You are of course assuming that exactly 65% of the geeky crowd are lazy, which doesn't necessarily follow even if 65% of the whole of/. are lazy;)
I just thought "which font size?", decided on showing the kids a book with a certain font size, and didn't bother to work out the rest because I didn't consider it useful. I don't often obsess over primes and other mathematical oddities. I guess I'm not really a hardcore geek. 55-60 times is pretty impressive though!
Google's built in calculator and unit converter is indeed an excellent resource:) It even converts currency at current values (don't know how often it updates the conversion rates though).
Yeah, I don't know how many petabytes (or whatever metric would be suitable) would be required, but it probably will still happen someday. I'm sure people 200 years ago would have considered it impossible to carry around video recordings of hundreds of theatre productions or concerts in their pocket, but these days a large percentage of people in the world can do that easily on their media player or phone. These same pocket devices that can do much more than computers that used to take up whole buildings 50 years ago.
Holographic storage for example could be a good medium for storing data far more densely than we currently can (though again I don't know much theory in this area). But just as 200 years ago people didn't have any clue about transistors, perhaps we will discover some completely new concept that will enable us to conveniently and cheaply store more data than even holographic storage can. Don't want to sound like a complete pie in the sky nutjob, but it's true that technology is currently advancing at a phenomenal rate, and we have no way of knowing the state of computing in 20 years, let alone 200.
Sorry, I forgot the part where the guy who patented the product tried to get money from the actual author. He probably would have gotten away with it otherwise. He's not just a criminal - he's a complete moron as well!
Perhaps the jerkoff who stole the source either didn't expect the guy to be able to prove anything, or just didn't think he'd put any money towards a court case since technically he hasn't lost any money in the first place.
I'm glad the author is standing up for what is right anyway, and hope the offeder's reputation is ruined enough that nobody will ever pay for his software again..
Indeed, it just needs a kickstart to break out of routine. I naturally go through phases of being interested in different stuff anyway, I just want to try and make my next several phases something that will get me coding some games again:)
Unfortunately my attempt to get myself back into photography again just involved me buying an expensive camera and it sitting on a shelf in my cupboard. Despite the fact that I've got into a habit of going for walks on weekends rather than driving everywhere, I never think to bring my camera with me. As a student I used to wander around town just taking pictures of random stuff and uploading it to deviantArt, but I stopped doing that when I got a job.
I need to think of some solid project to work on for the Pandora before I get it otherwise it may suffer the same fate:( A port of Racer could perhaps be a good way to do that. I had wanted to write a car simulator around the time that I started university - I loved Gran Turismo on the PSX but there weren't really any good alternatives for the PC at that time, and since the PlayStation was at home I had nothing to play! Unfortunately I didn't have much knowledge about how tyre grip, suspension etc worked back then (though I think I could have a good go at it these days). A year or 2 later I found Racer - seems that some other guy had the same idea around the same time as me, but actually carried it out. Perhaps I should just try writing my own from scratch - I'd learn a lot more that way really. I used to dream of working for a games development house until I found out that most people at these places get paid sod all;) So now I'm kind of like you, I've been doing a bit of web/database development, and some other engineering simulation apps for simulating performance of the tools our company designs and builds (I didn't design the equations for them, but at least I understand what is going on most of the time!), the rest of the time I'm IT support, but at least I get good pay and benefits like a company car, phone, etc. I do enjoy it, it's a great place to work, but it's not as intellectually challenging as working as a games developer would be..
I was going to do my honours in AI (and did all the course work including the AI classes, but just never actually finished my final project due to some stupid personal reasons), so yeah I do have a bit of an idea about this stuff;)
I was interested in neural networks towards the end of high school and found NNbot for counter-strike, which resulted in me doing my own bot. I actually read about neurons in the Psychology section of encyclopaedia and was thinking about how I'd simulate them, how you'd sort out which neuron to simulate first and so on considering they all run in parallel - I'm pretty sure that was before I had heardÂthe term neural network as well, but I'd never actually tried implementing one myself.
Then I found a counter-strike bot called NeuralNetwork Bot (NNBox), I was extremely impressed:) My dad saw how enthusiastic I was about it and helped me get started writing my own CS bot and I spent a year or more on that, it was pretty good in the end (the website is still up at planethalflife.com/teambot , but the last version was only compatible up to around Beta 1.4 and 1.5 because they released official bots for CS after that, and I was at University by then anyway).
Anyway, compared to writing that sort of AI (pathfinding to a goal, and my bots also laid information points along the route say if they died somewhere, they'd be more cautious the next time around, perhaps throw in a flashbang first, etc) I think I'd find Chess AI very dry. The final year project I was assigned was going to involve the board game diplomacy, and I just wasn't really motivated for it. The project I actually wanted to do involved a robot, can't remember exactly what it was even designed to do. The guy who was assigned to that project actually swapped with someone else! I couldn't believe it, I didn't even try to swap with him because I didn't see how anyone couldn't be interested in that project:p
Anyway, the whole reason I came to write this post was your paragraph
"I need to figure out a way to open this bottle. Out of all the knowledge I have about the world, what should I consider? Should I consider things I know about flowers? Well no. Duh. Should I consider things I know about cats? Well no. Duh. Maybe, I should consider things about *bottles* and gee, I don't know, what else.. hmm.. how about the parts of the bottle, maybe a common part of a bottle that most bottles have: lids. Now, do I know anything about *opening*? Why yes, I do.. I should probably look at all the ways I know to open something and see if any of the things I've opened in the past are like bottles. Maybe I should try to determine what *type* of lid this bottle has on it."
This got me thinking, if anyone ever designs an AI that learns its predicates by parsing information from the internet, we're in trouble. Something like Wikipedia would be mostly okay, but can you imagine if it got onto Myspace or Slashdot comments? Then created its own Myspace page or started commenting itself? Would be pretty funny to see some of the stuff it would come out with, its political and religious alignment etc:)
I didn't say that's not the way it should be, as not every job is about hard labour, running a company is high stress, technical knowledge takes years of training etc - I was just saying that pay doesn't really relate to intelligence.
I'm sure there are a lot of people like me who could have studied business management and get paid more if they actually found it interesting. Actually I wouldn't even need to do that - I could just go work offshore and almost make as much every day as I currently make in a week. I could easily do the whole living isolated for a couple of months at a time thing.
But anyway, I find it more interesting working with technology than working with stuff like human resources or even working offshore (though I would consider it just for a bit of a change maybe for a couple of months if we're struggling to assemble teams). In a similar vein, I vastly prefer playing FPS games to RTS games.
From there, it's just a matter of designing a computer that can learn, and giving it the online records of every regulation chess match ever played, and letting it figure out how to be the Grandest Master. Once its learning is in place, it's trivial to copy and redistribute that knowledge.
I think you're romanticising it somewhat. You're mostly right about the way humans play, though grand masters don't just play on instinct, they will also perform small mental simulations for several moves ahead.
The day will come where computers can just have a database of every single possble chess move in any game. Then it will no longer be a 'game' where computers are concerned, it will just be a matter of who plays first (unless of course the winning play comes down to starting second - I seem to do better when I play black myself, as I prefer to defend than to attack). Computers currently still trim down seemingly fruitless path choices sure, but if it they had as good algorithms as you think, they wouldn't need to be run on a supercomputer. Computer hardware is are far better than our own brains for games like chess. Any game with 'complete information' will eventually be able to be bruteforced by any computer, no smart algorithms required.
What is interesting for proper AI is games which involve 'incomplete information' ie an element of chance or some other unknown, like certain variations of poker or any game involving dice. That is a completely different category to games like chess and checkers where you know exactly what options your opponent has. Then the computer really has to be either taught good styles of play, or learn for itself as you suggest. Chess computers that can 'learn' will ultimately be completely unecessary when the game can be easily bruteforced on a home computer, or there is an online database somewhere that can tell you the exact move to play to maximise your chances of winning. When it comes to certain variations of poker against good human players, always playing the mathematical odds is not going to win the game.
So a chess board is nothing like an abstraction of the same problem as a computer that can learn what 'love' is. Consider that most humans (or perhaps no human) cannot define accurately, that makes it basically impossible for a computer to learn what it is. It is a function of inbuilt genetics, emotions, hormones, whatever. You could even say it has spiritual elements. A computer has none of those things, they would all have to be simulated to some extent, rather than learned.
You are right that the brain is an amazing computer, but it is good at different things from a Von Neumann modelled computer. We can't perform billions of mathematical operations a second like a supercomputer can. Then again, for the things that we do, we don't need to. We can often just 'tell' what is going to happen in certain situations because we have learned as you point out - we can see that something is likely to topple over and reach out to stop it. For a computer to do all these things would require a lot of work, especially if you want it to 'learn' to do them rather than just hardcoding in ways to interpret visual stimuli and build a 3D physics model of what is being seen, then simulate what is going to happen in the next few seconds, and then let it know how to move the hand, then how to move it with accurate 'hand to eye' co-ordination and so on. Anyway, sorry I'll stop rambling, but I think people greatly underestimate the difficulties involved with AI, and romanticize the idea of a computer learning to be 'human'. Often we don't want a computer to be human - we want it to be better than human, because humans make mistakes. A chess computer that makes mistakes is no good to anyone (unless they are just looking for an ego boost).
It wasn't flamebait. Possibly passive aggressive, that sounds a lot like me, but I don't think anyone is going to get a kick out of this card unless they are at least a serious amateur coder. At least physics cards have a few games that work with them - this thing is not meant for anything as common as gaming.
I was thinking about it a bit more and realised it would be a good way to get to know the cell architecture if you are planning on doing something with it in future, like getting into PS3 games development or any other areas where the Cell architecture has been designed to excel - but iamwhoiamtoday doesn't sound like he fits the bill for that. Nobody should buy one of these cards just because they have a free PCIe slot.
No, I've never written drivers so I'm quite ignorant in that area. I guess I was just thinking more of a linux library which takes advantage of the abilities of the card, but you would indeed need a driver for that to be useful.
Definitely. I used to do a lot of hobbyist coding at home (been 7 years since I left home for Uni), but after I started university I spent most of my free time socialising, and since then I've not done any coding outside of work because I've just been wanting to relax.
I miss doing my own projects at home though - the coding I do at work is pretty mundane. It can still be rewarding sometimes, but compared to creating a game that you can play, and creating AI opponents etc, writing up an engineering or productivity application is pretty dull! I just need to better organise my time - I'm in the habit of frittering it away watching movies and playing games these days:/
Yeah. I enjoy playing chess, but Chess AI is only useful in a very limited domain, and eventually it won't really be necessary once computers can just brute force every move. At least it could still be useful in the meantime for developing better tree search techniques?
As far as intelligence is concerned, playing board games doesn't really do much. Games which map out onto the real world like FPSes and driving games stand a better chance of developing AI techniques that are going to be used to make properly intelligent* or at least useful robots.
I have never considered income to be directly proportional to intelligence. It's kind of a bell curve. The smartest people are in the middle and probably get paid about right - management and unskilled labourers tend to get paid amounts quite disproportionate to the amount of work they do or how intelligent they are. Chess skill isn't directly related to intelligence (as in generally accepted IQ 'intelligence' - there are plenty of different types of intelligence recognised in Psychology) anyway - a highly intelligent man who hasn't ever played chess before would probably lose to a girl of average intelligence, as long as the child has a lot of experience.
the article summary "a Women's Grandmaster played two games against Rybka; the result? Rybka won both games!" is probably a troll because Rybka could beat any human
it's not the the number of cycles you have, but how you use them.
True to an extent, but using the same algorithm on each machine in a timed game would result in the 40-core box trouncing the PDA - especially in the mid-game where there are a crazy amount of possible paths.
Deep Blue had a database of endgames to use (and possibly starting techniques, been a while since I read about it) - a mobile could use the same technique to do well near the end without doing much calculation, though it's not exactly a very 'interesting' approach from an AI point of view. And it probably won't even survive until an endgame scenario if it's anything as bad as I am compared to a decent strength computer chess opponent.
This story, for example, apparently has zero comments.
Here I was thinking I could get a Turk related first post:( At least Hiarcs almost rhymes with Turk. I didn't actually realise that the name was a reference to this - should have known what with the rest of the series being full of references.
Is anyone really surprised that a computer beat a human grandmaster twice in a row? As computers get more powerful, it is inevitable that they will completely outclass humans in games with perfect knowledge. There were times when Kasparov fooled Deep Blue's algorithms a couple of times (presumably giving it a piece or two in return for a square with significant strategic advantage), but they just trained it so that it wouldn't fall for those tricks again. I honestly wouldn't be surprisd if a grandmaster never won a tournament game against a chess supercomputer ever again.
But you would have to add that time on to any other OS that was booting at the time as well, surely? Could you not disable the BIOS memory check? This is all before my time as far as PCs are concerned, the oldest PC I ever made serious use of was a 386 laptop with Windows 3.1.. before that it was all Macs and Amigas.
I loved Amigas, but it's been a decade since I was using one regularly (my last specimen was an A1200 with 30Mhz 68030EC, 16MB RAM and a 180MB HDD!). Workbench doesn't really count as 'modern' anymore, despite being one of the great OSes*.
*this is obviously just my personal opinion. I had for a while still hoped that Amigas would take over the world, because I am an idealistic fool
That's probably because iPhones aren't really advanced. I had a 3G touchscreen smartphone (HTC TyTN) about a year and a half before the first gen iPhone even hit the streets.. I have always liked Macs since I was a kid (we had a Mac Classic which I used to play games on, write up my homework on, and I even did a bit of coding on it), but iPods and iPhones don't interest me too much. I doubt you'd be able to do anything useful with a Cell PCIe card unless you are heavily into scientific research, cryptography, games development or something like that tbh.
Unless this thing is cheaper than a graphics card and your system can't handle HD graphics streams then it will remain pointless even for most geeks.
Did you see the article about the Pandora console? Now that is IMO a geek toy worth blowing some cash on (and thankfully I was on the mailing list so I got an order in before the site was slashdotted):)
Any PCIe card is a 'mac version' just as much as it is a 'PC version' - perhaps you mean will there be drivers or a developer API for the Mac - the good thing is that a lot of Linux geeks will be wanting this (probably good for University research projects), and if there is Linux support then basically you will already have OSX support.
The interesting question is, what are you planning to do with it that you can't already do fast enough with a multicore CPU, GPU or physics type add in card? Or do you just want this because it's there? I'm not meaning to criticize especially, I tend to waste a lot of money on gadgets myself..
Slashdot Mirror
Actually, if you type that in the first result you get is a link to this page. So yes it is that good, in a way! ;)
You are of course assuming that exactly 65% of the geeky crowd are lazy, which doesn't necessarily follow even if 65% of the whole of /. are lazy ;)
I just thought "which font size?", decided on showing the kids a book with a certain font size, and didn't bother to work out the rest because I didn't consider it useful. I don't often obsess over primes and other mathematical oddities. I guess I'm not really a hardcore geek. 55-60 times is pretty impressive though!
Google's built in calculator and unit converter is indeed an excellent resource :) It even converts currency at current values (don't know how often it updates the conversion rates though).
Dates in Grand Theft Auto IV don't count
Yeah, I don't know how many petabytes (or whatever metric would be suitable) would be required, but it probably will still happen someday. I'm sure people 200 years ago would have considered it impossible to carry around video recordings of hundreds of theatre productions or concerts in their pocket, but these days a large percentage of people in the world can do that easily on their media player or phone. These same pocket devices that can do much more than computers that used to take up whole buildings 50 years ago.
Holographic storage for example could be a good medium for storing data far more densely than we currently can (though again I don't know much theory in this area). But just as 200 years ago people didn't have any clue about transistors, perhaps we will discover some completely new concept that will enable us to conveniently and cheaply store more data than even holographic storage can. Don't want to sound like a complete pie in the sky nutjob, but it's true that technology is currently advancing at a phenomenal rate, and we have no way of knowing the state of computing in 20 years, let alone 200.
Sorry, I forgot the part where the guy who patented the product tried to get money from the actual author. He probably would have gotten away with it otherwise. He's not just a criminal - he's a complete moron as well!
I'm surprised it got as far as court.
Perhaps the jerkoff who stole the source either didn't expect the guy to be able to prove anything, or just didn't think he'd put any money towards a court case since technically he hasn't lost any money in the first place.
I'm glad the author is standing up for what is right anyway, and hope the offeder's reputation is ruined enough that nobody will ever pay for his software again..
Indeed, it just needs a kickstart to break out of routine. I naturally go through phases of being interested in different stuff anyway, I just want to try and make my next several phases something that will get me coding some games again :)
Unfortunately my attempt to get myself back into photography again just involved me buying an expensive camera and it sitting on a shelf in my cupboard. Despite the fact that I've got into a habit of going for walks on weekends rather than driving everywhere, I never think to bring my camera with me. As a student I used to wander around town just taking pictures of random stuff and uploading it to deviantArt, but I stopped doing that when I got a job.
I need to think of some solid project to work on for the Pandora before I get it otherwise it may suffer the same fate :( A port of Racer could perhaps be a good way to do that. I had wanted to write a car simulator around the time that I started university - I loved Gran Turismo on the PSX but there weren't really any good alternatives for the PC at that time, and since the PlayStation was at home I had nothing to play! Unfortunately I didn't have much knowledge about how tyre grip, suspension etc worked back then (though I think I could have a good go at it these days). A year or 2 later I found Racer - seems that some other guy had the same idea around the same time as me, but actually carried it out. Perhaps I should just try writing my own from scratch - I'd learn a lot more that way really. I used to dream of working for a games development house until I found out that most people at these places get paid sod all ;) So now I'm kind of like you, I've been doing a bit of web/database development, and some other engineering simulation apps for simulating performance of the tools our company designs and builds (I didn't design the equations for them, but at least I understand what is going on most of the time!), the rest of the time I'm IT support, but at least I get good pay and benefits like a company car, phone, etc. I do enjoy it, it's a great place to work, but it's not as intellectually challenging as working as a games developer would be..
I was going to do my honours in AI (and did all the course work including the AI classes, but just never actually finished my final project due to some stupid personal reasons), so yeah I do have a bit of an idea about this stuff ;)
I was interested in neural networks towards the end of high school and found NNbot for counter-strike, which resulted in me doing my own bot. I actually read about neurons in the Psychology section of encyclopaedia and was thinking about how I'd simulate them, how you'd sort out which neuron to simulate first and so on considering they all run in parallel - I'm pretty sure that was before I had heardÂthe term neural network as well, but I'd never actually tried implementing one myself.
Then I found a counter-strike bot called NeuralNetwork Bot (NNBox), I was extremely impressed :) My dad saw how enthusiastic I was about it and helped me get started writing my own CS bot and I spent a year or more on that, it was pretty good in the end (the website is still up at planethalflife.com/teambot , but the last version was only compatible up to around Beta 1.4 and 1.5 because they released official bots for CS after that, and I was at University by then anyway).
Anyway, compared to writing that sort of AI (pathfinding to a goal, and my bots also laid information points along the route say if they died somewhere, they'd be more cautious the next time around, perhaps throw in a flashbang first, etc) I think I'd find Chess AI very dry. The final year project I was assigned was going to involve the board game diplomacy, and I just wasn't really motivated for it. The project I actually wanted to do involved a robot, can't remember exactly what it was even designed to do. The guy who was assigned to that project actually swapped with someone else! I couldn't believe it, I didn't even try to swap with him because I didn't see how anyone couldn't be interested in that project :p
Anyway, the whole reason I came to write this post was your paragraph
"I need to figure out a way to open this bottle. Out of all the knowledge I have about the world, what should I consider? Should I consider things I know about flowers? Well no. Duh. Should I consider things I know about cats? Well no. Duh. Maybe, I should consider things about *bottles* and gee, I don't know, what else.. hmm.. how about the parts of the bottle, maybe a common part of a bottle that most bottles have: lids. Now, do I know anything about *opening*? Why yes, I do.. I should probably look at all the ways I know to open something and see if any of the things I've opened in the past are like bottles. Maybe I should try to determine what *type* of lid this bottle has on it."
This got me thinking, if anyone ever designs an AI that learns its predicates by parsing information from the internet, we're in trouble. Something like Wikipedia would be mostly okay, but can you imagine if it got onto Myspace or Slashdot comments? Then created its own Myspace page or started commenting itself? Would be pretty funny to see some of the stuff it would come out with, its political and religious alignment etc :)
I didn't say that's not the way it should be, as not every job is about hard labour, running a company is high stress, technical knowledge takes years of training etc - I was just saying that pay doesn't really relate to intelligence.
I'm sure there are a lot of people like me who could have studied business management and get paid more if they actually found it interesting. Actually I wouldn't even need to do that - I could just go work offshore and almost make as much every day as I currently make in a week. I could easily do the whole living isolated for a couple of months at a time thing.
But anyway, I find it more interesting working with technology than working with stuff like human resources or even working offshore (though I would consider it just for a bit of a change maybe for a couple of months if we're struggling to assemble teams). In a similar vein, I vastly prefer playing FPS games to RTS games.
From there, it's just a matter of designing a computer that can learn, and giving it the online records of every regulation chess match ever played, and letting it figure out how to be the Grandest Master. Once its learning is in place, it's trivial to copy and redistribute that knowledge.
I think you're romanticising it somewhat. You're mostly right about the way humans play, though grand masters don't just play on instinct, they will also perform small mental simulations for several moves ahead.
The day will come where computers can just have a database of every single possble chess move in any game. Then it will no longer be a 'game' where computers are concerned, it will just be a matter of who plays first (unless of course the winning play comes down to starting second - I seem to do better when I play black myself, as I prefer to defend than to attack). Computers currently still trim down seemingly fruitless path choices sure, but if it they had as good algorithms as you think, they wouldn't need to be run on a supercomputer. Computer hardware is are far better than our own brains for games like chess. Any game with 'complete information' will eventually be able to be bruteforced by any computer, no smart algorithms required.
What is interesting for proper AI is games which involve 'incomplete information' ie an element of chance or some other unknown, like certain variations of poker or any game involving dice. That is a completely different category to games like chess and checkers where you know exactly what options your opponent has. Then the computer really has to be either taught good styles of play, or learn for itself as you suggest. Chess computers that can 'learn' will ultimately be completely unecessary when the game can be easily bruteforced on a home computer, or there is an online database somewhere that can tell you the exact move to play to maximise your chances of winning. When it comes to certain variations of poker against good human players, always playing the mathematical odds is not going to win the game.
So a chess board is nothing like an abstraction of the same problem as a computer that can learn what 'love' is. Consider that most humans (or perhaps no human) cannot define accurately, that makes it basically impossible for a computer to learn what it is. It is a function of inbuilt genetics, emotions, hormones, whatever. You could even say it has spiritual elements. A computer has none of those things, they would all have to be simulated to some extent, rather than learned.
You are right that the brain is an amazing computer, but it is good at different things from a Von Neumann modelled computer. We can't perform billions of mathematical operations a second like a supercomputer can. Then again, for the things that we do, we don't need to. We can often just 'tell' what is going to happen in certain situations because we have learned as you point out - we can see that something is likely to topple over and reach out to stop it. For a computer to do all these things would require a lot of work, especially if you want it to 'learn' to do them rather than just hardcoding in ways to interpret visual stimuli and build a 3D physics model of what is being seen, then simulate what is going to happen in the next few seconds, and then let it know how to move the hand, then how to move it with accurate 'hand to eye' co-ordination and so on. Anyway, sorry I'll stop rambling, but I think people greatly underestimate the difficulties involved with AI, and romanticize the idea of a computer learning to be 'human'. Often we don't want a computer to be human - we want it to be better than human, because humans make mistakes. A chess computer that makes mistakes is no good to anyone (unless they are just looking for an ego boost).
It wasn't flamebait. Possibly passive aggressive, that sounds a lot like me, but I don't think anyone is going to get a kick out of this card unless they are at least a serious amateur coder. At least physics cards have a few games that work with them - this thing is not meant for anything as common as gaming.
I was thinking about it a bit more and realised it would be a good way to get to know the cell architecture if you are planning on doing something with it in future, like getting into PS3 games development or any other areas where the Cell architecture has been designed to excel - but iamwhoiamtoday doesn't sound like he fits the bill for that. Nobody should buy one of these cards just because they have a free PCIe slot.
Nobody said it was
Often, apparent differences in intelligence (e.g. income)
Way to self-pwn yourself, dumbass.
No, I've never written drivers so I'm quite ignorant in that area. I guess I was just thinking more of a linux library which takes advantage of the abilities of the card, but you would indeed need a driver for that to be useful.
Definitely. I used to do a lot of hobbyist coding at home (been 7 years since I left home for Uni), but after I started university I spent most of my free time socialising, and since then I've not done any coding outside of work because I've just been wanting to relax.
I miss doing my own projects at home though - the coding I do at work is pretty mundane. It can still be rewarding sometimes, but compared to creating a game that you can play, and creating AI opponents etc, writing up an engineering or productivity application is pretty dull! I just need to better organise my time - I'm in the habit of frittering it away watching movies and playing games these days :/
Yeah. I enjoy playing chess, but Chess AI is only useful in a very limited domain, and eventually it won't really be necessary once computers can just brute force every move. At least it could still be useful in the meantime for developing better tree search techniques?
As far as intelligence is concerned, playing board games doesn't really do much. Games which map out onto the real world like FPSes and driving games stand a better chance of developing AI techniques that are going to be used to make properly intelligent* or at least useful robots.
*for certain definitions of 'intelligent'
I have never considered income to be directly proportional to intelligence. It's kind of a bell curve. The smartest people are in the middle and probably get paid about right - management and unskilled labourers tend to get paid amounts quite disproportionate to the amount of work they do or how intelligent they are. Chess skill isn't directly related to intelligence (as in generally accepted IQ 'intelligence' - there are plenty of different types of intelligence recognised in Psychology) anyway - a highly intelligent man who hasn't ever played chess before would probably lose to a girl of average intelligence, as long as the child has a lot of experience.
the article summary "a Women's Grandmaster played two games against Rybka; the result? Rybka won both games!" is probably a troll because Rybka could beat any human
Exactly.
it's not the the number of cycles you have, but how you use them.
True to an extent, but using the same algorithm on each machine in a timed game would result in the 40-core box trouncing the PDA - especially in the mid-game where there are a crazy amount of possible paths.
Deep Blue had a database of endgames to use (and possibly starting techniques, been a while since I read about it) - a mobile could use the same technique to do well near the end without doing much calculation, though it's not exactly a very 'interesting' approach from an AI point of view. And it probably won't even survive until an endgame scenario if it's anything as bad as I am compared to a decent strength computer chess opponent.
This story, for example, apparently has zero comments.
Here I was thinking I could get a Turk related first post :( At least Hiarcs almost rhymes with Turk. I didn't actually realise that the name was a reference to this - should have known what with the rest of the series being full of references.
Is anyone really surprised that a computer beat a human grandmaster twice in a row? As computers get more powerful, it is inevitable that they will completely outclass humans in games with perfect knowledge. There were times when Kasparov fooled Deep Blue's algorithms a couple of times (presumably giving it a piece or two in return for a square with significant strategic advantage), but they just trained it so that it wouldn't fall for those tricks again. I honestly wouldn't be surprisd if a grandmaster never won a tournament game against a chess supercomputer ever again.
But you would have to add that time on to any other OS that was booting at the time as well, surely? Could you not disable the BIOS memory check? This is all before my time as far as PCs are concerned, the oldest PC I ever made serious use of was a 386 laptop with Windows 3.1.. before that it was all Macs and Amigas.
What part of 'modern' did you miss?
I loved Amigas, but it's been a decade since I was using one regularly (my last specimen was an A1200 with 30Mhz 68030EC, 16MB RAM and a 180MB HDD!). Workbench doesn't really count as 'modern' anymore, despite being one of the great OSes*.
*this is obviously just my personal opinion. I had for a while still hoped that Amigas would take over the world, because I am an idealistic fool
I dunno, the whole thing is frankly very ambiguous
That's probably because iPhones aren't really advanced. I had a 3G touchscreen smartphone (HTC TyTN) about a year and a half before the first gen iPhone even hit the streets.. I have always liked Macs since I was a kid (we had a Mac Classic which I used to play games on, write up my homework on, and I even did a bit of coding on it), but iPods and iPhones don't interest me too much. I doubt you'd be able to do anything useful with a Cell PCIe card unless you are heavily into scientific research, cryptography, games development or something like that tbh.
Unless this thing is cheaper than a graphics card and your system can't handle HD graphics streams then it will remain pointless even for most geeks.
Did you see the article about the Pandora console? Now that is IMO a geek toy worth blowing some cash on (and thankfully I was on the mailing list so I got an order in before the site was slashdotted) :)
Any PCIe card is a 'mac version' just as much as it is a 'PC version' - perhaps you mean will there be drivers or a developer API for the Mac - the good thing is that a lot of Linux geeks will be wanting this (probably good for University research projects), and if there is Linux support then basically you will already have OSX support.
The interesting question is, what are you planning to do with it that you can't already do fast enough with a multicore CPU, GPU or physics type add in card? Or do you just want this because it's there? I'm not meaning to criticize especially, I tend to waste a lot of money on gadgets myself..
The whole point is that this is a way to get Cell power in "Personal Computers", rather than supercomputers or games consoles.
a good whiskey called Knob Creek
Now that's a woody Bourbon if ever there was one!
I'll get me coat.