Story and origin of computer graphics by Prof. Donald Greenberg
October 28th, 10:00AM - Viewconference 2016, Turin (Italy). Prof. Donald Greenberg can be considered the father of computer graphics. He was, with his research, the first tests and the first 3d animations in the ‘60s the first person to have opened the way for the 3D world as we know it today.
Even today, Professor Greenberg’s studies are supported and funded by many IT companies such as Microsoft, Pixar, Dreamworks, Valve, Oculus, and many others.
He tells us that when he was young, while growing up he had no idea what he wanted to do in his life, especially because what he was doing - writing software to draw, in the early 60s! - was criticized by everybody as something useless.
Then he had a stroke of luck, because he met some advisors who directed him on the right way, giving to him the right confidence and self esteem.
In 1967 it was finally completed his first 3D simulation project at Cornell University:
One of animations created by Prof. Greenberg and his team in the '60's at Cornell University
In those years he realized together with his team (in a video of those years some people are shown: prof. Greenberg with 4 other people, some still working, some retired, but still all multimillionaires thanks to their research made in those years) a first animation project made with a computer; it was all pre-rendered, since at the time the computers were not powerful enough to be able to move even a few dozens of polygons in real time.
The video consisted of a fly-through of Cornell University; in a second time, they added a first-person part where the camera goes inside a tram (the video enters the tram, in this one runs a road for a while). It was a revolution. His team gained many enthusiastic feedback from Universities, companies and, of course, also they founds lots of funds in order to continue their experiments: their lives changed because thanks to these studies.
Today Prof. Greenberg is involved in the field of computer graphics and technology in general: at the moment they are studying virtual reality technologies and the difficulty to make realistic views: the main problem, at the moment, is that the human eye's resolution is about 432 megapixels while - to give a concrete example - a fullHD definition (1080p) is only 2.1 megapixel. The gap is really wide.
The biggest problem, as many of you know (because video games teach us something!) is rendering in realtime scenes for two eyes, at a good frame rate: 2 images (1 per eye) calculated 60 time each second. Today there are no technologies able to support this number of realtime calculations at a good resolution (today we speak of “VR Certified devices”, when a device can run realtime world at a 4k resolution… very far from 432 megapixels of the eye’s resolution).
The only way to do this today, he says, is using eye tracking algorythms: some studies have shown that while the eye observes a particular point, everything else around him is "black" (unseen). Then, tracing the movements of the fovea, we can figure out what part of the screen is observed by the user and render just that.
Current technologies cannot do this in an efficient way, because the hardware requirements are very expensive. However, the evolution of technology goes so fast that this (or other possible future solutions) will be available to the masses already within a few years.
The latest forecast of Prof. Greenberg is that VR displays in the future will have to be transparent, a bit like Magic Leap, or Google Lenses.