Getting familiar with the past

The last couple of weeks have been quite a mix between work and research. I was working on a freelance and a private project as well. Chasing a deadline for the freelance one was a good learning experience in many ways (more on that later) and at the same time taking breaks to continue my private project kept my mind fresh. During this time, I have also started researching the history of CG. I wanted to go through it in detail and take away the elements that I think are useful to me.

Now, this post will not list the entire history that I've went through, but rather contains the important elements that I think will help me better figure out where my place is in CG. Studying history is always beneficial, because you can learn a lot from previously encountered situations. You see how the top artists of this industry solved various problems. You can gain a better perception about how to approach different situations and by studying the past you can maybe have a slight idea about where this industry is heading to. This can help you better plan your career, thus ensuring a better employability in the long run.

Going way back

1950 - 1960:

The first signs of CG date back to the 1950's, the early days of computing. You had strip charts, pen plotters etc. But, in these times everything was so ineffective and expensive that there was no use for it, only for special departments like the U.S. military etc.

1960 - 1970:

As time progressed, so did technology. In between 1960-1970 wire-frame graphics appeared. This meant that computers could draw various shapes with the help of lines. Filled polygons was still a technology of the future, but we had lines.
Nowadays, there is a very important tool at a computer graphics artists' disposal. This is the pen tablet/drawing tablet (e.g. Wacom). The birth of this wonderful equipment dates back to the times of wire-frame graphics. An MIT student named Ivan Sutherland recognized the potential of man-machine interactivity and invented a lightpen with which you could interact with the display. As you moved the pen, the computer generated a new image based on your movement.

Sutherland is also responsible for many of today's commonly used algorithms in CG.
During this era, the display processor was invented as well. Scientists realized that the entire graphical workload was a bit too much for the host computer. As a result, a special purpose computer was invented. It was called the display processor (DPU). The host computer would compile a list of the tasks needed to be calculated and sent that list to the DPU. This resulted in a significant advance in speed.

1970 - 1980:

The next big step, which is actually the basis of today's graphics, came between the 70's and the 80's. A wonderful thing called Raster graphics was invented, which set the beginning of CG standards. Raster graphics means that the image is produced as an array (the raster) of picture elements (pixels) in the frame buffer. This technology allowed us to take our wire-frame graphics and fill in the spaces between the lines that built up a shape. It's fair to say this was a huge leap forward in visualization.
Artists start to experiment with CG to create animations. CG animation had a huge advantage over the previous cell-technique in terms of speed. In 1942 Walt Disney had 1500 artists drawing the images for Bambi, whereas today, theoretically, a single person could animate an entire Shrek movie. The 70's represented the first signs of this huge leap ahead in speed.
In 1974 the first animated short, "Hunger" was released, which was nominated for Oscar.
At the University of Utah, scientists developed a 3 dimensional object, the all famous, teapot. This 3D object is very commonly used today because of its interesting shape, which makes it very useful for test-rendering. After experimenting with the teapot, people realized that it's realism was all about the lighting. Lighting is everything when it comes to presenting your work, especially today. As an Environment Artist, I need to have excellent lighting skills to present my work as appealing as I can.

During this era the workstations and personal computers came along as well. Today, we don't make much of a distinction between the two, but in reality they have evolved from different roots. Early workstations were characterized by network-connection (client-server model) and a high level of interactivity. It's beneficial to study workstations in my opinion, because a lot of VFX and film companies turn to so called "Render Farms' today to render out their work. Render Farms are basically a huge network of workstations. The calculation power of hundreds of these machines is used to render out data, which otherwise would take unacceptably long. It doesn't sound unwise to optimize my work, prepare it for the render farm.





Adulthood of CG

1980 - 1990:

Going a bit forward to the modern days: realism came to CG between the 80's and 90's. Technologies like smooth shading, environment mapping, bump mapping came along. Smooth shading was essential, because previously each polygon had their own light reflection information built into them, separated from the neighboring faces. With smooth shading, all of the polygons' light reflection information is calculated as one, so you don't get the sharp edges between them. You get a smooth, clean result.

Pixar's Renderman was developed during this period which started to bring CG to a whole new level quality.
In 1982 Walt Disney created the first feature film combining live action footage with CG: Tron. The costs to make the movie went up to 17$ million  and 4$ million alone was spent on CG. 6 animators worked 9 months to create more than 1000 shots. Even with all the technological effort, Tron was unsuccessful. People started to critize Computer Graphics for its of believability and no emotional involvement with its audience. In my opinion this was a very good lesson for future artists working with this medium. Technology by itself is NOT ENOUGH to connect with your audience. In order to convey meaning through your work, you have to prioritize storytelling and content over technology.

Advertising agencies start having a big interest in CG. More and more ads pop up which use visual effects to persuade their viewers.
In 1983 the ad called "Sexy Robot" was released, which was the first computer animation that used the movements of a real person and translated them by using reference points painted on the body to a CG figure. This, all, for tinned food ad.

1990 - 2000:

In 1992, Silicon Graphics developed one of the fundamental building blocks of today's graphics, the OpenGL API. It allowed a whole new plethora of possibilities in software development. Hardware wise texture mapping, blending and new buffers such as the stencil buffer came along. Why are these buffers and mappings important for me? These techniques are still used and applied today. My aim is to become a better and more effective artist. Speed is very important nowadays. The quicker you can produce quality work, the better for everyone. In order to speed up my work it is essential to understand as much of the technical possibilities I can. By understanding the fundamental structure of the tools I am using, I get more freedom and could maybe come up with new solutions for my workflow.

In Toronto, Canadian programmers developed a more sophisticated CG program, called Alias. Alias was used in the movie Terminator 2 and also in Jurassic Park. Both of these films received Oscars for their visual effects. Later on, Alias joins up with Californian company, Wavefront to develop one of the most commonly used CG programs of even today, Maya.

In 1995 CG went through a major test. The first full feature-length film, Toy Story was released by Pixar. It had an immense success, thus showing the capabilities of this new industry. They've found the balance between storytelling and technique which skyrocketed its success. It was a whole new experience for the audience, yet enjoyable.
Producing quality movies in CG is very hard, because of the potential consequences of failure. It is much more costly to fail in CG, one of the main reasons why it takes 3-4 years to create a movie like the Toy Story.

The modern days

2000 - present:

Photorealism becomes possible. Powerful graphics cards for PCs dominate the market. (nVIDIA, ATi/AMD etc.).
The movie "Shrek" was released in 2001. It took 300 animators 4 years to create more than 1000 to create the film. This is once again proving the point that you have to have a perfectly polished product when you present it to the audience, otherwise the downfall is going to be a massive on. Shrek used very sophisticated character interactions and physical simulations (cloth, fluid) were used for the first time quite heavily. Shrek was a huge success.

Because the quality requirements were rising, Render Farms open up, to allow a possibility for film companies to render out their work. It took 500 computer servers 100 million hours in total to render out 86 minutes of Shrek.

As technology advances in terms of hardware and software as well, the cost to produce CG (let it be animation or still images etc.) keeps going down. This creates the possibility for smaller companies and even individual artists to produce animations. As the limitations start disappearing, the focus is more and more on story telling and art itself. Studies outside of pure CG is required nowadays. E. g. a modeler needs to have a good grasp on shape, form and anatomical studies. They need to study various objects in nature in order to meet the requirements of today. Technology by itself is not enough as I've mentioned before. Fine art skills help a lot.




I did not write about how the game industry was effected by the development of CG, as I dedicate one separate post to that topic.

Reference:   -   History of Computer Animation - by Discovery Science   LINK
                  -   Computer Graphics in Games - by Computer History Museum   LINK
                  -   Computer Graphics - Lecture 1 by Barbara Hecker   LINK