2 original location photo stitch and matte painting elements:

© 2008 Warner Bros. Ent. All rights reserved

3.3 multiple photographic and matte painting elements combined:

© 2008 Warner Bros. Ent. All rights reserved

3.4 3D projection texture map

© 2008 Warner Bros. Ent. All rights reserved

3.5 final composite

© 2008 Warner Bros. Ent. All rights reserved

3.6 View the final interactive panorama:

click here to view Rendezvous panorama
© 2008 Warner Bros. Ent. All rights reserved

3.7 View the final composite scene:

click here to view Casa Cristo Rendezvous fight scene compositing
© 2008 Warner Bros. Ent. All rights reserved

4. ELEMENTARY SCHOOL SCENE

click here to view Elementary School scene compositing
© 2008 Warner Bros. Ent. All rights reserved

The bubbles were indispensable part in the movie. Can you tell us something more about them?
One of the things I would say is that not all the bubbles were used for actors: there were many locations that were used for extreme car racing scenes The movie is loaded with acrobatic car combat as well...
There is a form of fighting we call car fu where cars can - thanks to special mechanisms - jump and drive on rock walls and do special flips and smash into one another. Moreover, they can drive over nearly any landscape. Part of what we used material provided by the World Unit to construct were endlessly looping roadways that acted like like cartoon conveyor belts and allowed us to stage elaborately choreographed animation. We were working closely with Dennis on a visual effects strategy that not only just factored editing actors in the dramatic scenes, but also factored very fast moving action and car races. Dennis extensively photographed the Alps and rode winding roads in the Alps, and we wanted to move the camera in a virtual way that allowed us to leap frog from turn to turn, from one mountain to another, effortlessly: we wanted to fly, to be able to be behind cars, spin around them and be in front of them, in a very virtual way. All that is a little bit derivative from the camera design that had happened in the Matrix.

Since we were using a far away location we had to understand and strategically consider where he would have put his camera, how these perspectives could have near and far vantage points that allowed us to visually smash zoom from bubble node to a bubble node. Within all these interconnected bubbles we would place computer generated cars, lot's of them. Often, we pushed the very limits of our faux lensing abilities to create very intense warp and vertigo effects, effectively bending and distorting the bubble in oblong egg shapes. It was very difficult for our visual effects venders, like BUF in Paris, SONY Imageworks in L.A , Evil Eye Pictures in San Fransisco, or our in house team, Exhaust, to spatially lock cars and people to these ever distorting backdrops, often they blended the best of techniques transitioning between multi layered composited HD images and 3D projected images onto model forms(like mountains or architecture).

I saw an example, and I think it's from a Realviz study (by Greg Downing), on city bicyclists riding through Petra in Jordan, which is funny since it came out in the middle of when we were planning our car races through these environments, and these ones are the extreme of that idea. (See Greg Downing's study example here and the related VRMag article here .)

Did the house World Team shoot all the imagery required for the movie?
Almost, except a few panos from the alps shot by Alex Gollegger and the Matterhorn from you guys (VRWay).

How do you see the future of the photographic virtual reality industry?
I absolutely see a converging point where several things happen, all around the same time. There will be converging points between simulation graphics and specialized photography, some of which is going to be based on many of the shooting methodologies which are happening now. There will be a time not too long from now where digital still cameras are going to be replaced with real time(or high speed) high definition cameras that are going to be arrayed for 360° full capture acquiring motion vr like surrounds in real time as a step toward UHD(Ultra High Def.) single lens capture. In either case, a supplimental stereo element should be part of the process to extrapolate depth information in parallel for creating stereo vr and later lock up with all manner of integrated computer graphics.

Actually, this stuff will be at some point in off and online game platforms. Emergent online game platforms will someday(5-8 years?) be able to support high definition photo elements en massapplied toward image based virtual components and in a manner that allows virtual people, places and things to be manifested and blended with simulation graphics. This stuff will churn interactive immersive experiences for industry, military, porn, political messaging and new psychology.
Certainly there will be more near term experimentation in virtual cinematography(including real time, aka RTVC), Sim and Sample Cinema and the hybridization of these new type movie forms and portals to game play.

These are baby steps toward what virtual reality might look like, right now we are focusing on the the visual components, because the sensory component is the greatest mystery so far. How do you make the leap in the sensory components of virtual reality?
Let's just stay pragmatic and say that the technological advances are coming, and will be driven by simple entertainment and industry oriented goals, but - whatever the boost - they inevitably will lead towards a deeper immersive experience. VR images can concretely be looked at as steps in front of us, and we are just at the early of discovering how to actually capture what is physically in front of us and converting it into a synthetic asset.

One last question, you are a long term reader of VRMAG, can you tell us why you like our magazine?
I do like VRMag because I find it to be one of the most innovative online publications. I might say that innovation has to do with the subject of immersive visual art and photography. Also its presentation format is very progressive.

I have been inspired, especially in the last couple years, by the efforts the magazine is making and have referenced many articles while making Speed Racer.

Experience Speed Racer at a Digital Cinema or IMAX for a maximum representation of it's look and feel.

The following video interviews by Xeni Jardin are courtesy of BoingBoing TV.

BBtv: Speed Racer is "poptimistic" -- interview with John Gaeta, part 1.

Speed Racer's "photo-anime" hyperreality: John Gaeta interview, part 2.

Go Speed Racer Go Music Video

Links:
VRMag's Speed Racer coverage
Speed Racer official site

Related articles in this issue:
SPEED RACER VISUAL EFFECTS SUPERVISOR KIM LIBRERI
SPEED RACER'S WORLD UNIT LEADER DENNIS MARTIN
LUBO HRISTOV VFX ENVIRONMENTS ART DIRECTOR ABOUT SPEED RACER
DIGITAL EFFECTS SUPERVISOR JAKE MORRISON ABOUT SPEED RACER