PDA

View Full Version : Panel Discussion: Has 3d gone far enough?


Array
08-08-2002, 08:42 AM
I saw this link on xsibase:

http://www.patmeier.com/pieros/siggraph/Siggraph02.htm

Grey
08-08-2002, 08:55 AM
doh... you have GOT to be kidding me...

The potential medical aplications are still years away...

the discussion was a simple publicity stunt, right?

CitizenVertex
08-08-2002, 05:04 PM
Originally posted by Grey
doh... you have GOT to be kidding me...

The potential medical aplications are still years away...

the discussion was a simple publicity stunt, right?

The point to the discussion if you looked at the presentation isn't in application of 3D it's a matter of whether we are at the point where continued research in the field will yield results worth the investment of time and money. We are currnetly at the point where we can simulate all matter of natural phenomena to create synthentic images that are indistinguishable from a photograph by the human eye. The only problem is we can't do it real time, but hardware is quickly catching up. This year we are on the cusp of a huge breakthrough in affordaple real time rendering with new technoloogy from ATI and Nvidia. Very shortly we will be able to produce cinematic quality visulas in realtime on the lowest end home PCs. Once we've reached that point is there any reason to continue development?

Grey
08-08-2002, 11:25 PM
We are currnetly at the point where we can simulate all matter of natural phenomena

I beg to differ. Simulation for the purpose of presentation and simulation for the purpose of research two very different things.

If you're talking about simulation for visual demonstration sure, whatever... that's the least of the potential, and it's been achieved...

Simulation for the purpose of research: we've but scratched the surface. Most who frequent this forum probably never even think about it.

Analysis programs that calcuate strudtural integrity of materials on a 3D model are still only at the very high end. Medical Research and 3D is something so obscure to the standard 3D artist, he or she has probably never even heard of the very basic aplication of it... and the hope for the future which won't come for several years.

When are we going to be using 3D to calculate the weather? First we need an accurate model of the antmosphere, then we need to be able to simulate it and that has yet to begin!

When these fields come to their peak, the DISPLAY work we're seeing today will look so increadibly primitive that if we even remember this conference occured, it will be laughed at, as I pointed out, as some kind of idiotic publicity stunt...

Array
08-09-2002, 12:06 AM
i think that the next step in rendering technology will be the simulation of light as a wave, and not a particle (as seen in photon mapping renderers). Henrik Wann Jensen, where are you??

Black&White
08-09-2002, 12:26 AM
HOME MOVIE BEGIN BEFORE BIRTH:applause:
i saw a ultra sound home device. which can generate 4d imagery of mother's abdomen and latter other parts too.

the result is 3d imagery

whats 4th d
it is sound.

I found this artice in StarComputing mag.

developed by GR Medical Systems
product : Volusion 730 4D Ultrasound Sytem.

Array
08-09-2002, 12:45 AM
4D? in terms of physics the fourth dimension is time. if you see an object in 4d, you are seeing it before it's creation, during it's existance, and after it's demise all at the same time. humans, naturally, cant accurately visualize this. for more information, read "Flatland". it's a physics classic :)

Black&White
08-09-2002, 12:53 AM
oops i read wrong 4d is time. But time is known with help of sound bouncing.:)

CitizenVertex
08-09-2002, 05:36 AM
Originally posted by Grey
If you're talking about simulation for visual demonstration sure, whatever... that's the least of the potential, and it's been achieved...

And that's all this panel was about. The discussion of whether further research into 3D graphics is necessary because we can already approximate reality to a stunning degree VISUALLY on computers. Sure they could continue on and simulate realisticly the effects of gravity on light and implement light as a wave, but they are asking themselves "Is it even necessary?" The hardware is coming to the point where we will be able to display 64 bit images with obscene geometry counts and realistic lighting and shading real time. Why continue research into computer graphics beyond that? If you can do it realtime or better you've hit the cieling. There is no where left to move.

Originally posted by Grey
Simulation for the purpose of research: we've but scratched the surface. Most who frequent this forum probably never even think about it.

Analysis programs that calcuate strudtural integrity of materials on a 3D model are still only at the very high end. Medical Research and 3D is something so obscure to the standard 3D artist, he or she has probably never even heard of the very basic aplication of it... and the hope for the future which won't come for several years.

When are we going to be using 3D to calculate the weather? First we need an accurate model of the antmosphere, then we need to be able to simulate it and that has yet to begin!

When these fields come to their peak, the DISPLAY work we're seeing today will look so increadibly primitive that if we even remember this conference occured, it will be laughed at, as I pointed out, as some kind of idiotic publicity stunt...

I don't really understand what this tangent you are going on is about. People already do all these things. What do you think Argonne and FermiLab are doing with these world's fastest Super Computers containing hundreds of CPUs? How do you think they come up with your nightly weather forecast? How does Volvo design a safer car? How does GE build a system that allows surgeons to operate remotely on highly intectious patients via a robot? They use computers to do very precise 3D simulations. You can't cheat physics simulations for medicine or the environment if your resulting data is to be of any use. If you want to be able to calculate global warming projections or map Genomics on your home PC accurately, sorry you're going to have to wait another decade or two. Lighting and Geometry is easy because it doesn't have to be accurately caluclated. 3D graphics just have to look correct when we see the final image. How we arrive there makes no matter. Scientific 3D modeling for anything is a totally different proposition. It MUST be precise. The mathematics to do realistic wheather and physics simulations on a computer all exists already. They blow up building and airplanes on super computers all the time to test integrity and do accurate physics models, we're just waiting for hardware to catch up.

Can we calculate accurately the effects of a Nuclear War on Global weather patterns in real time? No. But we can just about render a visual representation of it in real time.

eskil
08-09-2002, 02:51 PM
As a researcher i must say:

Research is not about solving problems, its about learning more, engineering is about solving problems by using the knowledge we got from research. Until the researchers have gathered the knowledge, we don't know if the engineers can make use of it. There for we must always research we simply don't know what it may lead to.

As an artist i must say:

That was one ugly web page.....

E

Grey
08-09-2002, 07:59 PM
eskil, you're looking for problems to solve? :D

I don't agree that research is driven by pure curiosity. I know biotech companies would take issue with that.

But I digress:

CitizenVertex: the point I'm trying to make is that the art of engineering and the art of artistic production are converging with 3D. Even if we're talking only about Character animation, it's still about simulating real dynamics. It doesn't take in-depth knowledge of 3D to know that we've barely scratched the surface.

When you're hardware and software is capable of accurately simulating the real world (not this century, but we'll be able to do it in portions) functions of entertainment and real-world development lines will be so blurred that they'll be virtually indistinguisable.

Whatever visual representations we can create now is largely interpretation.

What happens when a 3D Scanner willl be able to measure a face and accurately bone it for muscle function simulation? It is possible even with today's technology, but will likely require much heavier hardware capacity.

When that happens: does this not have both medical and entertainment applications? I submit that it does.

Adam
08-10-2002, 06:36 PM
hey this is a great interesting debate. keep it up.:beer:

Grey
08-10-2002, 07:40 PM
yeah, but someone has to call me an idiot and make me look like as much first :D

CitizenVertex
08-10-2002, 08:33 PM
Hey now, I don't know if that was in reference to my post, but I in no way meant to infer you were an idiot. I'm just an abrasive person some times:beer: I need to work on making points tactfully. I was just trying to clear up any misconceptions about what the panel was debating.

Peace, Grey.

Grey
08-10-2002, 08:36 PM
No worries CV! :D

I'm a very hard line debater and I pull no punches...

Peace to you too, my friend.

cdinic
08-11-2002, 04:45 PM
has anyone seen any animation of a tree swaying in the brezze as if a storm was rolling in? this is a simple, common event in real life I have never seen re created on a computer to any level of reality. I think such and animation is years away.

how about a simple talking head. I have never seen a syntheticly created facial animation that my little 10 year old cousin couldn't point out as fake.

these are two example of very common event that are far from being achived syntheticly

how can we even think about stopping development

-Chris

genesis max
08-12-2002, 12:26 AM
Has 3d gone far enough?
Nah! We always want more :arteest:

CGTalk Moderation
01-13-2006, 02:00 PM
This thread has been automatically closed as it remained inactive for 12 months. If you wish to continue the discussion, please create a new thread in the appropriate forum.