Stereo vision is only noticeable to about 30 feet out.
Is't not as simple as all that of course. The brain is marvelous at detecting differences between the two images, something which is very evident when you look at the ground from the window of a plane during take off. Even with everything much further away than 30 feet, there sill is a pronounced 3D-effect.
The limit depends on what you're looking at basically. All of our senses are "auto calibrating" which is one of the reasons why it's easier to detect changes rather than absolutes. Feeling hot/cold can be a matter of several degrees, but even minute temperature changes are detectable.
Of course there IS a hard upper limit to where our 3D-vision stops, which is why the night sky looks like dots on a hemisphere.
"What I mainly have problems with is their lack of proof that (a) the files really were the copyright works they claim, (b) did the p2p users have the right to have a legal copy? (c) Are they providing an itemized list? (d) Do they have actual authorization to act as agents of the copyright holders?"
I've read some of the original danish articles, so I can help you out with some of these questions.
(a) They didn't download anything so they do not really know.
(b) Even if the users in question owned the material, they did not own the right to distribute it on Kazaa. That's the issue here. Leechers are off the hook.
(c) They have kazaa screenshots. I don't know if the bills sent out are itemized or not.
(d) Yes.
IS evidence based on hacking a network admissible in court?
These goons hacking into your file lists was not a consensual act...
The "find more from same user"-function in Kazaa hardly constitutes hacking. A court order was obtained to get the ISPs to map IP-adresses with users.
I'm quite curious to see if the new improved threading will help Java performance at all (especially swing) which because of the event model spawns new threads like there is no tomorrow.
The light issue does not become any less of a problem when shooting digital video. You would still be exposing the image capture device (the CCD in this case) to half the amount of light that you'd have when shooting at 24fps, and you'd still get all the problems that go along with it.
Of course, but CCDs are already a lot more light sensitive and are still getting better.
I agree with all your points on the problems of going digital. It is a large step, and many will delay taking it as long as possible. George Lucas was one of the first to shoot a film completely digital, and is sort of a pioneer in that regard, but it will sure take a while before all films are shot digitally. It will probably be that way with faster framerates too.
People who complain about flicker and suchsort when going to see a movie are probably watching the movie in a shitty theater with a substandard projector.
My main peeve with cinema is the fact that to hide the flickering, and i suppose also to avoid damage to the film, each frame is flashed twice when the film is displayed, but his causes a very disturbing temporal aliasing effect as the motion predictors of the human vision system fail half the time.
The result is that fast panning scenes are almost unbearable to watch. Instead of a smooth sweep it appears as a jittered mess. It's much less of a problem for slow motion, and I suspect the effect may be linked to the eyes trying to track the action on screen. Some people might not be able to see it at all, but everyone I've explained it to have been able to see it, and of course they all hate me now.:)
With 48 real frames per second and no double shutters, this problem would dissappear completely. To give the illusion of perfectly smooth motion it is a requirement to have a new frame for each screen exposure.
The points you make about exposure times and amount of light are valid, but as more film makers move to digital cameras, this becomes less of a problem.
What's eatin' you, pal? Who's twisting your arm to buy Doom3? "Poorly desgined new graphics features", and I suppose you can do better?
I'm pretty confident that Id will release something before the game is out, as they always have in the past. You'll be able to test the finished thing and then decide if you want to buy the game or not. I cannot envision why it should bother you the least that people have high hopes for Doom3 and are eager to try it out, unless you're upset because your hardware can't keep up.
Something a wise man once told me: No matter how smart you are, never make fun of people who are happier than you are. They must be doing something right.
On TNG, you had occasional sexual tension between Picard and Dr. Crusher, but that was generally very subtle. On Enterprise, though, we're treated to Archer's bizarre sexual fantasies in which he and T'Pal basically fuck in the Detox chamber.
That's because Enterprise is not only less PC than TNG, although it is more mainstream, and because of the fact that they deal with real people, not pillars of morality and wholesomeness. The people on TNG act like they have no sexuality at all. One may start to wonder if that replicated food REALLY contains ALL the required nutrients for proper hormone production.
Then again, the first Enterprise has no holodeck. The crew on TNG can probably use that to relieve themselves, not that they'd ever show it on screen. The closest they've come is showing Vulcan foreplay which really can't count for obvious reasons.
Seems to me that something that you're not going to keep forever is best kept on the hard drive.
I can't see myself having an archive of video that I at any given time would want to overwrite that wouldn't fit on an 180GB HD with MPEG4 compression.
Given the price of a CD-R it seems a bit pointless to overwrite things.
I'm pretty sure this COULD help vnc performance, if not at the client side, then at the server side, which often spends quite a bit of CPU fetching those pixels.
Picasso, Michael Angelo, Da Vinci, Van Gogh, Rembrandt....
None of them knew what a paint brush was until they got interested in painting.
They were also geniouses, not just some guys who thought painting something would be cool. (And it's Michelangelo)
Re:What all 3D games are really missing
on
New DOOM III Shots
·
· Score: 1
Physics is infact the next big thing in gaming.
There are several real time physics libraries out there which several game companies have licensed. Expect to see them in action in Deus Ex 2, and Doom3 (But I suspect Ids making their own phyisics package, at least I haven't heard that they've licensed one).
The first things we'll see are objects falling over in a realistic way. Often referred to as rigid body dynamics, which is exactly what it sounds like.
Ragdoll physics for bodies which will now tumble violantly down staircases, rest against walls instead of falling into them, etc. Realistic particle physics that will let steam and smoke move and be affected by the turbulence caused by the characters or missiles (seriously simplified, but still pretty darn cool)
All these features are already in the new Unreal Engine. Also expect to see some soft body dynamics like cloth, rubber and jello for the next iteration of physics engines. The features are there already, but are currently a bit too computationally intensive for todays games.
I hope the anti-gun people can see that restricting guns is the equivalent to the gun people trying to restrict the ownership and use of inflated pigskins, or golf clubs (or as they are known on the street, cop clubbers). It's just a piece of sporting equipment in many cases.
How often do kids kill themselves or other kids by the accidental misuse a of golf club or football?
Reflection maps are barely correct when you're mapping a sphere and when the objects reflected are infinitely far away, and they become increasingly inaccurate as objects get closer to the object that is supposed to be reflecting them.
That said, NVidia didn't buy Exluna for no reason, and that a rib->nv30 thingy exists or is on the drawing board is a pretty safe bet, but these images where not generated that way.
Would you PLEASE stop talking about stuff you obviously know nothing about? There is no PGA in the TNT. A bios/microcode update cannot make up for the lack of vertex and pixel shader silicon. In fact: you cannot even emulate the vertex and pixel shader path in software because there is no way of inserting it into the correct rendering path on the TNT2.
You cannot emulate rendering 16 textures at once by rendering several times either because there is not enough framebuffer alpha accuracy to do it.
You're living in a land of make believe, with elves and fairies and little frogs with funny green hats!
Cg looks like that announced so far for OpenGL2 - could nVidia just be trying to repeat OpenGL2 functions using their own identical and properitary Cg extentions instead?
That would be strange indeed, as NVidia is one of the major reasons why OpenGL is still alive. I'm guessing they are just waiting for OpenGL2 to play catchup.
I don't see how this compares to Glide in any way. Glide was irreversably linked to the hardware, as it was a paper thin API just consisting of some header files with function declarations that plugged straight into the hardware registers of the Voodoo-part.
You basically can't get any more low level, a design choice that made sense at the time where there was no 3D-API to speak of on the Windows platform, and CPUs were too slow to do fancy things in software.
Cg is almost the exact opposite. A generic language that compiles into hardware specific code. As more GPUs are supported by other compilers, it becomes easier for developers to write shaders that work everywhere. It's much like a C-compiler in the way that it lets you ignore some, but not all, of the HW-spesific issues.
Cg-compilation is fast and the compiler is small, so one could imagine pixel-shader source code being included in the product/game/whatever and compiled for the specific hardware just before runtime.
I always get annoyed whenever anyone talks about Glide like it's the best 3D API ever made. Claims like this is usually comming from people clinging to their ageing V5 pretending that it's still cutting edge.
In reality it's barely an API at all. It is a set of header files with function declarations that plugs data straight into the HW. It's like calling C64 basics poke and peek functions a machine code api. You have low level access, sure, and thus things are about as fast as they can be, but you have to do EVERYTHING yourself. Glide implementation and hardware was so tightly linked that the one stifled development of the other, holding 3DFX back in the feature department. Even 3DFX said that Glide was on it's way out when the V5 was released because of that fact.
You're getting things confused. Carmack bitched about the GF4 MX-series not having any pixel-shader and vertex-shader functionality, and recommended people not to buy those cards, at least not for playing doom III.
The GF4 Ti-series is a different matter. They are basically souped up GF3s and will be able to run Doom III just fine, at least faster than GF3s.
Still, I'd wait to see whatever card is best when Doom III is released.
Which is what was wrong with the sierra-games to start with. LucasArts understood that, and didn't
just jump onto the bandwagon of graphical adventure games, they kicked Sierra off it.
Sierra has yet to figure out that killing off the
player whenever he takes a wrong turn is what
killed them.
Have you ever wondered why, if our eyes are an analog interface, we see the wheels on vehicles appearing to rotate in the opposite direction they should? I wonder why, and I've never found a satisfactory answer.
Because you're either watching them on TV, or there's a strobing lightsource nearby. And virtually all electrical lightsources do strobe because of the alternating current.
Why not limit games to 28 or 30 fps and work on building things like accurate motion blurs?
Becuase 30fps is not enough to eliminate flicker. This might be possible on non CRT-screens, but 85fps would be better, past this point motion blur would be more interesting.
Motion blur is an aliasing problem, namely temporal aliasing not unlike the spatial aliasing people usually talk about, and they can be solved in much the same way by using supersampling or multisamping methods.
The design of the 3dfx allowed the cards to do temporal and spatial supersampling at the same time. non-motion blurred objects would just shift slightly (half a pixel) to create spatial antialiasing while fast-moving objects would be placed in an intermediate position between the last frame and the destination for the next frame, and result in temporal-antialiasing. Of course, there would be no spatial antialiasing for those fast-moving objects, but what wasn't really needed.
The irony here is that antialiasing be it temporal or spatial will usually require just as much oomph from your graphics card as increasing the resolution and framerate will require, which will create _real_ detail instead of faking it, using the limits of the human vision to create the motion blur instead, so bottom line is this: It is better to increase resolution than use antialiasing as long as the monitor can keep up, and even the fastest of graphics cards can't push anything remotely photorealistic at 1600x1200 at 85fps. If you've maxed out your display, and you've still got fillrate to burn, spatial and temporal antialiasing is more worthwhile.
Slashdot Mirror
aside from amazing progress in special effects (The Matrix).
The Matrix didn't have a single innovation in special effects, it just popularized a few.
Isn't "always on" and "not worrying about how much time you spend" pretty much the same thing?
It's the absence of minute-rates which makes "always on" practical. I can be always on with ISDN, but it'll cost me.
Stereo vision is only noticeable to about 30 feet out.
Is't not as simple as all that of course. The brain is marvelous at detecting differences between the two images, something which is very evident when you look at the ground from the window of a plane during take off. Even with everything much further away than 30 feet, there sill is a pronounced 3D-effect.
The limit depends on what you're looking at basically. All of our senses are "auto calibrating" which is one of the reasons why it's easier to detect changes rather than absolutes.
Feeling hot/cold can be a matter of several degrees, but even minute temperature changes are detectable.
Of course there IS a hard upper limit to where our 3D-vision stops, which is why the night sky looks like dots on a hemisphere.
"What I mainly have problems with is their lack of proof that (a) the files really were the copyright works they claim, (b) did the p2p users have the right to have a legal copy? (c) Are they providing an itemized list? (d) Do they have actual authorization to act as agents of the copyright holders?"
I've read some of the original danish articles, so I can help you out with some of these questions.
(a) They didn't download anything so they do not really know.
(b) Even if the users in question owned the material, they did not own the right to distribute it on Kazaa. That's the issue here. Leechers are off the hook.
(c) They have kazaa screenshots. I don't know if the bills sent out are itemized or not.
(d) Yes.
IS evidence based on hacking a network admissible in court?
These goons hacking into your file lists was not a consensual act...
The "find more from same user"-function in Kazaa hardly constitutes hacking. A court order was obtained to get the ISPs to map IP-adresses with users.
You realize java does bounds-checking on EVERY array access?
This is incorrect. In many cases, typically in loops, hotspot will limit bounds-checking on
all but the first and last iterations.
I'm quite curious to see if the new improved threading will help Java performance at all (especially swing) which because of the event model spawns new threads like there is no tomorrow.
The light issue does not become any less of a problem when shooting digital video. You would still be exposing the image capture device (the CCD in this case) to half the amount of light that you'd have when shooting at 24fps, and you'd still get all the problems that go along with it.
:-)
Of course, but CCDs are already a lot more light sensitive and are still getting better.
I agree with all your points on the problems of going digital. It is a large step, and many will delay taking it as long as possible. George Lucas was one of the first to shoot a film completely digital, and is sort of a pioneer in that regard, but it will sure take a while before all films are shot digitally. It will probably be that way with faster framerates too.
"Shown in 48fps where available".
People who complain about flicker and suchsort when going to see a movie are probably watching the movie in a shitty theater with a substandard projector.
:)
My main peeve with cinema is the fact that to hide the flickering, and i suppose also to avoid damage to the film, each frame is flashed twice when the film is displayed, but his causes a very disturbing temporal aliasing effect as the motion predictors of the human vision system fail half the time.
The result is that fast panning scenes are almost unbearable to watch. Instead of a smooth sweep it appears as a jittered mess. It's much less of a problem for slow motion, and I suspect the effect may be linked to the eyes trying to track the action on screen. Some people might not be able to see it at all, but everyone I've explained it to have been able to see it, and of course they all hate me now.
With 48 real frames per second and no double shutters, this problem would dissappear completely. To give the illusion of perfectly smooth motion it is a requirement to have a new
frame for each screen exposure.
The points you make about exposure times and amount of light are valid, but as more film
makers move to digital cameras, this becomes less
of a problem.
What's eatin' you, pal? Who's twisting your arm to buy Doom3? "Poorly desgined new graphics features", and I suppose you can do better?
I'm pretty confident that Id will release something before the game is out, as they always have in the past. You'll be able to test the finished thing and then decide if you want to buy the game or not. I cannot envision why it should bother you the least that people have high hopes for Doom3 and are eager to try it out, unless you're upset because your hardware can't keep up.
Something a wise man once told me: No matter how smart you are, never make fun of people who are happier than you are. They must be doing something right.
On TNG, you had occasional sexual tension between Picard and Dr. Crusher, but that was generally very subtle. On Enterprise, though, we're treated to Archer's bizarre sexual fantasies in which he and T'Pal basically fuck in the Detox chamber.
That's because Enterprise is not only less PC than TNG, although it is more mainstream, and because of the fact that they deal with real people, not pillars of morality and wholesomeness. The people on TNG act like they have no sexuality at all. One may start to wonder if that replicated food REALLY contains ALL the required nutrients for proper hormone production.
Then again, the first Enterprise has no holodeck. The crew on TNG can probably use that to relieve themselves, not that they'd ever show it on screen. The closest they've come is showing Vulcan foreplay which really can't count for obvious reasons.
Seems to me that something that you're not going to keep forever is best kept on the hard drive.
I can't see myself having an archive of video that I at any given time would want to overwrite that wouldn't fit on an 180GB HD with MPEG4 compression.
Given the price of a CD-R it seems a bit pointless to overwrite things.
Or if you're running tightvnc.
I'm pretty sure this COULD help vnc performance, if not at the client side, then at the server side, which often spends quite a bit of CPU fetching those pixels.
Picasso, Michael Angelo, Da Vinci, Van Gogh, Rembrandt
None of them knew what a paint brush was until they got interested in painting.
They were also geniouses, not just some guys who thought painting something would be cool.
(And it's Michelangelo)
Physics is infact the next big thing in gaming.
There are several real time physics libraries out there which several game companies have licensed. Expect to see them in action in Deus Ex 2, and Doom3 (But I suspect Ids making their own phyisics package, at least I haven't heard that they've licensed one).
The first things we'll see are objects falling over in a realistic way. Often referred to as rigid body dynamics, which is exactly what it sounds like.
Ragdoll physics for bodies which will now tumble violantly down staircases, rest against walls instead of falling into them, etc. Realistic particle physics that will let steam and smoke move and be affected by the turbulence caused by the characters or missiles (seriously simplified, but still pretty
darn cool)
All these features are already in the new Unreal Engine. Also expect to see some soft body dynamics like cloth, rubber and jello for the next iteration of physics engines. The features are there already, but are currently a bit too computationally intensive for todays games.
Some links.
Havok
Open Dynamics Engine
Mathengine
I hope the anti-gun people can see that restricting guns is the equivalent to the gun people trying to restrict the ownership and use of inflated pigskins, or golf clubs (or as they are known on the street, cop clubbers). It's just a piece of sporting equipment in many cases.
How often do kids kill themselves or other kids by the accidental misuse a of golf club or football?
Reflection maps are barely correct when you're mapping a sphere and when the objects reflected are infinitely far away, and they become increasingly inaccurate as objects get closer to the object that is supposed to be reflecting them.
That said, NVidia didn't buy Exluna for no reason, and that a rib->nv30 thingy exists or is on the drawing board is a pretty safe bet, but these
images where not generated that way.
Would you PLEASE stop talking about stuff you obviously know nothing about? There is no PGA in the TNT. A bios/microcode update cannot make up for the lack of vertex and pixel shader silicon.
In fact: you cannot even emulate the vertex and pixel shader path in software because there is no
way of inserting it into the correct rendering path on the TNT2.
You cannot emulate rendering 16 textures at once by rendering several times either because there is not enough framebuffer alpha accuracy to do it.
You're living in a land of make believe,
with elves and fairies and little frogs with funny green hats!
Cg looks like that announced so far for OpenGL2 - could nVidia just be trying to repeat OpenGL2 functions using their own identical and properitary Cg extentions instead?
That would be strange indeed, as NVidia is one of
the major reasons why OpenGL is still alive. I'm guessing they are just waiting for OpenGL2 to play
catchup.
I don't see how this compares to Glide in any way.
Glide was irreversably linked to the hardware, as it was a paper thin API just consisting of some header files with function declarations that plugged straight into the hardware registers of the Voodoo-part.
You basically can't get any more low level, a design choice that made sense at the time where there was no 3D-API to speak of on the Windows platform, and CPUs were too slow to do fancy things in software.
Cg is almost the exact opposite. A generic language that compiles into hardware specific code. As more GPUs are supported by other compilers, it becomes easier for developers to write shaders that work everywhere. It's much like a C-compiler in the way that it lets you ignore some, but not all, of the HW-spesific issues.
Cg-compilation is fast and the compiler is small, so one could imagine pixel-shader source code being included in the product/game/whatever and
compiled for the specific hardware just before
runtime.
I always get annoyed whenever anyone talks about Glide like it's the best 3D API ever made. Claims like this is usually comming from people clinging to their ageing V5 pretending that it's still cutting edge.
In reality it's barely an API at all. It is a set of header files with function declarations that plugs data straight into the HW. It's like calling C64 basics poke and peek functions a machine code api. You have low level access, sure, and thus things are about as fast as they can be, but you have to do EVERYTHING yourself. Glide implementation and hardware was so tightly linked that the one stifled development of the other, holding 3DFX back in the feature department. Even 3DFX said that Glide was on it's way out when the V5 was released because of that fact.
You're getting things confused. Carmack bitched about the GF4 MX-series not having any pixel-shader and vertex-shader functionality, and recommended people not to buy those cards, at least not for playing doom III.
The GF4 Ti-series is a different matter. They are basically souped up GF3s and will be able to run Doom III just fine, at least faster than GF3s.
Still, I'd wait to see whatever card is best when Doom III is released.
Which is what was wrong with the sierra-games to start with. LucasArts understood that, and didn't
just jump onto the bandwagon of graphical adventure games, they kicked Sierra off it.
Sierra has yet to figure out that killing off the
player whenever he takes a wrong turn is what
killed them.
If you do it once, it's a parody.
If you do it twice...
actually, my wireless optical mouse works for like 4 months on two fresh double A batteries
But it wouldn't last a week with an internal fan.
Have you ever wondered why, if our eyes are an analog interface, we see the wheels on vehicles appearing to rotate in the opposite direction they should? I wonder why, and I've never found a satisfactory answer.
Because you're either watching them on TV, or there's a strobing lightsource nearby. And virtually all electrical lightsources do strobe because of the alternating current.
Why not limit games to 28 or 30 fps and work on building things like accurate motion blurs?
Becuase 30fps is not enough to eliminate flicker. This might be possible on non CRT-screens, but 85fps would be better, past this point motion blur would be more interesting.
Motion blur is an aliasing problem, namely temporal aliasing not unlike the spatial aliasing people usually talk about, and they can be solved in much the same way by using supersampling or multisamping methods.
The design of the 3dfx allowed the cards to do temporal and spatial supersampling at the same time. non-motion blurred objects would just shift slightly (half a pixel) to create spatial antialiasing while fast-moving objects would be placed in an intermediate position between the last frame and the destination for the next frame, and result in temporal-antialiasing. Of course, there would be no spatial antialiasing for those fast-moving objects, but what wasn't really needed.
The irony here is that antialiasing be it temporal or spatial will usually require just as much oomph from your graphics card as increasing the resolution and framerate will require, which will create _real_ detail instead of faking it, using the limits of the human vision to create the motion blur instead, so bottom line is this: It is better to increase resolution than use antialiasing as long as the monitor can keep up, and even the fastest of graphics cards can't push anything remotely photorealistic at 1600x1200 at 85fps. If you've maxed out your display, and you've still got fillrate to burn, spatial and temporal antialiasing is more worthwhile.