People have been using octrees for ray tracing for decades. Sparse ones. For voxel geometry as well as other primitives. They compare poorly to more recent methods. Structures like this have already been implemented on GPU, with some success, but GPUs are going to need some more changes before it gets really good. Carmack is going to have to be more specific in order to not be on crack.
Currently at 8th place. I've seen this story all over and I don't know why it does not link to the actual standings:
http://www.netflixprize.com/leaderboard
Oh wait, I know why. Because it makes the story look dated!
A great irony (or an intentional tease?) of quantum entanglement is that the physics behind it works out exactly such that it cannot be used for faster-than-light communication.
Huh? You can play all you want without being online. Even if you've already run online updates on your game. I'm not sure if the steam you're not using is the same steam that I'm using...
I've spoken with people who worked on large-scale projects to create IT systems that would monitor patients, gather statistics for datamining (to improve the treatments and learn new correlations), and even automatically suggest diagnoses and drugs (with references and reasons for the suggestions) to help doctors save time. I know two separate groups who tried similar projects. Both projects failed for the same reason: Doctors did not want their actions to be logged. Doctors want the freedom to skirt protocols when they feel it is appropriate, without worry of repercussions from insurance and / or drug companies. If their actions are logged, then doctors would lose all their power, and that power would be completely transferred to whoever writes the protocols. There was also quite a lot of resistance from the doctors who didn't want to deal with a PDA or whatever while working, but that was a less significant problem.
... from who you are out of their field of view I think that's the problem right there. Most people have roughly a 180 degree field of view, especially when it comes to detecting eyes or faces looking in their direction. I can tell if someone is looking at me even if they are standing slightly behind me to one side, because I have a 210 degree peripheral field of view. That's with my eyes locked straight ahead. Also, people tend to scan their environment a little bit with their eyes, which increases that field of view without even turning their head. Furthermore, people tend to turn their heads a little once in a while, again increasing the field of view.
My hypothesis is that people look back at you nervously because they saw your staring at them, given that you were more than likely in their field of view, even when you thought you weren't. As a rule of thumb, if you aren't looking at the back of their head, then they may very well see you within a few seconds of ordinary milling around.
I'm not postulating anything, nor am I engaging in wishful thinking. I simply won't go so far as to assert that there is no "magic" me in my head that is not covered by current physics. Complexity theory may get involved in there somewhere, but I won't throw away the possibility that consciousness could be some kind of intrinsic property of the universe, as fundamental as anything else in our present theories. We simply have no idea. Until someone comes up with a more or less complete model of consciousness, I will not pass judgement. Maybe I've misunderstood, but I took your original post as an indication that you had already closed your mind to the possibility of new phyisics being discovered with relation to consciousness.
While it's hard to argue against you that we're all just meat popsicles, your description of consciousness does nothing to address the mystery of the "ghost in the machine". The existence of my consciousness seems hardly necessary if I am simply a stimulus/response machine. Why do I perceive colors? Couldn't my body just react to different wavelengths of light hitting my retina and sending signals to my brain without requiring the color "experience"? I would argue that we have absolutely no understanding of consciousness at this point in time, and coupled with an incomplete understanding of physics, I would not rule out the possibility that consciousness may involve new physics not covered in our standard model.
I've seen hundreds of bees crawling around, apparently unable to fly, on several occasions here in California. I see a lot of them on sidewalks, on tennis courts, on my terrace, on concrete steps. They crawl around, buzzing pitifully. I've tried to help them, but they can't fly, or at least can't fly for more than a couple seconds at a time. They all die within a matter of hours after that. I've swept dozens of dead bees of my terrace. It's kind of sad, and disturbing. When I heard about these "hive collapses", I thought of all those dead bees I saw. Could be related.
At first I thought this must be at least equivalent to infinitesimal analysis, or taylor series expansion and taking limits, but after reading TFA I realise it's not. The axioms he presents are simply the NaN axioms, and they still lose information just like the "standard" axioms do. Loss of information is the thing that limit analysis avoids in order to get sensible answers near zeros, but even that doesn't work for some non-smooth functions. Anyhow, he suggests implementing his axioms to replace the IEEE axioms, but if you did that you would still lose information when multiplying by zero and such, so that later divisions by zero still give you NaNs, and you haven't gained a thing. You get practically the same result by simply ignoring floating point exceptions on current hardware! His assertion that this would improve the floating point model by avoiding these problems is therefore false.
Web 2.0 is so 2005! I'm already developing applications for Web 3.0! My company has a roadmap for the application extending all the way into Web 6.0 in 2010! It's all done in PowerPoint and Visio!
Deja vu is just a hash collision in the brain. There is no mystery. There are no "glitches in the matrix" required. No magic needed. No unusual mental processes at work. It's just a hash collision in your recall system. Not that our brains use hashing per se, but just to put it into lay terms. Lay terms around here, at any rate.
Any database that allows people to determine the identities of all the people at any scene, whether it is a crime scene or otherwise, is a civil liberties issue. You were at WHAT social gathering? With WHOM? Now we're going to all have to start behaving like Ethan Hawke in GATTACA, scrubbing off all our dead skin cells before we go out.
It could be compression, or if they have any budget it's probably mastering.
Mastering is a very cool process where the audio engineer removes or reduces frequencies that the ear is less sensitive to, and also removes or reduces frequencies that are not important for the sounds being played, thereby leaving more dynamic range for the important frequencies. The important frequencies can then be cranked up a notch or two without clipping. In addition, as mentioned elsewhere, the signal can be compressed a little bit to get even more oomph, but this does affect the quality of the sound.
ProTools, which is used by many recording studios, has these features built in. It's easy to tell a mastered song from an unmastered song when playing in WinAmp, for example, because a mastered song will have higher, longer-lasting peaks in the little frequency visualizer, but will still sound crisp and clean.
Jeff Gilchrist's Archive Comparison Test has been around for years, and covers many more archivers and uses several different data sets, on several different platforms. It has even been cited in compression literature:
For raster images PNG is as good as it gets at the moment
BMF will save you a good 30% to 50% more than PNG.
MRP is a newer format that I just noticed today while searching for BMF websites. It's still in vaguely research stages, but it clearly beats even BMF. Note that this website doesn't even show comparisons to PNG, since PNG fell off the "top 10" many winters ago.
If you want to talk about market penetration, then PNG has severe advantages. But there's not much stopping someone from making a MRP plug-in for Firefox, is there? Well, okay, yes there is. Shut up.
Those orange tic-tac boxes were the best! Great for the outermost walls of my lego fortresses. And you could put lego men's heads inside the little square hole in the corner!
Hear hear! When Occam's Razor comes up in discussion like this, I usually point out that nearly every scientific discovery has been a counter-case for the notion that Occam's Razor actually indicates anything. As we gain more information from observations, theories often require revision, often becoming more complex. It is, as you say, a rule of thumb. A sensible work ethic.
"The only problem is Mills' theory is supposed to be impossible when using current rules of quantum mechanics."
Yes, and galaxies spin too fast using current rules of General Relativity. This has been verified by many top scientists. Oh, wait, no they don't!
First, 1% of $20.5 billion is $205 million. But this only supports your point:) Second, revenues of $100,000 would earn major profit for you, for sure, but innovation funds only begin to get interested when there is "hockey stick growth" into the 7-digit or 8-digits numbers at the end of your 3-year business plan. Your measly 6-digit profit figures won't even get a sideways glance.
Slashdot Mirror
People have been using octrees for ray tracing for decades. Sparse ones. For voxel geometry as well as other primitives. They compare poorly to more recent methods. Structures like this have already been implemented on GPU, with some success, but GPUs are going to need some more changes before it gets really good. Carmack is going to have to be more specific in order to not be on crack.
Currently at 8th place. I've seen this story all over and I don't know why it does not link to the actual standings: http://www.netflixprize.com/leaderboard Oh wait, I know why. Because it makes the story look dated!
No mention of lucid dreaming == fail.
A great irony (or an intentional tease?) of quantum entanglement is that the physics behind it works out exactly such that it cannot be used for faster-than-light communication.
It's so obvious, in fact, that Bryce Bayer mentioned similar designs in his original Bayer filter patent. Check it out here:
http://www.quadibloc.com/other/cfaint.htm
I think it's getting press now because Kodak is actually considering implementing the idea in consumer cameras.
Huh? You can play all you want without being online. Even if you've already run online updates on your game. I'm not sure if the steam you're not using is the same steam that I'm using...
I've spoken with people who worked on large-scale projects to create IT systems that would monitor patients, gather statistics for datamining (to improve the treatments and learn new correlations), and even automatically suggest diagnoses and drugs (with references and reasons for the suggestions) to help doctors save time. I know two separate groups who tried similar projects. Both projects failed for the same reason: Doctors did not want their actions to be logged. Doctors want the freedom to skirt protocols when they feel it is appropriate, without worry of repercussions from insurance and / or drug companies. If their actions are logged, then doctors would lose all their power, and that power would be completely transferred to whoever writes the protocols. There was also quite a lot of resistance from the doctors who didn't want to deal with a PDA or whatever while working, but that was a less significant problem.
... from who you are out of their field of view I think that's the problem right there. Most people have roughly a 180 degree field of view, especially when it comes to detecting eyes or faces looking in their direction. I can tell if someone is looking at me even if they are standing slightly behind me to one side, because I have a 210 degree peripheral field of view. That's with my eyes locked straight ahead. Also, people tend to scan their environment a little bit with their eyes, which increases that field of view without even turning their head. Furthermore, people tend to turn their heads a little once in a while, again increasing the field of view.My hypothesis is that people look back at you nervously because they saw your staring at them, given that you were more than likely in their field of view, even when you thought you weren't. As a rule of thumb, if you aren't looking at the back of their head, then they may very well see you within a few seconds of ordinary milling around.
I'm not postulating anything, nor am I engaging in wishful thinking. I simply won't go so far as to assert that there is no "magic" me in my head that is not covered by current physics. Complexity theory may get involved in there somewhere, but I won't throw away the possibility that consciousness could be some kind of intrinsic property of the universe, as fundamental as anything else in our present theories. We simply have no idea. Until someone comes up with a more or less complete model of consciousness, I will not pass judgement. Maybe I've misunderstood, but I took your original post as an indication that you had already closed your mind to the possibility of new phyisics being discovered with relation to consciousness.
While it's hard to argue against you that we're all just meat popsicles, your description of consciousness does nothing to address the mystery of the "ghost in the machine". The existence of my consciousness seems hardly necessary if I am simply a stimulus/response machine. Why do I perceive colors? Couldn't my body just react to different wavelengths of light hitting my retina and sending signals to my brain without requiring the color "experience"? I would argue that we have absolutely no understanding of consciousness at this point in time, and coupled with an incomplete understanding of physics, I would not rule out the possibility that consciousness may involve new physics not covered in our standard model.
Not a surprise. If you look at the history of the contest, you will find that the University of Waterloo has won twice:
http://en.wikipedia.org/wiki/User:ReiVaX/SWERC
It's pretty common for multiple Canadian teams to place in the top 10 or 20, year after year.
I've seen hundreds of bees crawling around, apparently unable to fly, on several occasions here in California. I see a lot of them on sidewalks, on tennis courts, on my terrace, on concrete steps. They crawl around, buzzing pitifully. I've tried to help them, but they can't fly, or at least can't fly for more than a couple seconds at a time. They all die within a matter of hours after that. I've swept dozens of dead bees of my terrace. It's kind of sad, and disturbing. When I heard about these "hive collapses", I thought of all those dead bees I saw. Could be related.
At first I thought this must be at least equivalent to infinitesimal analysis, or taylor series expansion and taking limits, but after reading TFA I realise it's not. The axioms he presents are simply the NaN axioms, and they still lose information just like the "standard" axioms do. Loss of information is the thing that limit analysis avoids in order to get sensible answers near zeros, but even that doesn't work for some non-smooth functions. Anyhow, he suggests implementing his axioms to replace the IEEE axioms, but if you did that you would still lose information when multiplying by zero and such, so that later divisions by zero still give you NaNs, and you haven't gained a thing. You get practically the same result by simply ignoring floating point exceptions on current hardware! His assertion that this would improve the floating point model by avoiding these problems is therefore false.
Web 2.0 is so 2005! I'm already developing applications for Web 3.0! My company has a roadmap for the application extending all the way into Web 6.0 in 2010! It's all done in PowerPoint and Visio!
Deja vu is just a hash collision in the brain. There is no mystery. There are no "glitches in the matrix" required. No magic needed. No unusual mental processes at work. It's just a hash collision in your recall system. Not that our brains use hashing per se, but just to put it into lay terms. Lay terms around here, at any rate.
Any database that allows people to determine the identities of all the people at any scene, whether it is a crime scene or otherwise, is a civil liberties issue. You were at WHAT social gathering? With WHOM? Now we're going to all have to start behaving like Ethan Hawke in GATTACA, scrubbing off all our dead skin cells before we go out.
It could be compression, or if they have any budget it's probably mastering.
Mastering is a very cool process where the audio engineer removes or reduces frequencies that the ear is less sensitive to, and also removes or reduces frequencies that are not important for the sounds being played, thereby leaving more dynamic range for the important frequencies. The important frequencies can then be cranked up a notch or two without clipping. In addition, as mentioned elsewhere, the signal can be compressed a little bit to get even more oomph, but this does affect the quality of the sound.
ProTools, which is used by many recording studios, has these features built in. It's easy to tell a mastered song from an unmastered song when playing in WinAmp, for example, because a mastered song will have higher, longer-lasting peaks in the little frequency visualizer, but will still sound crisp and clean.
Jeff Gilchrist's Archive Comparison Test has been around for years, and covers many more archivers and uses several different data sets, on several different platforms. It has even been cited in compression literature:
http://www.compression.ca/
Does this mean that another small town, on the other side of the mirror, will be cast in a giant shadow for part of the day?
For raster images PNG is as good as it gets at the moment
BMF will save you a good 30% to 50% more than PNG.
MRP is a newer format that I just noticed today while searching for BMF websites. It's still in vaguely research stages, but it clearly beats even BMF. Note that this website doesn't even show comparisons to PNG, since PNG fell off the "top 10" many winters ago.
If you want to talk about market penetration, then PNG has severe advantages. But there's not much stopping someone from making a MRP plug-in for Firefox, is there? Well, okay, yes there is. Shut up.
Those orange tic-tac boxes were the best! Great for the outermost walls of my lego fortresses. And you could put lego men's heads inside the little square hole in the corner!
See my journal entry on the subject: http://yro.slashdot.org/~glyph42/journal/65573
Hear hear! When Occam's Razor comes up in discussion like this, I usually point out that nearly every scientific discovery has been a counter-case for the notion that Occam's Razor actually indicates anything. As we gain more information from observations, theories often require revision, often becoming more complex. It is, as you say, a rule of thumb. A sensible work ethic.
"The only problem is Mills' theory is supposed to be impossible when using current rules of quantum mechanics."
Yes, and galaxies spin too fast using current rules of General Relativity. This has been verified by many top scientists. Oh, wait, no they don't!
First, 1% of $20.5 billion is $205 million. But this only supports your point :) Second, revenues of $100,000 would earn major profit for you, for sure, but innovation funds only begin to get interested when there is "hockey stick growth" into the 7-digit or 8-digits numbers at the end of your 3-year business plan. Your measly 6-digit profit figures won't even get a sideways glance.