A cold foreign environment, devoid of appeal. An empty desolate space with which we design.
When we first started learning 3d at Calarts I hated it. Our first assignment was to put apples on a table in 3d space. All we had to do was manipulate them in the environment. I yelled, "THIS IS LIKE LANDING A SPACESHIP", while my friend Charlotte Jackson, laughed and did it easily.
I made a big joke out of it, but I was legitimately frustrated.
I felt like if I couldn't do something so simple, it wasn't my fault. I felt it was a problem with the way that the program was designed.
Odds are, if you feel like the interface is unintuitive, thats because it is. There is almost certainly a better way to do whatever it is that you are doing. The flaw you sense could have a simple workaround, or it could be woven inextricably into the framework of the interface itself. Either way, it should be addressed. In order to find spontaneity, we must dismantle the barriers that separate us from our work.
We have it drilled into our heads. I hear it over and over again, from faculty students and professionals. The same rhetoric that continues to define the way many of us interface with a tool that has enormous potential.
"You cant be spontaneous in CG"
I think this outlook is not only false, it is actually thwartive. Thinking this way prevents artists from diving into work and experimenting. It seems everyone is taught that their are pitfalls in cg, pit falls that when pushed through the pipeline will prevent the work from functioning at all. Often times these fears are baseless and do not come from actual experience.
I am not saying that theres no such thing as bad topology, what I am saying is we should figure that out for ourselves, by trying things out.
Once I realized this, I really got interested in CG. I started doing things that I was taught I couldn't do. I rigged piles of broken polygons. I plugged inverse values into explosion inputs and connected arbitrary nodes to each other, just because I felt like it. What I found was that the sandbox is much bigger than I thought, and a lot of the things I was told I couldn't do cause no problems at all.
below, you will find an example of a simulation which has intentionally strange values.
below, you will find an example of a simulation which has intentionally strange values.
I would consider this a doodle, a rough sketch. It took very little time to create. Its sort of like a gesture drawing in 3d; an attempt to explore the qualities of the medium and to challenge its boundaries.
Most of what we are taught to avoid in CG comes from big studio workflow, which is (in my opinion) a homogenized style. As lovely as much of it is, there is very little aesthetic experimentation. CG feature films invariably include sculptural forms in a realistically lit environment. There is a lot of time spent experimenting and pushing CG towards realism at big studios, consequently, the line between whats real and what isn't is blurring.
Here is a link to a fantastic blog post about current lighting techniques (which is where I got the images of the shrimp and the guy)
http://www.fxguide.com/featured/the-art-of-rendering/
These tools are incredibly freeing to the modern animator. The conceptual box that I work in is HUGE. In theory, I have the capacity to create work that is tactile... hyperreal. In addition, I maintain the ability to create work that is completely graphic and stylized.
So, where are the experimental cg films? What is cartoon CG? We have a medium that can create an inconceivable amount of lighting situations yet we rarely see people pushing the limits of non-realistic lighting in big productions.
This is partially because animation in feature films is hardly spontaneous. The intention of the crew is to execute an existing aesthetic, not to discover a new look. The visual experimentation takes place before production. I have an immeasurable appreciation for this type of work. We call it visual development. Big studio workflow maintains a segregation of visual experimentation. All the stylistic choice making happens at this stage. After that, each stage of production does what it can to achieve the visual benchmark set by the development stage.
I'm not necessarily condemning this method. In a feature production it is important that the entire film has a consistent look. Creating a clear bench mark for the look of the film helps to maintain that consistency. However, this is where I feel like the production loses its spontaneity.
I was taught not to draw from photographs. That was always one of the biggest pieces of advice I would get from faculty and students when I was applying to Calarts. This is because a photograph is a still representation of life. It is a unique interpretation of an existing moment. Doing a drawing of something which has already been digested is an easy way to create a stiff, uninteresting drawing. The gesture of the photo is only diluted by the drawing, which is essentially a double abstraction. In addition elements in the photo might not deserve interpreting because they only exist as an attribute of the photo. Intricacies such as lens flares and aberrations create visual interest in a photo, whereas they make no sense in a drawing. The term used to describe this type of nostalgic representation is skeuomorph.
In CG film making, particularly big studio workflow, a double abstraction that I have observed is orthographic modeling.
2d design has innumerable elements worked in to it that function as shortcuts for describing volumetric forms. Modeling to orthographic drawings is like tracing a photograph. The drawing itself has perspective worked into it, whereas the orthographic cameras in 3d have literally no perspective whatsoever. Tracing a strong 2d design in this way would not produce a 3d design with the same qualities. Of course, a lot of time is spent working on the design afterward, in 3d space. However, after the orthographic process, the mentality is that we should try to meet the threshold of appeal that was established in the drawing.
This is of course impossible, they are different things. This type of workflow creates a mentality that is incredibly restrictive for the modeler. They are expected to perfect the double abstraction; denied the creative freedom to develop a language that is unique only to 3d.
I think an ideal alternative would be to have a strong relationship between the 2d designers and the modelers, or even (if possible) combine those positions. The same level of visual experimentation should be applied to each stage. Modelers should be expected to have a strong sense of design! The mark making and shape language of a 2d design should inspire inspired the sculpt which has its own layer of improvisation. The drawings should be observed, not copied. Maybe then we might begin to see methods for representation that are completely unique to 3d. I am not suggesting I know what those things are, thats what makes it exciting.
Here we have a Giacometti painting, and a Giacometti sculpture:
The similarities in tone and resonance are striking, but fundamentally their implementation must be different. The painting has qualities that the sculpture cannot possess, such as its 'lost edge' --its vague areas of darkness around the shoulders. The sculpture has a tactile quality that the painting could never possess; an image could never do justice to its form. One can only compare them subjectively. Each medium has its own set of rules, within which improvisation creates a unique set of marks.
This is but one example of studio workflow that lacks iteration. Ultimately I would like to see a production where the artists are trusted to experiment with the visual aesthetic of the project at every level, under creative direction.
Most of what I find to provide the most compelling experimentation comes from commercial work, student and independent films, and interactive material. I hope that the buddhist mistakes that we discover bleed into larger productions. I see no reason why that won't happen naturally as artists familiarize themselves with this amazing tool. I leave you with... a bunch of cool stuff:
Courtmétrg from La Mécanique Du Plastique on Vimeo.
The Zoo from ZEITGUISED on Vimeo.
Loom from Polynoid on Vimeo.
TRI▲NGLE from Onur Senturk on Vimeo.
SHOWREEL 2012 from Gero Doll on Vimeo.
We can now simulate things like caustics, bounced light, and refraction on our desktops at home. The amount of CG artists achieving uncanny realism in their sculpts and renders is staggering. Animation now finds itself with a much higher ceiling. It can achieve absolute realism; much such as baroque painters achieved realism in oil.
http://www.fxguide.com/featured/the-art-of-rendering/
These tools are incredibly freeing to the modern animator. The conceptual box that I work in is HUGE. In theory, I have the capacity to create work that is tactile... hyperreal. In addition, I maintain the ability to create work that is completely graphic and stylized.
So, where are the experimental cg films? What is cartoon CG? We have a medium that can create an inconceivable amount of lighting situations yet we rarely see people pushing the limits of non-realistic lighting in big productions.
This is partially because animation in feature films is hardly spontaneous. The intention of the crew is to execute an existing aesthetic, not to discover a new look. The visual experimentation takes place before production. I have an immeasurable appreciation for this type of work. We call it visual development. Big studio workflow maintains a segregation of visual experimentation. All the stylistic choice making happens at this stage. After that, each stage of production does what it can to achieve the visual benchmark set by the development stage.
I'm not necessarily condemning this method. In a feature production it is important that the entire film has a consistent look. Creating a clear bench mark for the look of the film helps to maintain that consistency. However, this is where I feel like the production loses its spontaneity.
I was taught not to draw from photographs. That was always one of the biggest pieces of advice I would get from faculty and students when I was applying to Calarts. This is because a photograph is a still representation of life. It is a unique interpretation of an existing moment. Doing a drawing of something which has already been digested is an easy way to create a stiff, uninteresting drawing. The gesture of the photo is only diluted by the drawing, which is essentially a double abstraction. In addition elements in the photo might not deserve interpreting because they only exist as an attribute of the photo. Intricacies such as lens flares and aberrations create visual interest in a photo, whereas they make no sense in a drawing. The term used to describe this type of nostalgic representation is skeuomorph.
In CG film making, particularly big studio workflow, a double abstraction that I have observed is orthographic modeling.
2d design has innumerable elements worked in to it that function as shortcuts for describing volumetric forms. Modeling to orthographic drawings is like tracing a photograph. The drawing itself has perspective worked into it, whereas the orthographic cameras in 3d have literally no perspective whatsoever. Tracing a strong 2d design in this way would not produce a 3d design with the same qualities. Of course, a lot of time is spent working on the design afterward, in 3d space. However, after the orthographic process, the mentality is that we should try to meet the threshold of appeal that was established in the drawing.
This is of course impossible, they are different things. This type of workflow creates a mentality that is incredibly restrictive for the modeler. They are expected to perfect the double abstraction; denied the creative freedom to develop a language that is unique only to 3d.
I think an ideal alternative would be to have a strong relationship between the 2d designers and the modelers, or even (if possible) combine those positions. The same level of visual experimentation should be applied to each stage. Modelers should be expected to have a strong sense of design! The mark making and shape language of a 2d design should inspire inspired the sculpt which has its own layer of improvisation. The drawings should be observed, not copied. Maybe then we might begin to see methods for representation that are completely unique to 3d. I am not suggesting I know what those things are, thats what makes it exciting.
Here we have a Giacometti painting, and a Giacometti sculpture:
The similarities in tone and resonance are striking, but fundamentally their implementation must be different. The painting has qualities that the sculpture cannot possess, such as its 'lost edge' --its vague areas of darkness around the shoulders. The sculpture has a tactile quality that the painting could never possess; an image could never do justice to its form. One can only compare them subjectively. Each medium has its own set of rules, within which improvisation creates a unique set of marks.
This is but one example of studio workflow that lacks iteration. Ultimately I would like to see a production where the artists are trusted to experiment with the visual aesthetic of the project at every level, under creative direction.
Most of what I find to provide the most compelling experimentation comes from commercial work, student and independent films, and interactive material. I hope that the buddhist mistakes that we discover bleed into larger productions. I see no reason why that won't happen naturally as artists familiarize themselves with this amazing tool. I leave you with... a bunch of cool stuff:
Courtmétrg from La Mécanique Du Plastique on Vimeo.
The Zoo from ZEITGUISED on Vimeo.
Loom from Polynoid on Vimeo.
TRI▲NGLE from Onur Senturk on Vimeo.
SHOWREEL 2012 from Gero Doll on Vimeo.
This was brilliant Charley.
ReplyDelete