Slashdot Mirror
Matrix-Like VR Coming in the Near Future?
Anonymongoose writes "A researcher at Brookhaven National Lab reckons it could be just a few years before computers can pass through the uncanny valley. The article refers to this as a 'Graphics Turing Test': 'a computer can be considered intelligent if it can create an artificial world capable of fooling a person into believing it is the real thing.' Michael McGuigan has been performing some interesting experiments using Brookhaven's Blue Gene/L supercomputer and has shown that it can produce realistic lighting effects in real time. McGuigan's original research paper (pdf) is available online."
But how are we going to fit a full VR.8 onto an 8" floppy?
...welcome our new Matix overlords.
Future Tech Prediction Checklist:
"Researchers" did or said something: x
"A few years" before the tech is out: x
Promises to change the way we think of computers: x
Shitty PDF "research paper" that was probably written by a half drunk college kid: x
I had a couple of hundred television channels, and canceled my satellite service because there was never anything worth watching on.
Having a realistic world doesn't impress me. I'm holding off to see what they do with it before getting excited.
...I prefer 2D games.
Circumcision is child abuse.
If you can't build a large hadron supercollider in the game and get new insights into particle physics, in real time, then it fails the test. This is NOT near future.
Music speeds up when you yawn, but does not change pitch.
Gimme an 'O', gimme an 'R', and gimme an 'N'!
It won't have to fool me into believing it's the real thing; I WANT to believe. I'm quite willing to ignore some gaping holes in any VR representation (but not others, nudge & wink).
(In fact my "Top Ten" List would contain more than a couple of anime characters)
Sure, using a Blue Gene/L you can run a radiosity simulation on top of raytracing in approximately real time. Big freaking whoop!
But will we have the model and shading tools, not to mention the physics engines and such to simulate a realistic environment in 5 years? 10? 20? Curiously the article fails to investigate this.
Instead they have a nicely shaded clump of colored balls. Maybe they'll do a teapot next!!!
THIS THING CAN TURN ON A DIME, MACROSSZERO STYLE ALSO FUCK BETA, ~NYORON
But having a direct neural interface, that can mimic all five senses at once, is another thing altogether.
(Not to mention being able to do it for hundreds of thousands of people, some of whom might be spaced out all over the world, with no appreciable lag... Oh, and having many separate strong AIs all running on the same hardware...)
Have you been touched by his noodly appendage?
What the heck are they going to be rendering? I mean, i keep hearing how it'll be ultra-realistic...but will people suddenly start rendering things like never before? Excuse me for my incompetence though.
Off topic rant..
Ok. So, this new discussion system is kinda half-cocked. I wanna revert it to the good ol' html click boxes. Where the heck is the option under my preferences?
import system.cool.Sig;
Blue Gene/L possesses 18 racks, each with 2000 standard PC processors that work in parallel to provide a huge amount of processing power â" it has a speed of 103 teraflops, or 103 trillion "floating point operations" per second. By way of comparison, a calculator uses about 10 floating operations per second....
He found that conventional ray-tracing software could run 822 times faster on the Blue Gene/L than on a standard computer.... This allowed it to convincingly mimic natural lighting in real time.
So, does this mean in a few years they'll have a computer that can actually run Crysis with a decent FPS?
....you believe everything you read. You take the blue pill, you're skeptical of such far fetched allegations (and you get a hardon that lasts and lasts).
These posts express my own personal views, not those of my employer
Well supposedly vista sp2 will include support for directx 11 which they claim will support raytracing. I don't know how powerful graphics cards will be by then, but to be honest that seems like kind of a long shot to me.
Weaksauce as they say...
I have a lot of karma to burn* so why the fuck not...
The poster sounds like a pup to use the phrase "Matrix-Like". Back when the Wachowski brothers were in high school, Gibson had already formulated the term "cyberspace" in Burning Chrome, which was a "Matrix-Like" VR before there was even a Matrix. Give credit where credit is due!
* I find people who post something along the lines of "I have a lot of karma to burn" before posting a rant end up getting modded plus points. Let's see what happens!
'a computer can be considered intelligent if it can create an artificial world capable of fooling a person into believing it is the real thing.'
That's true if the user has no senses other than vision. It might look real, but the user will know it isn't , so the "Turing" test will be redundant, right?
Has there been any significant progress on other sensory technologies to compliment this?
I can't wait till they get porn on it!
Hey! Look a Distraction!
Super Computer + Ray Tracing = . . .!!!
This is something that has always made me wonder. When computer graphics reach the point where you cannot readily tell if the image you are seeing is real or synthetic, how will this affect video game violence?
Can you imaging Grant Theft Auto X with full realistic imaging? How would that affect someone when they go beat a whore to death with a baseball bat and the mind cannot as easily dismiss the disturbing imagery as virtual.
Physical properties of the materials in the simulated environment would drastically fall short of reality I would imagine. Simulating a realistic appearing universe seems trivial when you try to make the same universe look realistic when digging through sand, or cupping your "hands" to hold some water. How could you possibly say you have a realistic environment when it falls apart the moment you try to interact with the fluid dynamics of the system? I'm sure it is possible to make it convincing with the appropriate hardware, however I doubt that current (and relatively near future) systems would be able to at once handle the computationally intense task of a lifelike appearance as well as handling the smallest physical properties of objects dynamically interacting with each other.
Take the Crysis engine for example. Although we have the facets of gravity and collision detection seeming lifelike, the impacts of your interaction on the environment are limited by your means of interface: i.e. keyboard and mouse as well as the pre-defined scale of your interaction: i.e. you can't interact with the sand or water directly.
All I mean is it seems a recursively controlled depth of interaction control system on top of a material property and interaction control system that could reproduce realistic environments in a way that would pass this sort of "Turing test" would be computationally infeasible without allowing for loss of generality.
My two cents...
When I played DOOM, I found myself trying to look around the corner of the inside of the computer screen.
It was immersive enough to fool me...
---
I type this every time.
Oh yeah, it was back in the early 90s. Yup, way back then.
-- Boycott Shell
about the matrix. I want to learn how to master kung-fu in a day, and fly a helicopter in a few seconds.
That's real power. Imagine if everybody could know every thing. that means everyone would push new boundaries and the wheel wouldn't ahve to keep getting invented.
The second kick ass thing was the ships.
The Kruger Dunning explains most post on
A researcher at Brookhaven National Lab reckons it could be just a few years before computers can pass through the uncanny valley.
We can use it for the heads-up display for our flying cars (just a few years away) powered by practical fusion (just a few years away) while travling to the clinic for our immortality tratements (just a few years away).
Thank god all the best things humanity will ever invent are going to be practical at the same time (just a few years away).
I remember when I first heard of VR around 1989-90. I'm talking about the big headset with two screens one for each eye drawing a slightly different angle, with it also having head tracking and then draw the screens appropriately.
I thought, what an amazing idea! This seemed like the closest thing you could have to a holodeck (kind of like a holodeck in reverse). Anyway, some games came out in the arcades. One company in particular was virtuality. They had this game called Dactyl Nightmare that I tried a couple of times. It was like a fps where you and a friend were pitted in an arena against each other with a gun. There was also this pterodactyl flying around that would randomly try and grab one of you. Anyway, neat simple idea to showcase VR. Problem was, it was certainly not ready for prime time.
The screens were extremely low res. I mean it seemed lower than 320x240 per screen. But what really ruined the immersion factor was the frame rate. It felt like it was in the teens at best. Most of the time it felt like a slideshow.
Anyway, they had a couple other games at the time, and they were pretty much the same experience.
I still think it's a great idea, just way ahead of it's time. The problem was they were trying to do 3d (on two screens no less) in a 2d world. At that time, I think virtua racing/fighter just hit the scene. Almost all games were 2d still, and most certainly with the consoles/home computers.
I checked their wiki entry just now and there was a sequel to dactyl nightmare which came out about 3 years later that ran on a 486, so I could just imagine what the first ran on.
Anyway, the idea seemed to flop, but I always thought it was an idea ahead of it's time. Certainly we could do two screens at say 640 480 at 60 fps. It's been 16-17 years since I tried this and thought by now the idea would resurface.
The whole Matrix simulacrum spiel is such a load of shite I find it utterly bizarre that people are still entertaining it.
I'm *sure* that the computer will fool some people into thinking what it makes is real, because THOSE PEOPLE ARE STUPID. It's not that the machines will become intelligent, it's that we're bending the curve on what we think is intelligence to something really stupid - we'll just lower the bar, or collectively enter our idiocracy and think "Hey - fooled me!"
"Gee Johnny, why don't you stop drooling on yourself for a minute and tell me: is the machine intelligent?"
"Id da macheen telligent? Duhh YEAH Boss! Id be willy telligent! Can I have cookie now?"
RS
Shoes for Industry. Shoes for the Dead.
Some black guy gave me a choice between three pills, a red pill, a blue pill, and a yellow pill he called acid that let me go inside the computer. It was trippy
Orbis terrarum est non altus satis
My computer generated diffusely-reflecting sphere sitting on a diffusely-reflecting surface looks exactly the same as my diffusely-reflecting sphere sitting on a diffusely-reflecting surface.
"It's all just meme meme around here"
Can you taste it ?
see more here, and GET SERIOUS about advocating for a ROBOTIC WAGELESS ECONOMY in the next 10 years, with the VR RV to smooth out desires...
VR is the new RV
The Future is already here, just unevenly distributed... THE ROBOTIC WAGELESS ECONOMY NOW! http://RoboEco.com/slash
Realistic lighting effects ... immersive virtual reality.
Does anyone else feel like maybe there's a step or two missing there?
sic transit gloria mundi
I'd rather have Tron-like VR. Screw your stylish coats and sunglasses, I want a lightcycle and gridbugs.
As copyright owner of this comment, I authorize everyone to defeat any technological measure which limits access to it.
If we just need a few years of development to get truly photo-realistic images in real time, then why can't why make those realistic images right now in less than real-time? I mean, sure Hollywood visual effects are great, but they are never perfect. And, that's with a zillion artists working day and night to make frames that often take many hours to render when all is said and done. And, when it comes to people, they aren't even great. Crossing the uncanny valley isn't about FLOPS. It's about creating the content to throw those FLOPS at. It's going to take a long, long time before you have the algorithms in place that can simulate, animate, and render a realistic person. Not that it won't happen. It probably will. I just think we may wind up with hardwrae to run those algorithms before we wind up with those algorithms. So, just pointing at hardware advances and shouting is probably a bit misleading.
And it will be ready just in time for Duke Nukem Forever.
Bemopolis
"I guess the moral of the story is, don't paint your airship with rocket fuel." -- Addison Bain
If the requirement is that the interaction with other "humans" in the simulation be realistic, then you've got two components: simulation of human behavior/interaction/conversation, and graphics. And graphics is comparatively trivial, so really, you can ignore that part of the formula.
What are you left with? A regular old Turing test.
Additionally, reality is really, really high res. And let's not forget that the relationship between required processing power and resolution/poly count is non-linear. So even while graphics are the easy part of this over-hyped "Turing test IN 3D!!11"... 3D truly indistinguishable from reality is still a long way off.
You'll know when it comes. Pixar's films will stop looking like cartoons.
Will everything taste like chicken there?
This story reads like a kid bragging about his new video card.
New punctuation update "~" (no quotes) at the end of a line to indicate sarcasm.
That's a fantastic idea.(!)
The graphics in today's games are already pretty damn good. Yet, even with the great graphics we have today, there is no immersive VR available. After 20 years of waiting I still can't go into a shop and buy a VR headset that covers my entire field of view with decent resolution. The best you can buy is something with the field of view of a postage stamp stuck to your glasses.
At the moment all of these great games are still stuck behind a little screen. By now we should really be inside the games. When I am in front of a tall building in a game, I should be able to tilt my head to look up and see how high it is. There doesn't seem to be any drive to bring good VR into the market, yet I'm sure it would sell well. People would love the idea of really feeling like they are inside the game, looking around.
Go Blackwhitek iss .. c om where many are chatting this online, also i met hundreds of cute black and white ladies...
Doesn't it need to pass the normal Turing Test before? A realistic environment is cool, but if it's filled with NPCs whose only lines are "There are many guards in the castle", I'll get bored soon.
was just added to the feature list of Duke Nukem Forever
I get your point but I doubt it'd happen. These are the days when society arrests 17 year olds for having normal sexual relations and they both get charged as being sex offenders.
Virtual worlds are defiantly going to be restricted in the future to what you can and can't do. In fact that am I talking about, they are today. Games like GTA are getting a bad rep for their evil situations.
I disagree, although communication with other people in a 3d environment would be a must if the technology was ever going to go anywhere, I think anyone would be convinced by a room full of interact-able static objects.
No, this is not the Elder Scrolls, its obviously Duke Nukem Forever. I mean they've been working on it for ten years now. Guess what else takes 10 years, FDA clinical trials. Duke Nukem Forever obviously comes with a direct brain neural interface. The game hasn't come out yet because the FDA hasn't approved the surgery to implant the interface.
You'll only knows how Kung Fu and how to fly the Helicopter in the Matrix. In the real world, you will still be only a regular dude. When they learn kung fu, they basically just donwload a hack or enter the appropriate cheat code.
The big virtual penis avatars of Snow Crash.
-- Posted from my parent's basement
They make computers that look like people now?
Wow! Ray Tracing!!! With one virtual photon per pixel!! That's like 10e-100th of what we see every time we look at something. Reality is just around the corner, I can feel it!
Come back to me when someone's figured out real-time radiosity...
http://fantasylab.com/
http://www.geomerics.com/
crap....
Waiter! Red pill, please!
Simulating physical rwality (without intelligent beings in it) does not require intelligence. It does require extreme attentioan to detail, is computationally very demanding, but has no connection whatsoever to the turing test.
Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
So the it will be as realistic as a 2d i*j array? I don't get it.
Field of view is not the problem. It is that it doesn't recognize where your head is. If the screen reacted to you moving your head around it would be better. If you had full field of view goggles that didn't react to the position of your head it would feel crappy and awkward. And if it didn't react perfectly to what reality would do it will make you sick as a dog. For some reason this is reminding me about irreducible complexity... Anyways i'm going to take a guess that position sensing earbuds/headphones need come out first.
I want a bit that floats around and answers my questions.
(C) Kaki Sain, 2011. By reading this, you have illegally copied my property to your brain.
The good news is, it won't be happening in 20 to 30 years, unless something completely astonishing happens. Look at games today - they're a far cry from 8 bit NES games, graphically - but they're fake. They're so, obviously, absolutely, positively *fake*.
...But when we do get there, yes. There's going to be a whole new field of psychobullshit and undoubtedly, a great deal of drugs to deal with it advertised on TV. ;) And I'd expect a lot of laws. I'd really expect some mandated feature for games that pointedly, without question, make it clear that you *are* in a game when you're bashing that prostitute's skull in. And like today, for 99.9% of the population, that's all there'll be to it. The vast majority will be able to distinguish between a game and the real world. And like today, there'll be a very small minority who can't separate the two.
Even when we get to the point of photorealistic graphics (and no - we're not there yet. Maybe with still images, but nothing else...) - there's yet another problem. It's still going to be fake. Look at water. I've yet to see realistic water in a game. Now add in wind. How much computational power is going to be needed to get individual leaves on a tree blowing exactly how they should? Shadows. Lighting. No, we're very, very far from having 'OMG TEH MATRIX!'
BTW, screw the Matrix, I'm waiting for my cyberbrain. Mmm, virtual immortality. And hot cyborg chicks. And thermoptic camouflage. And child-like, talking AI in the form of blue spidery tanks.
*text* games, you insensitive clod!
Tsunami -- You can't bring a good wave down!
I want to learn how to master kung-fu in a day, and fly a helicopter in a few seconds. That's real power.
That's nothing. Now if you could learn how to be a ninja in a day, that's what I'd call REAL Ultimate Power!!!!!!!!!!!!!!!!!!
You just got troll'd!
Yes, games are interactive, and there's that...
But the photorealism has been there in movies and TV for some time now.
If it's going to affect people, the damage is already (being) done.
Oh, and keep in mind... you don't have to beat the whores to death with baseball bats. That's the interesting thing about GTA. It's a tradition that goes back to Ultima -- let the players do whatever they want, even if some of these things might be downright horrible. It's a test of their morality (or lack thereof), and they still have the chance to do the "right thing".
Don't thank God, thank a doctor!
Oh no, they learned kung fu, and they kept that in the real world.
What you don't get to take to the real world is your virtual muscles (which can be as toned as you want), and the ability to do things that are physically impossible. But that's no reason the knowledge would suddenly go away.
Don't thank God, thank a doctor!
Why is Brookhaven National Labs working on this? That's a tax-funded national lab. If they have people and resources to compete with Hollywood, they need a budget cut.
But as for still frames and modelling, we're getting there:
Sexy Girl - http://forums.cgsociety.org/showthread.php?f=121&t=532817
Tattoo Guy - http://forums.cgsociety.org/showthread.php?f=121&t=550192
The Artist - http://forums.cgsociety.org/showthread.php?f=121&t=472843
As for realtime photorealistic animation though, we're a long, long way from there. Lighting is one hurdle, the bigger hurdle is content. Models, textures, rigs... forget rendering, all of this takes a lot of time to BUILD for a photoreal environment.
Its one thing to come up with a realistic model and scene for a photo-realistic still frame. Its another, to rig all of those models so that they can interact with each other in a pre-determined way. Its something altogether entirely different to do this in real time without predetermined paths and choreographed actions, and modelling all viewable elements based upon the degree of movement that a user has within the space.
This is very much highlighted in the differences between high-poly count models (for detailed stills) and low-poly models (used for 3D games). The "art" for immersive environments like simulated 3D gaming (fps, racing sims etc) is to come up with a convincing representation of a real world object with the lowest poly count possible.
Currently the difference between these polycounts is massive.
Alternative theory: This has happened already and you are experiencing it now...
"Not an actor, but he plays one on TV."
"Matrix-like"? Doesn't that kind of imply a totally transparent full-body haptic control system on top of totally realistic real-time 3d graphics? Maybe the latter is possible with today's fastest supercomputers, but somehow I doubt the former is.
Actually, this will permit the first real turing test. The testee must properly emulate all forms of human discourse from body language to an IQ test to (simulated) driving in heavy traffic and deciding who to flip off.
Now that we can make sufficiently realistic worlds for AI algorithms to interact in, we can start using sped-up genetic algorythms to develope strong AI!
If video games influenced behavior the Pac Man generation would be eating pills and running away from their problems.
As someone who has dabbled in making a game or two, I can say for sure that a good virtual reality will never come to light until completely procedurally generated content comes into play. The sheer volume of content required even for an incredibly linear game such as Half-life 2 is huge, with nearly every object in the game being reused multiple times. In a virtual environment that could easily be mistaken for the real world, every single item that a person could imagine would have to be created on the spot. Virtual reality will only work if we have everyone in the world working all the time to create content for it, because that's what takes the real time in creating something.
Oh dear... Imagine the fallout from a fully immersive VR GTA... Jack Thompsons of the future (or perhaps Cyborg Thompson, since I think he'll go to the grave kicking and screaming, like all Christians who've actually read their Bible should) will be calling it a "murder simulator", and to lend credence to the claims, the US Government will make America's Army VR (no playing as Terrorists, so Counter-Strike 7 will still outsell it), and recruit through it. On the other hand, we can put Cyborg Thompson into one of these things, and he need never trouble society again. A tailor made simulation where he always wins, and becomes the Pope should suit him. That, or a sim where he finds out he's really the second coming of Jesus.
Admit it. You post strawman arguments as AC so you get modded Insightful for refuting them, rather than Troll
The visual effects usually don't look so great next to real images. Although the movie Beowulf wasn't all great all the time (actually a lot of it was sub-par, I think), there were scenes when it seemed close enough. Kind of like when you see The Lion King on stage, it quickly becomes a non-issue that everything is a puppet, and the puppeteers are clearly visible. I think if people want to be fooled, they will allow themselves to be fooled. If you go to watch a movie with a preconceived idea that you'll hate it, then you will most often hate it (unless it's a triumph of modern cinema). To people with an open mind about the technology, who will give into it when they want to be in a simulation, it won't matter that it's not quite real. And in the event it is good enough to trick people, they'll have a reverse-placebo effect if they know it's a simulation. They'll see flaws that aren't really there.
Admit it. You post strawman arguments as AC so you get modded Insightful for refuting them, rather than Troll
The really funny thing is watching all you people talk about this as if it is the future.
You have no way of knowing whether you are in a convincing artificial reality right now.
In fact, Hegel - back in the 1830s - already taught that all "reality" is virtual. It is *essentially* appearance. It is all a show, folks. It is meaningless to discuss "real reality versus artifical reality", because there is no absolute distinction between them. They are just "more real" and "less real" in relation to each other.
We philosophers knew all about the problems of virutal reality and knowledge of the world back in the 1600s and 1700s and 1800s - long before computers were invented.
Computers just help the people with no imagination to get the problem a few centuries late.
There. Was that trollish enough?
I am anarch of all I survey.
I really suspect that Strong AI is a myth, and a pipe dream, and that the Turing test is somewhat a hoax. I am rather fond of Searle's Chinese Room thought experiment, on this count. The appearance of intelligence does not necessarily mean intelligence. The turing test just proves that a machine can fool some people, but P.T. Barnum learned this long ago.
Off topic, but in regards to you sig...
While the viewing of child porn might be somewhat like thoughtcrime, the main problem is the CREATION of child porn, which is a real crime, and rather nasty to its victims. People consuming child porn are participating in this REAL crime by creating the demand. Thus people consuming child porn share at least some of the blame over the exploitation of children that the creators are guilty of.
If there is no market, there is no product. This, I guess, is the reasoning, outside of the natural abhorrence regular people have at the whole topic. Personally, even if it borders on thoughtcrime, I support cracking down on BOTH the consumers and creators of child porn. Sometimes this goes to far, but the cause is still just.
As for the terrorism bit, you probably are half right. I think that the government does half of it because we WANT them too, or at least they think we want them too. Also there is a prevalent thought in Washington that we the people are too dumb to look out for our own interests. At times I agree.
Sometimes it has been used for power, fear has, but most the time I'm guessing its misguided "ends justify the means" reasoning (sexified by calling it realpolitik). Also a large bit of social dynamics come in, a lot of people would lose jobs and power if we were safe, and thus unconsciously they try to keep up safe, but not TOO safe. This probably isn't intentional. Viva la Foucault.
A patriot must always be ready to defend his country against his government. -edward abbey
Nope
A myth? A pipe dream?? Your brain, sir(and I know you are a man because of the venue), is turing complete, emlatable by any other turing complete processor. I'll see your Searle and raise you a Dijkstra. I don't care whether it "really understands" any more than I care which part of my brain "understands" what I am typing right now, or even if I "understand" what I am typing now. If I act like I understand than I am fully as human as you, and so is our Turing Complete computer. While I agree that the current Loebner prize is no more than a test of the gullability of the typical person, Strong AI is possible. On other topics: Sure! There is no bank robbing! No such thing at all! We know this because all the publishers of those steamy bank robbing flicks were all put into prison. My point here is not that bank robbing is more or less evil than molesting a child, but that committing a crime and recording the commission are two different things. If child porn weren't illegal, the evil people who make it would be caught a hundred times faster! And for a quick jaunt back to the topic: VR child porn can be created without harming a child. Furthermore, it may be (I think it is the case) that those who buy the disgusting stuff would strongly prefer the ethical variety. Legalization would also free the archives so that previously created material could be consumed instead of anything new. Many of us pirate movies, and that creates no incentive for the maker to make more... I submit that information itself should never be illegal. too dumb?? I need the government to protect me? You're off the deep end there...
If video games influenced behavior the Pac Man generation would be eating pills and running away from their problems.
A myth? A pipe dream?? I'll see your Searle and raise you a Dijkstra. I don't care whether the machine can "really" think any more than I case, in this context, whether I can think. If I act like a person then you are ethically obligated to treat me like a person. One may as well call any person a mere product of his genes and upbringing, and make a slave of him because he only acts human! So ask yourself, when a machine can object to being turned off, how is that different from you objecting to your own death?
Information should never be illegal. If it were legal to own child porn, the creators could be caught a hundred times faster! Furthermore, consumption, without payment, creates no value to the creator. Ever pirate a movie? Copy an MP3? You do realize that the artists didn't get a penny when you did that... I also submit that VR child porn would be much cheaper and easier to produce, and that many consumers would strongly prefer the ethical kind.
I, for one, would strongly prefer to lose the utility of a few more terrorist incidents than the utility lost by the millions of needless searches, the loss of rights, the minimum wage workers on power trips, and the overly-powerful politicians. If I spend 20 seconds per day thinking about how to drive safer, I'll make myself a hundred times safer than spending the same time worrying about OMG the terrorists.
If video games influenced behavior the Pac Man generation would be eating pills and running away from their problems.
1)Find a bathtub or similar container
2)Place a number of random objects into the container
3)Fill container with soapy water to half-cower the objects in question
4)start stirring the water.
Even if you manage to render that realistically your supercomputer is going to completely choke trying to work out 3D fluid dynamics with surface tension in real time. For extra fun you can throw in some hydrocarbons with a melting point in the vicinity of room temperature, thus forcing the simulation to take into consideration temperature dependent viscosity as well. If you want to really push it you add a pigment that goes transparent above a given temperature and then focus a strong lamp at one end of the container, thus causing the rendering algorithm to have take into consideration how the liquid's color will change with a non-fixed temperature. The temperature variations are of course dependent upon how light scatters in the rather complicated system. In addition, since the heating of the fluid will depend upon how light interacts with the system, the physics of the fluid now depends on the rendering, and the rendering obviously depends on the physics. So all in all you have a system composed of two fluids with different thermal and electromagnetic properties, different densities and viscosity. The density, temperature, color and electromagnetic properties of the fluids vary with time and position and depend on one another as well as on how light propagates through the entire system.
In summary, good fucking luck.
Not even close.
It works for plasticky scenes with lots of mirrors and refractive glass balls but not much else.
No sig today...
So how many petaflops will be needed before tasty wheat tastes like anything other than chicken?
Why would anyone engrave "Elbereth"?
Because raytracers are useful, and there aren't that many free raytracers with good performance, a reasonably complete set of features and support for distributing the workload across hundreds of processors.
That's because they didn't use antialiasing for that teapot render. They do support anti-aliasing, according to the documentation.
Besides, the SPD teapot scene isn't supposed to look awesome, it's supposed to show that your rendering algorithm is functioning properly.
Povray is great, but it isn't particularly fast, and it until the current beta releases they haven't even supported multi-core parallelism in the standard distribution, much less MPI.
Also, tachyon is, according to the changelog, at least 13 years old. Not as old as povray perhaps, but not incredibly modern either.
If you want to see pretty renderings, check out pbrt. If you want to see performance, download a demo of arauna. Both are considerably more modern. If you just want to complain about terrible renderings and pointless projects, then you should check out my ray tracer.
great. They can save $ when they CG the bodies in front of the shagging.
http://www.skybooksusa.com/books/wgordon.htm
Believe it.
As far as the second two quotes - well, hence the comment about pork. I still don't understand why anyone claiming to be a Christian pulls anything out of Leviticus - if you want to quote it, then follow all of it.
Oh, is that a nylon/cotton blend you're wearing?
As far as Paul went - as much as I hate using wikipedia, this is the only site I can find that discusses the issue with the translation from Greek - namely, why didn't Paul use the common Greek word for homosexual. (Not even mentioning Paul's sometimes self-contradictory views on other subjects, from circumsion to marriage & Gentiles following Levitical law.)(Caveat : the last Christian church I attended only accepted parts of the New Testament as inarguable - the Gospels & Acts, specifically. Everything else was up for debate.)
Have you been touched by his noodly appendage?
The Wise adapts himself to the world. The Fool adapts the world to himself. Therefore, all progress depends on the Fool.
Hmm, this is pretty interesting; I never thought about the way things worked there - did they just "patch" the software (so that the function returns TRUE, or whatever the environment expects it to return) or did they actually learn those abilities?
At the same time, you say that the knowledge stays, therefore they can do kung fu in reality, but can they? A lot of knowledge relies on hardware features (ex: moving fast enough). What happens when a matrician wants to do an uber-kick in reality, but realizes they don't have the horse-power for it? "Feature not supported by hardware" is shown on the screen?
The saddest poem
Anyone can learn to throw a punch or kick; these skills are fairly simple. The tricky part of a fighting art is learning when to use them, how to create openings, how to place yourself in a strong position relative to your opponent's weakness.
The human body can be built up remarkably quickly through simple repetition of movement with enough focus. There are more efficient ways, but all that should be required to acquire the physical strength to perform kung fu is knowledge of kung fu.
What happens when a matrician wants to do an uber-kick in reality, but realizes they don't have the horse-power for it? They'd pull something.
I'm not so concerned about the technical issues as I am of the social issues.
I'm reminded of the problems that arose when "The War of the Worlds" was broadcast on the radio and some people thought it was real. That was audio. Then, IIRC, there was a scene in "The Moon is a Harsh Mistress" where the moon colony made up a video of a leader announcing something, but it wasn't real (sorry about the lack of details - I read it a LONG time ago - I'm sure someone here can elaborate/clarify).
Yes, there are still some technological hurdles to overcome in both hardware and software, but at some point I believe it will be possible to generate a scene that is, for all intents, indistinguishable from reality. Then what?
The geek in me can't wait for the day for us to have this power. The human in me fears for the day it comes.
BUT WHEN WILL IT SPARKLE??
Massmarket VR helmets had head tracking over decade ago, it just happened that they haven't sold all that well and got lost and forgotten in computer game history. Today you can buy TrackIR, doesn't have the VR googles, but does all the head tracking for gaming on a normal monitor.
God this annoys me. You know even the most sophisticated simulation comes nowhere near matching the physical world. If people are not familiar with the physical world cos they spend to much fucking time online and not seeing what actually happens then they will be easily fooled. But if you take the time to actually look at the seemingly mundane then you will see things that are never simulated. Because they are subtle but surprising.
I was going to mention some examples but that would be playing into the hands of the simulators. And yeah I've spent plenty of time online, even a sucker for WoW, but it aint the real world.
Don't be fooled fellow /.ers see the real world first.
Bitter and proud of it.
> 'Graphics Turing Test': 'a computer can be considered intelligent if it
> can create an artificial world capable of fooling a person into believing
> it is the real thing.'
"Ahhh, you almost had me fooled! But I am a Slashdotter, and know that no chick who looks like Natalie Portman and Jessica Alba's love child would ever want to haul me into a bar bathroom and give me a hummer!"
(-1: Post disagrees with my already-settled worldview) is not a valid mod option.
So I was looking at the last picture on that site and I decided it wasn't realistic because the man's face didn't contain enough blemishes. Then I realized that when you see a close-up photograph of a person in a magazine, the picture is modified to remove blemishes. People have gotten used to looking at modified photographs of people and accepting that they are real. So maybe fooling people isn't so hard because our notion of "real" has been distorted anyway. All you have to do to make something look "realistic" is to make it look like the modified, idealized, and simplified notion of reality that we have all been conditioned to believe.
Head mounted displays (HMDs) I should know. I work for VR company in Santa Barbara. Modern VR looks a bit like Half-Life 2:
http://www.worldviz.com/solutions/methodist.html
How hard would it be to hook up some TV goggles to an FPS where each eye is its own display, slightly off-center of each other and see if this can partially fool someone? Just being inside any 3D environment, regardless of the detail level at this point "could" be convincing, couldn't it?
Doesn't this part need to be worked out before the photo-realistic part of it needs to be? If I knew the world I was in was fake that's fine, but being inside of it could be very cool I would think. I mean, drop me in any of the latest FPS games with those graphics and if the "head turning" interface, treadmill in all directions (or something like that), frames/sec, and sound worked correctly, it would be a good start.
http://skelman.com
It's actually much simpler than that... and I thought this was obvious to everyone.
Go back and watch the original movie. Neo lies down, jacks in. Tank picks up a disc, inserts it into the ship's computer, and pushes a button. Neo's eyes flutter for a bit, as computer/data sounds are made.
They are downloading the knowledge and experience into his brain. (I'm fairly sure the word "downloading" did appear on screen at this point, too.) Chances are, these are memories of actual training, possibly some wiring of his own chemistry... Think of it as installing software, not "hacking".
Then, after he's spent 12 hours straight downloading stuff into his brain: "I know Kung Fu." "Show me." At which point, he enters an actual training simulation with Morpheus. He already knows kung fu at this point -- Morpheus is teaching him to push beyond the boundaries of the artificial world. This is the "hacking".
These skills don't carry over to the real world, obviously, because the real world is real, and can't be manipulated that way -- at least in the first movie, when things made sense. However, things like flying a helicopter would, if there were any helicopters remaining in the real world -- anyone can learn this skill.
What is learning? It's copying information into your brain, and integrating it with the information you already know. Skills can be seen as software -- and if you see the brain as a computer, then learning facts is just downloading files, and learning a skill is just installing software. Yes, there's a physical component, too -- his muscles in the real world won't be able to do everything he learned just because he learned them -- but building muscle is much easier than training to become a world-class martial artist. (To abuse the metaphor, your body is just the "hardware" that it runs on.)
Don't thank God, thank a doctor!
We're just a few years away from being MatRix-rolled...
The article seems to be using the term "Graphics Turing Test" to mean a test similar in design to the Turing-test, not another test for artificial intelligence. Neither it nor the paper mention artificial intelligence except in defining the Turing-test itself.
Supposing the article did mean "Graphics Turing Test" in terms of AI, it's a ridiculous idea. Physics simulation has nothing to do with intelligence. The only intelligence involved would be in dictating the actions of human (and possibly animal) elements in a way that convinced a third-party observer. And if an AI were capable of passing that, it would be capable of passing the traditional text-only test. It wouldn't mean anything more than the original, in terms of "intelligence".
Please, tell me more!
Except that the skills learned are PURELY at the brain level: interface your brain with the matrix, which is a big piece of software, then flight a piece of helicopter software.
Basically, this is a video game, except that instead of having a finger/gamepad interface, the brain interface with the Matrix through the ships computer.
So instead of grabbing a gamepad and pressing buttons, you jack something in your skull and send brainwaves over.
Its not because I can control an helicopter with a GamePad that I can they flight an helicopter. You could argue however than it is because the gamepad doesnt simulate the real helicopter interface. That is true, and indeed there are flight simulators used to train real pilots, but even in that case, you cant fly the real thing after some practice with the real thing. You might argue that this is because the simulator is an imperfect representation of the real thing, and I would agree.
But now the question is, is flying the helicopter in the Matrix a correct representation of the real world?
The Matrix is designed precisely to be a perfect representation of the real world, so that the brain cannot actually make the distinction between the real world and the Matrix, so if normal people where to learn something in the Matrix simulation, in the same way they would learn in real world, I have no doubt they would remember it when out. This is supported by the fact that Neo still knows how to walk and talk when he is outside. (And I assume here that he learn walk and talk as part of the simulation, interacting with his parents, this is howver an assumption).
But when the "fly the helicopter" program is downloaded or when loads of guns are "created" one can argue that it is not the same learning process as the real world and not the same control mechanism.
This is obvious when you think that a lot of the things the people who are "aware" of the matrix do are not possible in the real world. For example some of the kung-fu moves they use are not remotely realistic, the speed they move is impossible. And i'm not even talking about stopping bullets.
The Matrix is a Software Simulation that a few can control with their brainwaves. I dont see any indication in the movies that the humans actually developed a way to alter their brain to learn new things. They have a way to alter the matrix using their brain, and it is definitely dangerous as some interaction with the matrix could basically "overload" your own brain and kill you.
If the human had invented a way to jack up their brain, they would certainly use it to gain a physical advantage in the real world.