I hope they'll do new classic Wii-controller games in full HD.
Judging from their demo videos, you have one screen-controller for the hardcore games and regular old Wiimotes for multiplayer and casual stuff. They even showed weird stuff like putting the screen-controller on the ground showing a golf-ball and then swinging with a Wiimote.
In some games, you actually want to keep things hidden from the other players.
Rumor right now is that the WiiU will only support a single screen controller and they didn't demo anything that used two screen controller. All multiplayer they showed used one screen controller and the rest of player would get regular Wiimotes. So it was essentially asymmetric multiplayer, as different player would play with different controls.
Yes, because the iPad and iPad don't seem to be doing well at all.
. WiiU doesn't have a iPad level touch screen, it has a very basic single-touch screen, likely like the DS and 3DS and while those are functional, they aren't really all that great.
It isn't really that practical to render everything on the console and stream it uncompressed to four separate tablet remotes
Rumor is that the WiiU only supports a single screen-controller, the rest will be regular Wiimotes. Also the whole compression/uncompression will likely be done in hardware, not software, so I wouldn't expect the thing to be of much use as standalone. Also the whole WiiU would be prohibitively expensive if the thing would be any good as a standalone and able to compete against Vita.
But when will people stop trusting the Intertubes security implicitly and just blindly dumping all their personal info into various "secure" web sites and Internet connected systems?
When companies will stop requiring the data to gain access. In the PSN case for example you have to give name and address, even so that is completly unneeded for operating the free part of the service. It will even go so far as to do a bit of error checking on the data, so you can't just enter random stuff as address, it has to be a valid one. And once there faking the information actually becomes work, it is no longer a case of just not entering it and thus most people will provide real data.
The way to get companies to limit data collection of course requires some new laws, i.e. only allow companies to collect data necessary for the given transaction and not more then that.
User education of course is needed as well, but that only goes so far and won't really fix the underlying problem.
That is only true in a very theoretical sense and completely wrong in practice. Almost everything that burns CPU on your computer today is easily parallelizable, your video encoder doesn't really care if the other CPU is crunching away on the next scene in the movie, most image filters work just fine when applied to only a section of an image, a game could easily do AI for each unit in parallel and your webbrowser shouldn't really care if each webpage is rendered on a separate thread either.
The part of the code that isn't parallelizable certainly exist, but it is almost never that part of the code that is actually CPU intensive, especially not on your Average Joe's computer. The only problem with parallelization is that neither programs nor programming languages are prepared for it, so it is a lot more troublesome to add then it should be.
That doesn't really address the problem I mentioned, all it does is hide your IP and I am not even sure if that matters in BitCoin as as far as I understand you don't really transfer BitCoins by sending them over directly, but by signing them and putting them in the public database. Thus you wouldn't even need a direct IP connection.
Third, the Creative Commons have gone through 5 versions of licenses (1.0, 2.0, 2.1, 2.5 and 3.0) in just over 4 years and (as far as I know) there is no forwards or backwards compatibility between the different versions.
There is forward compatibility, since I think version 2.0, 1.0 lacked it:
You may Distribute or Publicly Perform an Adaptation only under the terms of: (i) this License; (ii) a later version of this License with the same License Elements as this License; (iii) a Creative Commons jurisdiction license (either this or a later license version) that contains the same License Elements as this License (e.g., Attribution-ShareAlike 3.0 US)); (iv) a Creative Commons Compatible License.
I perceive things. That's not motion of atoms, or if it is
Why is that not motion of atoms? My digicam can see things too, all without no magic just a few electrons moving around. What exactly makes consciousness different, other then that the brain is much better at the task? How can I test if an object has consciousness?
Is it true that this interpretation (which I am told is the most accepted), requires the presence of consciousness?
No, its just a very common misinterpretation. An observer in quantum mechanics is essentially everything that interacts with a particle. So if two particles collide, one of them is an observer. It has nothing to do with consciousness.
if you don't know something, we name it "quantum physics"
We know quantum mechanics quite well and can make extreme precise predictions with it. The only problem with QM is that it doesn't follow our intuition, but that really shouldn't be much of a surprise given that our intuition has evolved so we can make decisions in a macroscopic world, not on an subatomic level.
what would be the point of evolving any kind of conscious awareness at all if consciousness is simply a detached observer of events in the brain?
The way I see it, consciousness is mainly about building a model of the world so that your brain can make high level decisions instead of just having reflexes directly reacting to sense input. Naturally decisions your brain makes need to become part of the model, so the brain feeds them back into the model after it has already made them. And the whole feeling of self is just a part of that model so that you don't end up eating your arm when you are hungry.
As for QM playing a role in consciousness, yeah, of course it plays a role, but that role will be no larger then it plays in any other system. It is an unimportant low level implementation detail, not some fundamental magical requirement for consciousness.
"Observer" has nothing to do with a conscious human observer, but just refers to whatever tool or technique you use to measure a physical effect. At large scale you can shine a light on an object and observe it by the photons are reflected without problem, if you want to observe a single atom on the other side by shooting photons at it, the photons will have an effect on the atom.
While BitCoin is anonymous, from what I understand all transactions in BitCoin are public. Wouldn't that make it rather easy to trace it back and de'anomynize the transactions as soon as I make a transaction with the real world? Say I go shopping at Amazon, I naturally have to tell Amazon, my name, where I live, etc. as I wanna get a package from them, I also have to tell them my BitCoin address as thats how I would pay. My address now however is no longer anonymous, it is attached to a name that Amazon knows. BitCoin of course allows the creation of an arbitrary number of addresses, but I don't see them helping much, as you wouldn't have any money attached to them, if you transfer money to them, you again can be trivially traced, as one would just need to look who was the previous owner of the money. Do that with a few transactions and you might soon find the real address.
Don't know all to much about BitCoin, so is that really the case, I did I miss something fundamental?
Not every piece of human creation needs to be updated, upscaled, "improved" and redone.
Things change and what once looked good on your 14" CRT TV in your bedroom doesn't look so good when scaled up to a 44" LCD TV in your living room. Results of course vary from game to game and this algorithm in particular seems to smooth out some things more then needed, but hqx and the other scalers available in emulators often give quite solid results, that smooth out the pixels, without destroying the original look behind them. Especially fonts often get much more readable when smoothed by a proper scaler instead of just blown up without any kind of interpolation. And of course when you don't like it you can still just switch them off. Its good to have options.
How so? A video is just a series of fixed pictures, one after the other.
Yes, but it are pictures of the same object in motion. Super simplified example: Imagine you have a one-pixel camera. A single frame of video of that will tell you nothing by itself, as its just a pixel, but if you take the camera, move it around in the right pattern and add the images together with the correct offset you can get a full 2D picture of whatever you are photographing. Same can be done with a 2D camera, you have to reconstruct how the images are shifted relative to each other and then add them together, that way you get a higher resolution then you have in a single frame, its called Super resolution. That is of course more useful for real photos then it is for video games.
But for video games you would get plenty of benefit as well, as multiple frames tell you much more about the structure of the object, in a simple Space Invaders you might not be able to tell which pixel is part of the body and which is part of an arm or leg, in motion that however becomes pretty clear and thus you can avoid blurring pixels together that really represents different parts of an objects.
Yes, but we are not talking about CSI style super zoom that is used to read number plates, but about making pixel art look pretty at hi-res. And as far as pixel art goes, there often is no original information, because the pixel aren't the result of scaling down a larger image, but they themselves are the original. Now of course one can't get the one true real hi-res representation of a low-res image, but on the other side current algorithms do make a ton of very obvious mistakes. With the space invader you might not be able to tell how exactly its antennae where supposed to look like in the artists mind, but you can easily infer that they should be in the same style in both frames of animation. Yet with this algorithm and with most others, they aren't, as the interpolation happens frame by frame, without taking the animation as a whole into account and thus the results you can are even worse then the original pixel art. You might also be able to tell that it actually are antennae in the first place, not just things that stick out of its body. Same with things like eyes, its easy to tell for a human that those pixels are supposed to be eyes and should be handled special, yet current scaling algorithm handle them just like any other pixel and blur them into pixels that serve a completely different purpose.
There is nothing that prevents a scaled up result from being much better then a blurry mess. Its not easy to get that, but all the information is there in the original pixel art, one just needs to identify it properly.
Wait, your example of "infinite super zoom" is CSI, not Blade Runner?
The Blade Runner infinite zoom is actually a lot more realistic then the CSI one, as the CSI one works with current day cameras and essentially creates information out of nothing, with the BladeRunner zoom you don't really know what kind of source data is has access to. It is not unthinkable for example that in the future a paper photograph will come with an embedded SD card or RFID that contains either a much higher resolution version of the photo or an access key to the photo stored in the cloud. Furthermore some of the 3D panning seen in BladeRunner, which go beyond what a 2D photo would be able to do, can be accomplished via modern microlense array cameras and lightfield capture, not unthinkable that they will be common place one day. And as far as the insane resolution goes: That's essentially a software problem, human eye is something equivalent to a 1 megapixel camera if I remember correctly and yet it gives your brain a 600 megapixel field of view, all done by moving across the screen and stitching the results together, modern cameras should be able to replicate that sooner or later on the fly, they after all already can do that, it just requires a bit of setup.
Reconstructing image details accurately is still impossible and always will be.
Why should that be always impossible? There is nothing all that hard in it, humans can do it pretty easily. The problem with these algorithms is that they are all pretty primitive and only really take local pixels into account, not the whole image, thus in the example in the article you have edges in the back and belly of dolphin, which are trivial to tell that they should be smooth, for a human at least. The highlights on th gradients are also wrong, as the algorithm apparently doesn't take them into proper account. While that might not be easy to fix, it isn't exactly impossible.
Those were point&click adventures not RPG. And while LucasArts is kind of dead in that area, the people working on those titles in the first place are still alive and doing games, namely TellTale Games and DoubleFIne, and they have put out quite a few adventure games in the last years.
It is not a myth, it was very damn real for many years. It is just that in the last five years most games have moved to consoles as their primary platform and thus you can run those games on an older PC as well, as they are just not that demanding to begin with. Real PC games with PC as primary platform are still quite a bit more demanding and you won't run something like ArmAII on a five year old computer in maximum details at any tolerable speed. With Crysis1 you couldn't even get a PC when it was released that could run the game on maximum details. Also the Crysis2 demo I tried ran rather crappy on my four year old PC (not a hardcore gamer PC, just midrange with a new HD5670).
Ray tracing is much faster for very detailed geometry (very high triangle count) compared to scanline rendering.
Yes, but the issue is that modern hardware can already render as much polygons as there are pixels on the screen (i.e. Uncharted 2 has 1.2 million triangles per frame and only 1280*720 = 921'600 pixel). So that is really more of a theoretical benefit of ray tracing then one that matters in practice and geometry shader will make that even less of an issue in the future, as they can insert triangles on the fly to smooth out round objects.
Raytracing is based on a particles of light, while in real world you have the particle/wave duality. Anyway, you are not going to set up double slit experiments in your games anytime soon, so that is a non-issue in practical terms.
The only difference is that computers are working from your eye going forward,
Yeah, but that is exactly what makes computer graphics look so computer generated. You say it like its a small thing, but by shooting rays from the eyes you ignore all indirect lighting and indirect lighting is the very thing that makes things look realistic, without it everything looks like a really shitty computer generated picture. That you get good refraction and reflection doesn't really help you a bit, as neither of these effects really matters much, real world simply isn't build out of shiny glass spheres. Those bits of reflection and refraction that you can get in reallife can easily be faked with environment maps and a bit of displacement mapping, the errors that you get that way are completely negligible.
And anyway, essentially computer game graphics have already long moved past past direct lighting and all the hip stuff these days are things like subsurface scattering, photon mapping, ambient occlusion, god rays, depths of field and a bunch of other stuff that mostly happens as post-processing step. So if you do the rendering via raytracing or by painting triangles doesn't really matter a lot, as what makes the graphics look the way they do happens afterwards.
Thats not to say that it wouldn't be nice to have fast hardware raytracing, but we don't, rendering triangles still gives you much better results as it is simply much faster.
Gameplay/playability is always more important than graphics. It has ALWAYS been the playability of a game that's made the big hits.
Its not an either or, both graphics and gameplay go hand in hand. Almost all big successes in gaming history have been enabled due to advances in graphics that could then be utilized for new gameplay. You couldn't have done Mario64 on a SNES or Zelda:OOT, Doom on 286 wouldn't have been much fun either, FinalFantasyVII without CD, that wouldn't have worked, and games like Monkey Island or Maniac Mansion where one of the graphically most impressive titles of their times (full bitmap background instead of tilemaps, large sprites, sprite scaling, etc.).
And of course the fun part: CoD isn't graphical powerhouse, in fact most people by now call that engine kind of outdated, instead the engine is optimized for playability, instead of shiny graphics effects you get 60fps and super smooth controls. I still consider it a shitty game, but certainly not because they didn't spend a lot of time fine tuning the core gameplay.
Slashdot Mirror
I hope they'll do new classic Wii-controller games in full HD.
Judging from their demo videos, you have one screen-controller for the hardcore games and regular old Wiimotes for multiplayer and casual stuff. They even showed weird stuff like putting the screen-controller on the ground showing a golf-ball and then swinging with a Wiimote.
In some games, you actually want to keep things hidden from the other players.
Rumor right now is that the WiiU will only support a single screen controller and they didn't demo anything that used two screen controller. All multiplayer they showed used one screen controller and the rest of player would get regular Wiimotes. So it was essentially asymmetric multiplayer, as different player would play with different controls.
Yes, because the iPad and iPad don't seem to be doing well at all.
.
WiiU doesn't have a iPad level touch screen, it has a very basic single-touch screen, likely like the DS and 3DS and while those are functional, they aren't really all that great.
It isn't really that practical to render everything on the console and stream it uncompressed to four separate tablet remotes
Rumor is that the WiiU only supports a single screen-controller, the rest will be regular Wiimotes. Also the whole compression/uncompression will likely be done in hardware, not software, so I wouldn't expect the thing to be of much use as standalone. Also the whole WiiU would be prohibitively expensive if the thing would be any good as a standalone and able to compete against Vita.
But when will people stop trusting the Intertubes security implicitly and just blindly dumping all their personal info into various "secure" web sites and Internet connected systems?
When companies will stop requiring the data to gain access. In the PSN case for example you have to give name and address, even so that is completly unneeded for operating the free part of the service. It will even go so far as to do a bit of error checking on the data, so you can't just enter random stuff as address, it has to be a valid one. And once there faking the information actually becomes work, it is no longer a case of just not entering it and thus most people will provide real data.
The way to get companies to limit data collection of course requires some new laws, i.e. only allow companies to collect data necessary for the given transaction and not more then that.
User education of course is needed as well, but that only goes so far and won't really fix the underlying problem.
Most problems are not parrallizeable
That is only true in a very theoretical sense and completely wrong in practice. Almost everything that burns CPU on your computer today is easily parallelizable, your video encoder doesn't really care if the other CPU is crunching away on the next scene in the movie, most image filters work just fine when applied to only a section of an image, a game could easily do AI for each unit in parallel and your webbrowser shouldn't really care if each webpage is rendered on a separate thread either.
The part of the code that isn't parallelizable certainly exist, but it is almost never that part of the code that is actually CPU intensive, especially not on your Average Joe's computer. The only problem with parallelization is that neither programs nor programming languages are prepared for it, so it is a lot more troublesome to add then it should be.
That doesn't really address the problem I mentioned, all it does is hide your IP and I am not even sure if that matters in BitCoin as as far as I understand you don't really transfer BitCoins by sending them over directly, but by signing them and putting them in the public database. Thus you wouldn't even need a direct IP connection.
Third, the Creative Commons have gone through 5 versions of licenses (1.0, 2.0, 2.1, 2.5 and 3.0) in just over 4 years and (as far as I know) there is no forwards or backwards compatibility between the different versions.
There is forward compatibility, since I think version 2.0, 1.0 lacked it:
I perceive things. That's not motion of atoms, or if it is
Why is that not motion of atoms? My digicam can see things too, all without no magic just a few electrons moving around. What exactly makes consciousness different, other then that the brain is much better at the task? How can I test if an object has consciousness?
Is it true that this interpretation (which I am told is the most accepted), requires the presence of consciousness?
No, its just a very common misinterpretation. An observer in quantum mechanics is essentially everything that interacts with a particle. So if two particles collide, one of them is an observer. It has nothing to do with consciousness.
if you don't know something, we name it "quantum physics"
We know quantum mechanics quite well and can make extreme precise predictions with it. The only problem with QM is that it doesn't follow our intuition, but that really shouldn't be much of a surprise given that our intuition has evolved so we can make decisions in a macroscopic world, not on an subatomic level.
what would be the point of evolving any kind of conscious awareness at all if consciousness is simply a detached observer of events in the brain?
The way I see it, consciousness is mainly about building a model of the world so that your brain can make high level decisions instead of just having reflexes directly reacting to sense input. Naturally decisions your brain makes need to become part of the model, so the brain feeds them back into the model after it has already made them. And the whole feeling of self is just a part of that model so that you don't end up eating your arm when you are hungry.
As for QM playing a role in consciousness, yeah, of course it plays a role, but that role will be no larger then it plays in any other system. It is an unimportant low level implementation detail, not some fundamental magical requirement for consciousness.
"Observer" has nothing to do with a conscious human observer, but just refers to whatever tool or technique you use to measure a physical effect. At large scale you can shine a light on an object and observe it by the photons are reflected without problem, if you want to observe a single atom on the other side by shooting photons at it, the photons will have an effect on the atom.
Its a computational or logic processing device.
Something that is easily confirmed by observation (for reasonably wide definition of computational device of course).
Maybe its more of a network card, linking our actual conscientiousness to our bodies.
Something that somebody made up on the spot with no evidence what so ever.
Its a fundamental difference but it would explain telepathy, ...
You are doing it wrong. You are trying to explain things that don't exist with things you don't understand.
While BitCoin is anonymous, from what I understand all transactions in BitCoin are public. Wouldn't that make it rather easy to trace it back and de'anomynize the transactions as soon as I make a transaction with the real world? Say I go shopping at Amazon, I naturally have to tell Amazon, my name, where I live, etc. as I wanna get a package from them, I also have to tell them my BitCoin address as thats how I would pay. My address now however is no longer anonymous, it is attached to a name that Amazon knows. BitCoin of course allows the creation of an arbitrary number of addresses, but I don't see them helping much, as you wouldn't have any money attached to them, if you transfer money to them, you again can be trivially traced, as one would just need to look who was the previous owner of the money. Do that with a few transactions and you might soon find the real address.
Don't know all to much about BitCoin, so is that really the case, I did I miss something fundamental?
Not every piece of human creation needs to be updated, upscaled, "improved" and redone.
Things change and what once looked good on your 14" CRT TV in your bedroom doesn't look so good when scaled up to a 44" LCD TV in your living room. Results of course vary from game to game and this algorithm in particular seems to smooth out some things more then needed, but hqx and the other scalers available in emulators often give quite solid results, that smooth out the pixels, without destroying the original look behind them. Especially fonts often get much more readable when smoothed by a proper scaler instead of just blown up without any kind of interpolation. And of course when you don't like it you can still just switch them off. Its good to have options.
How so? A video is just a series of fixed pictures, one after the other.
Yes, but it are pictures of the same object in motion. Super simplified example: Imagine you have a one-pixel camera. A single frame of video of that will tell you nothing by itself, as its just a pixel, but if you take the camera, move it around in the right pattern and add the images together with the correct offset you can get a full 2D picture of whatever you are photographing. Same can be done with a 2D camera, you have to reconstruct how the images are shifted relative to each other and then add them together, that way you get a higher resolution then you have in a single frame, its called Super resolution. That is of course more useful for real photos then it is for video games.
But for video games you would get plenty of benefit as well, as multiple frames tell you much more about the structure of the object, in a simple Space Invaders you might not be able to tell which pixel is part of the body and which is part of an arm or leg, in motion that however becomes pretty clear and thus you can avoid blurring pixels together that really represents different parts of an objects.
Because the information is simply not there?
Yes, but we are not talking about CSI style super zoom that is used to read number plates, but about making pixel art look pretty at hi-res. And as far as pixel art goes, there often is no original information, because the pixel aren't the result of scaling down a larger image, but they themselves are the original. Now of course one can't get the one true real hi-res representation of a low-res image, but on the other side current algorithms do make a ton of very obvious mistakes. With the space invader you might not be able to tell how exactly its antennae where supposed to look like in the artists mind, but you can easily infer that they should be in the same style in both frames of animation. Yet with this algorithm and with most others, they aren't, as the interpolation happens frame by frame, without taking the animation as a whole into account and thus the results you can are even worse then the original pixel art. You might also be able to tell that it actually are antennae in the first place, not just things that stick out of its body. Same with things like eyes, its easy to tell for a human that those pixels are supposed to be eyes and should be handled special, yet current scaling algorithm handle them just like any other pixel and blur them into pixels that serve a completely different purpose.
There is nothing that prevents a scaled up result from being much better then a blurry mess. Its not easy to get that, but all the information is there in the original pixel art, one just needs to identify it properly.
Wait, your example of "infinite super zoom" is CSI, not Blade Runner?
The Blade Runner infinite zoom is actually a lot more realistic then the CSI one, as the CSI one works with current day cameras and essentially creates information out of nothing, with the BladeRunner zoom you don't really know what kind of source data is has access to. It is not unthinkable for example that in the future a paper photograph will come with an embedded SD card or RFID that contains either a much higher resolution version of the photo or an access key to the photo stored in the cloud. Furthermore some of the 3D panning seen in BladeRunner, which go beyond what a 2D photo would be able to do, can be accomplished via modern microlense array cameras and lightfield capture, not unthinkable that they will be common place one day. And as far as the insane resolution goes: That's essentially a software problem, human eye is something equivalent to a 1 megapixel camera if I remember correctly and yet it gives your brain a 600 megapixel field of view, all done by moving across the screen and stitching the results together, modern cameras should be able to replicate that sooner or later on the fly, they after all already can do that, it just requires a bit of setup.
Reconstructing image details accurately is still impossible and always will be.
Why should that be always impossible? There is nothing all that hard in it, humans can do it pretty easily. The problem with these algorithms is that they are all pretty primitive and only really take local pixels into account, not the whole image, thus in the example in the article you have edges in the back and belly of dolphin, which are trivial to tell that they should be smooth, for a human at least. The highlights on th gradients are also wrong, as the algorithm apparently doesn't take them into proper account. While that might not be easy to fix, it isn't exactly impossible.
Can we get some new RPG from LucasArt now?
Those were point&click adventures not RPG. And while LucasArts is kind of dead in that area, the people working on those titles in the first place are still alive and doing games, namely TellTale Games and DoubleFIne, and they have put out quite a few adventure games in the last years.
The PC upgrade cycle is a myth.
It is not a myth, it was very damn real for many years. It is just that in the last five years most games have moved to consoles as their primary platform and thus you can run those games on an older PC as well, as they are just not that demanding to begin with. Real PC games with PC as primary platform are still quite a bit more demanding and you won't run something like ArmAII on a five year old computer in maximum details at any tolerable speed. With Crysis1 you couldn't even get a PC when it was released that could run the game on maximum details. Also the Crysis2 demo I tried ran rather crappy on my four year old PC (not a hardcore gamer PC, just midrange with a new HD5670).
Ray tracing is much faster for very detailed geometry (very high triangle count) compared to scanline rendering.
Yes, but the issue is that modern hardware can already render as much polygons as there are pixels on the screen (i.e. Uncharted 2 has 1.2 million triangles per frame and only 1280*720 = 921'600 pixel). So that is really more of a theoretical benefit of ray tracing then one that matters in practice and geometry shader will make that even less of an issue in the future, as they can insert triangles on the fly to smooth out round objects.
It's an exact reproduction of the real world.
Raytracing is based on a particles of light, while in real world you have the particle/wave duality. Anyway, you are not going to set up double slit experiments in your games anytime soon, so that is a non-issue in practical terms.
The only difference is that computers are working from your eye going forward,
Yeah, but that is exactly what makes computer graphics look so computer generated. You say it like its a small thing, but by shooting rays from the eyes you ignore all indirect lighting and indirect lighting is the very thing that makes things look realistic, without it everything looks like a really shitty computer generated picture. That you get good refraction and reflection doesn't really help you a bit, as neither of these effects really matters much, real world simply isn't build out of shiny glass spheres. Those bits of reflection and refraction that you can get in reallife can easily be faked with environment maps and a bit of displacement mapping, the errors that you get that way are completely negligible.
And anyway, essentially computer game graphics have already long moved past past direct lighting and all the hip stuff these days are things like subsurface scattering, photon mapping, ambient occlusion, god rays, depths of field and a bunch of other stuff that mostly happens as post-processing step. So if you do the rendering via raytracing or by painting triangles doesn't really matter a lot, as what makes the graphics look the way they do happens afterwards.
Thats not to say that it wouldn't be nice to have fast hardware raytracing, but we don't, rendering triangles still gives you much better results as it is simply much faster.
Gameplay/playability is always more important than graphics. It has ALWAYS been the playability of a game that's made the big hits.
Its not an either or, both graphics and gameplay go hand in hand. Almost all big successes in gaming history have been enabled due to advances in graphics that could then be utilized for new gameplay. You couldn't have done Mario64 on a SNES or Zelda:OOT, Doom on 286 wouldn't have been much fun either, FinalFantasyVII without CD, that wouldn't have worked, and games like Monkey Island or Maniac Mansion where one of the graphically most impressive titles of their times (full bitmap background instead of tilemaps, large sprites, sprite scaling, etc.).
And of course the fun part: CoD isn't graphical powerhouse, in fact most people by now call that engine kind of outdated, instead the engine is optimized for playability, instead of shiny graphics effects you get 60fps and super smooth controls. I still consider it a shitty game, but certainly not because they didn't spend a lot of time fine tuning the core gameplay.