Question: How did you initially start working on ways to improve 3-d graphics?
Euclideon invented a new way to run unlimited amounts of point cloud data in real time. Normally 3d graphics are made out of flat shapes called polygons. Point cloud data is where you make everything out of little atoms. Other systems have been made that can run 3D objects made from atoms, but our system is many times faster and can run atom based graphics in unlimited quantities, We have been using 64 atoms per cubic millimeter so our graphics are of a much higher detail than most people are used to.
Question: So there are never enough polygons
The number of polygons in computer games has been increasing by about 25% per year, but games are severely constrained by limited polygon budgets. Even with powerful GPUs, there simply isn't enough processing power available to do realistic 3-d worlds using polygons. So different approaches were examined to do 3-D. I was examining this problem, when I came up with the idea of using "atoms" to render 3-D objects instead of polygons. This idea is currently done using software, and allows for virtually limitless detail in objects.
Question: How is Euclideon getting funding for the project?
Australia has a very large technology company called Mincom, the head of Mincom sold his shares a few years ago and was looking for something new and interesting, he and one of the Mincom directors put in some money here and became my directors, so I instantly got the best board for technology in the country. We also received one of the largest government grants in 2010 after a long assessment process. We know have 23 people and a very nice office.
Question: How are atoms different than voxels?
The word "voxel" means different things to different people. There is a "voxel system", but that system is different than ours. If when you say “voxel” you mean a little floating atom in computer space then yes we use voxels but we use unlimited numbers of them.
Question: Although your demo is quite impressive, there are claims that it reuses models to keep memory requirements reasonable.
No, we made our demo in 3 weeks, so we only reused models in order to speed production. But since then we have been working with large laser scanned areas where everything is unique.
Question: But how do you create an environment with large numbers of highly detailed, unique 3-d objects without exorbitant memory requirements?
Regarding the memory, If we were making our world out of little tiny atoms and had to store x, y, z, color etc… for each atom, then yes it would certainly use up a lot of memory. But instead we've found another way of doing it. I could say we use less memory than what the current polygon system uses, but if I did that I think I'd exceeded my quote of unbelievable claims for the day. So well leave that for future demonstrations.
Question: Have you looked into the benefits of transitioning from software rendering to GPU acceleration?
We have examined that, but we are still seeing substantial improvements just on the software side. So there is currently no compelling need to use GPUs at this point.
Question: Given the power of GPUs, it isn't clear to me how a CPU only solution could be so much better than a GPU accelerated solution.
I'm not predicting the demise of GPUs, and I'm sure that in the future we will incorporate GPUs. Most of our technology is at this point proprietary, so I cannot divulge many details.
Question: Do your models use any polygons?
Our hope is to get 100% conversion from polygons to unlimited detail, and we are on track to do that. So we want to have the highly detailed, well-lit, movie-quality models that you would see in a movie such as Avatar completely converted to atoms. We do have recycling techniques to reduce memory usage, so memory requirements will not be exorbitant.
Question: Does this technology have applications outside the games industry?
Absolutely! We have been approached by many other industries, including the architecture industry and the graphics industry. More than a few industries need to run laser-scanned data capture from an airplane or car. So we had to decide on which tasks we wanted to divert our limited resources, and we decided to allocate half of our resources to these tasks. So this has resulted in a delay or two in the games department.
Question: Does atomistic detail represent the future of gaming and graphics?
I see the gaming industry as being transformed over the next five years due to a number of technological developments. I think there is consensus that the future of 3D graphics is to make them out of atoms rather than polygons; we have just brought that future a lot closer. I don't see the transition from polygons to atoms being abrupt; I think that atoms will first be used in backgrounds. Bit by bit, characters and models will be replaced by atoms, and over the next few decades I think we will see polygons retire after their years of loyal service to us all. Then after 20 years have passed they can make a popular reappearance as a beloved “retro” style like blocky graphics do in Minecraft ;) .
Question: What demos/products do you plan on releasing during the next couple of years?
I know that gamers won't like hearing this, but the first commercial releases of unlimited detail will not be game related. Games, however, won’t be too far behind.
Question: Could atoms be used to make computer-generated images for movies and trailers?
The movie industry is definitely interested in this technology. They would like to reduce rendering times as much as possible. I honestly don't know how much of a role this technology will play in future special-effects, but I can say that atomistic rendering has generated considerable excitement in the entertainment industry. At this point the technology has so many potential applications, including neuroscience, architectural visualization, cgi, etc., that we had to make difficult decisions as to how to allocate limited resources.
Question: Is Euclideon seeking venture funding?
We have been contacted by various companies, but at this point we simply don't need any more money. The more outside funding we receive, the less control of the company we have. Given the potential of this technology, I want to keep as much control over the company as possible. We are currently well-financed, and by the time the money runs out, we will be selling products for revenue.
Question: When will Euclideon emerge from stealth mode?
We will release a non-gaming product in 2012. Regarding games, we will probably remain in stealth mode for a while longer, since we don't want to release the technology before it has been properly developed and tested.
Question: Are there any plans to take Euclideon public?
I have thought about that, but at this point we don't see it as a likely scenario for the foreseeable future.
Question: What will be the long-term affect of this technology on society?
I think the line between the real world and the virtual is starting to blur. I am not sure that is the best thing for humanity, but it seems to be the direction we are all headed in. I think laser scanning and photogrammetry brings things from the real word to the virtual world, 3D printing brings the virtual things to the real world, augmented reality overlaps the two, and Euclideon technologies allow you to handle the huge amounts of atomic data that all of this is creating.
If you liked this article, please give it a quick review on ycombinator or StumbleUpon. Thanks