Its a negative feedback loop, and one that some men find very hard to see their way clear from
Actually, it's a positive feedback loop. Negative feedback works to cancel the orignal disturbance while positive feedback reinforces the original disturbance. In your example the subject is led through frustration to do more of what made him frustrated in the first place: positive feedback. This positive feedback will lead to oscillation if the system's loop-gain is greater than 1.
Ahh, but all science _IS_ quantum physics, in that if you apply the rules of QM to your system, it will give the correct answer.
Ahh yes, the reductionist theory of science. Are you familiar with the field of non-linear dynamics (chaos theory)? Maybe you could work out human biology using quantum physics but there are two serious problems. First, we don't have the physics yet to do it, even with infinitely fast computers. It may "exist" (in the platonic sense) but we don't have it. Second, you are mixing quantum theory with DeCartes' "clockwork universe". Basically, maybe we could figure everything out with math, but only if we knew initial conditions to infinite precision. Impossible.
Also, quantum theory isn't the end-all-be-all. We don't know how to use it to explain charge-parity-time (CP) violation yet, for example. There are theories, but not proven.
If you really think we can do everything with quantum mechanics, I dare you to solve some fluid-dynamics problems with an abacus.
I guess all the scientists during World War II and before were "chalkboard scientists", because the use of the chalkboard permeated all fields of science, from physics and chemistry to biology and medicine.
Wait, they were all "pencil and paper" scientists as well. Damn, were they well-educated or what?
And Archimedes was a "stick and dirt" scientist, right?
Give me a break. The computer is a tool. A very powerful tool, in fact indispensible now, but a tool nontheless. I'm an Electrical Engineering researcher, and I spend a lot of time writing computer programs for my research in a variety of languages. Please don't call me a Computer Scientist, though, or I might just throw up. I use oscilloscopes a lot too. Does that make me an "oscilloscope scientist"?
Computer science is a well developed discipline in which very smart people devise new ways to solve problems. People in other fields, like me, use what computer scientists come up with. We are not computer scientists in our own right.
I remember from my digital electronics class that in many ways, it makes the most sense to have tri-state (instead of binary) electronics
You shouldn't use the term "Tri-State" is this context. Tri-State, a copyright of National Semiconductor, means drivers that can be put into a high-impedence output mode and so be disconnected from a bus in a simple way. What you are referring to is called "Multiple Valued Logic" and has been researched forever. It has found it's way into a few products (ROMs most notably) but in general is more work than it's worth.
The last time I checked, there wasn't any sort of high frequency clock signal running down my spine.
That's because your brain is an analog computer, not a digital one. As for "if the brain can do it, it must be possible", that is simply not true. We are still in the dark ages as to what the brain actually does and how it actually does it; and we won't be able to use any of our discoveries in information processing technology for the forseeable future.
And no, artificial neural networks are NOT analogous to how the brain operates. It is most useful to think of them purely as mathematical creations. They are orders of magnitude simpler than the networks found in the brain; and their operations are at BEST guesswork.
Now that's the real trick, isn't it? In modern integrated circuit design the interconnect uses up more area than the transistors. Even if you could do all the wiring in other layers (by the way, only VERY recently have ICs come out with lots of layers. One or two wiring layers was the standard for YEARS) you would still need lots of vias to move the signals between layers and down to the transistors.
And as for cost, I just checked MOSIS and if you needed a 6.5mm by 6.5mm square silicon chip fabbed, MOSIS would charge you about $70000 for a lot of 25! Kind of pricey for a 68k processor, don't you think?
There are a lot of reasons silicon is useful, and I'd be VERY suprised if people started printing chips out on their desks.
Can you imagine how delicate a processor of this nature would be. I would expect that the tiniest magnetic field could potentially disrupt the entire process. If we're talking quantum computing, then the spin state can represent a near-infinite number of positions, correct? If this is the case, the slightest fluxuation in the spin state would cause data-corruption.
Corporate Research Director, circa 1960, discussing invention of Integrated Circuit : Can you imaging how delicate a processor of this nature would be? The tiniest speck of dust could potentially disrupt the entire process. If we're talking digital computing, the superposition of all transistor outputs can represent a near-infinite number of positions, correct? If this is the case, the slightest fluxuation in substrate impedence could cause cross-talk and lead to data-corruption.
Isn't it just amazing we can build GHz processors in sub 0.2 micron technology with 100s of millions of transistors on a single substrate, given the sorry state-of-the-art in 1960? I'm very skeptical of quantum computing (I'm even more skeptical of biological computing), but I think off-the-cuff dismissals of the technology is dangerous. People have a way of making such statements look really dumb 40 years on.
However, this is purely based on assumptions. I could be completely wrong. I still do math using my 10 fingers.
THIS is your question?? What would we do differently?? A better question would be HOW. With no computers, try designing (let alone manufacturing) a chip.
Actually, people used to design chips without computers all the time. Computers are used to make the process more efficient. They are necessary now only because of the extreme complexity of current chips, not because of any inherent need of computers.
Basically, we would re-populate the world in a very similar way that we design a compiler. We would design a very simple computer using very simple gate-level chips that don't need computers to be designed. Then, we would build the computer (it would be as large as a mainframe). We could then use this computer to design a more powerful computer. And so on. This is like compiler design in that a very rudimentary compiler is implemented and then used to compile its own components.
I'm a professional integrated circuit designer, and I can tell you that there are a lot of useful circuits that I could design without a computer. For high performance stuff, of course I would need a powerful computer for extensive simulation, but designers could easily design simple analog and digital circuits without them.
As for the semiconductor processing needed to make these simple chips, a lot of manual fab equipment exists, mostly in university teaching and research labs. While we would have to put the sub-micron fab lines at Intel in storage, I can tell you that we could build simple 2 micron chips by hand at UC Santa Barbara TODAY.
In summary, this is not something to worry about. Because, presumably, human knowledge would not be erased, I think we could bootstrap outselves back to modern technology in about 10 years. It would take that long only because of the many generations of simpler computers used to design more powerful computers required.
It's called silicon. If you heat it up (a process called annealing) the lattice reorganizes into a lower energy state. This has been known for many years and is one reason why we all use silicon chips and not gallium-arsenide. (there are many other reasons as well of course).
You can even repair a broken pixel in a TFT Flat-Panel display by putting into an oven at 350degrees Fahrenheit. I don't recommend this, of course, because all the other components of your laptop will melt!
Well that's irony for ya, I guess. With the mod-ups of my recent above post, I just hit 50.
That's not irony. For something to be ironic it needs to imply the opposite of reality. For example, a big guy named "tiny" is ironic because in reality he is the opposite of tiny. Also, Pacific Bell DSL's customer service is ironic in that "service" is the exact opposite of what they provide.
I mean, think about it...ram is not _that_ pricey. There must be a lot of research dollars to compensate for. You can get 512megs of pc133 for ~US$200, so why does it cost outrageous sums for the drives?
Solid-State hard drives are so expensive because they use SRAM, not the DRAM you are referring to. SRAM, or Static RAM, is an entirely different memory technology from DRAM, or dynamic RAM. It's main selling point is that it is MUCH faster than DRAM. The problem with it is it is much less dense than DRAM and uses a lot more power. These things make it much more expensive than DRAM which is why you don't use it to expand the memory of a PC. In fact, SRAM is the kind of memory used in on-chip cache in microprocessors because of its extreme speed.
What's the big deal? Am I just missing somehting? I'm really not trying to start a flame war here, I just don't get it...
You are missing two things: speed and volatility.
1. SPEED: A Solid-State hard disk is made out of static RAM (SRAM) not the dynamic RAM (DRAM) that consitutes the user RAM in a PC. SRAM is what is used in on-chip cache and is MUCH faster than DRAM because it stores information actively and has physical amplifiers in each memory cell (usually SR-latches), rather than passively storing the information on a capacitor as in DRAM. Because of this it is also much more expensive and burns more power than DRAM. That is why these solid-state hard disks are so expensive.
2. VOLATILITY: When your computer crashes, or you shut it down, your RAM disk is GONE. This means you have to periodically write it to a physical hard-disk. With a solid-state hard disk, it looks to the computer just to be an amazingly fast hard drive, and no memory-management overhead is required. This is a big deal to large data warehouses and data mining operations.
The real selling point to the solid-state hard drive is the speed. Internal SRAM can operate upwards of 1 GHz, and although it can't communicate with the outside world at that speed of course, with advanced high-speed digital signaling technologies you can achieve latencies and throughput unheard of with regular hard-disks and even DRAM based RAM-disks.
But then, this is still vapor, so they're probably counting on a.13u process being readily available by then.
No, it's not. The Solid-State Circuit Society requires all papers presented at ISSCC to be based on measurements of physical prototypes, not simulations. So, the chip has been fabricated, and it does work, or it wouldn't be at the conference.
I assume the mean that the wafer is 21.7x21.3mm^2, this is a little under an inch to a side
No, the wafer is probably eight inches or more in diameter. The 21.7x21.3mm^2 refers to the size of the silicon die. Many of these will be fabricated simultaneously on one wafer, which is how semiconductor manufacturers get economies of scale.
The package size depends most strongly on the material used and the number of pins the chip requires (for I/O, power, ground, etc.)
Re:Yeah but, chip making isn't as easy as writing
on
Open-Source Processors
·
· Score: 2
Designed chips is very much like programing.
Yeah, if you want gigantic, slow, hot, wasteful chips. If you want high-performace chips, like we're talking about here, you would at least have to design the datapath by hand. And have you ever synthesized memory, by chance? Probably not. I bet you would say "just get a memory compiler". What about clock generation and distribution? What about high speed digital signaling?
On the other hand, attitudes like yours are good for people like me who design chips for a living. You see, when people that think "designing chips is like programming" start trying to design chips, that is true job security for people who actually know what they're doing.
A couple of weeks ago, when this story first came out, didn't Sega representives deny reports they were dropping Dreamcast, and now they confirm it? It seems to me they were lying before. Why do we as a society allow people to get away with bald-faced lies like this? If something is true shouldn't they agree or at very least have no comment? It makes me distrust everything they have to say. Who gives a damn about their future products, if they lie to their customers?
This kind of stuff (blatant lying with no consequences) happens all the time in business and politics. When are we going to demand people start telling the truth?
Metal? But metal is a conductor? The reason we use SEMIconductor is we can, after appropriate doping, select whether they conduct or insulate with a control voltage. How do you suppose we would do that with METAL?
Re:I'm sure this is all wrong
on
Plastic Valley?
·
· Score: 2
The reason they use silicon is because IT'S CHEAP! IT IS JUST (refined) SAND! Far better would be Gallium Arsenide or some similar compound. But that is orders of magnitude more expensive to do in volume (at the moment).
I have to disagree with you strongly here. Gallium Arsenide is NOT far better than silicon.
1. It is impossible to integrate GaAs to anywhere near the integration levels of Si because GaAs has such a HIGH defect rate. Trying to do a Million Transistor chip in GaAs would be futile; you would have no yield at all.
2. GaAs has a large amount of "surface" charge and it does not have a high quality native oxide. This means MOSFETs are next to impossible to create in GaAs (or any III/V semiconductor).
3. While GaAs does have a very high electron mobility (which is why it is faster) it actually has a LOWER hole mobility than silicon. This means complementary structures such as CMOS are not useful in GaAs. This is a real problem because complementary structures have no static power dissipation.
4. GaAs, like all III/V semiconductors, is mechanically brittle. This is because it is formed as a superlattice of two different materials with slightly different lattice constants. This is a big reason why it is more expensive to make than silicon, it is just plain harder. More to the point, because it is a bulk material, silicon can actually repair it's own lattice when heated. This is a process called annealing and is used in chip manufacturing.
5. It is difficult to design high performance, low power analog circuits in GaAs. It is used for stand alone power amps in wireless devices, but mixed analog/digital systems are quite hard in III/V materials. If you think mixed analog/digital systems aren't the future, why do you suppose Intel is buying every analog company it can get its hands on?
In summary, I would submit silicon is dominant for two reasons. First, because it is the best. Second, because it is cheap. It would be more correct I think to say we use CMOS over bipolar silicon technology because it is cheap.
I don't see any reason to use the same compiler - it's the end result which matters.
Then you obviously aren't thinking. If you don't use the same compiler, then how do you know you aren't simply testing how good your Itanium to X86 cross-compiler is and not how good the hardware is.
To put it another way, imagine that we wanted to see if Red Hat Linux or Windows ME was faster on a specific processor. If I did such a test, using the best Windows C complier on the market and, for Linux, using some weak complier I cobbled together in my undergrad compilers class, everyone on Slashdot would be flaming the hell out of me when Windows wiped the floor with Linux because it was such an unfair test.
In my hypothetical test, I would say the results reflected the difference in quality of the compliers, not the operating systems.
You don't seem to understand how a wireless system is partitioned.
All 5 million people could be online at once, but they are not served by a single base-station. There would be one base-station for every 1000 or so people. So, we would need 5e6/1000 = 5000 base-stations for all 5 million people to be online at once and still only use 200 MHz of bandwidth.
Cell-phone networks work the same way. Each "cell" is served by a base-station that can handle maybe 100 simultaneous calls in its area. When a call is made from cell-phone to cell-phone, each cell phone is actually communicating with its nearest base-station and the two base-stations are communicating with each other over a fiber-optic link. Because base-stations dynamically assign bandwidth to individual cell-phones requesting a connection, when two cell phones are talking to each other they could be on totally different frequencies.
In summary, when a wireless network is partitioned using cells, the bandwidth requirement is dependent on the number of users per base-station, not on the total number of users. Therefore, increasing the number of users from, say, 10000 to 5 million, only requires additional base-stations, NOT additional spectrum.
Wireless communication is great for cell phones and GPS and a bunch of other things, but when you start talking monstrous bandwidth, you need cable. Say, for instance, that 5 million New Yorkers want internet connections of 2 Mb/s each, and that your wireless technology can pack 10 bits per Hertz (that's really tight packing!).
5 million x 2 million / 10 = 1x10^12
So to give those folks their internet access, you need 1 THz (terahertz) of electromagnetic spectrum. But the whole usable spectrum is only about 300 GHz, and the FCC probably wants some of it for little things like radio stations, air traffic control, military communication, etc.;) Wireless connections won't work because there are too many people.
I don't agree with your example. You are assuming that all 5 million New Yorkers are simultaneously connected to the same base-station. Your example seems to "prove" even cell-phones are impossible. In reality, a network of base-stations, each connected to each other and the backbone by fiber, would be used to implement the last mile. If one base-station was used per 1000 customers, then only (1e3 * 2e6) / 10 = 2e8 MHz if required. So, using your assumption of 10 bits per Hertz, only 200 MHz of bandwidth is needed to implement the network city-wide. Agreed, it would be damn hard to find an economical A/D converter if we needed a high SNR, but it certainly IS possible.
And of course the Golden Rule is also reflected in Kant's Categorical Imperitive. When deciding if something is ethical,
ask yourself "what if everybody did it?"
That is a truly ridiculous argument for "an- eye-for-an-eye", mate. Do you think it is ethical for you to go down to the market for a bottle of Pepsi? Well, according to the Categorical Imperitive, no, because if EVERYONE when down to the market anarchy would ensue with riots and murders leading to pitched battles as the supplies of Pepsi dwindled, and could eventually lead to the downfall of Western Civilization. Give a break.
As for your, er, analysis of the Prisoner's dilemma, "tit for tat" maximizes only the two prisoner's COLLECTIVE expected utility. The best result for a given prisoner is to sell out the other prisoner given that the other prisoner doesn't talk. That's why it's called a dilemma, mate. Tit for Tat did NOT win, because if you believe the other prisoner is honest, you can screw him and do better for yourself than if you were honest.
I am a scientist and I don't believe an eye for an eye is OK or workable. We live in an obstensibly civilized society, and to forgive is divine.
For example, I claim that the following program is legal C++ code:
cout
However, I wouldn't dream of claiming that the program is "object-oriented" simply because C++ is perceived to be an "object-oriented" language.
You know what? cout is an object oriented function! You can send it strings, characters, integers, and floating point variables without having to indicate to the complier what you are sending, as you do when using printf() in C. Try a better example. In fact, since C++ is a superset of C, I would submit that THIS is a truly non-OOPed C++ program:
int main() {
printf("%s\n","hello world!");
}
Now, THAT would compile with g++ and doesn't use
anything OOP.
This in a way relates back to the Singapore story of the early 90's where that one American committed a crime in Singapore and recieved the "Singapore punishment" of a lashing. America begged and pleaded to stop the "cruel and unusual" punishment, but it was carried out.
This American was a student at the Singapore American School and was a classmate of a very good friend of mine. About 6 months or so before he got arrested (for graffiti, by the way) he secretly put two firearms into my friends bag on a flight from Malasia to Singapore (this guy and my friend were on the basketball team together). Thank God the guns were never found because my friend could have gotten life in prison or the death penalty. This guy, whose parents were on 60 minutes whining about how their "Good Kid" had made a mistake, deserved a LOT more than a caning. True, he is now scared for life, but he should have thought about the consequences of his actions and not that "He's an American, dammit!"
This guy was a grade A a**hole and he needed a serious attitude adjustment.
Slashdot Mirror
Actually, it's a positive feedback loop. Negative feedback works to cancel the orignal disturbance while positive feedback reinforces the original disturbance. In your example the subject is led through frustration to do more of what made him frustrated in the first place: positive feedback. This positive feedback will lead to oscillation if the system's loop-gain is greater than 1.
Ahh yes, the reductionist theory of science. Are you familiar with the field of non-linear dynamics (chaos theory)? Maybe you could work out human biology using quantum physics but there are two serious problems. First, we don't have the physics yet to do it, even with infinitely fast computers. It may "exist" (in the platonic sense) but we don't have it. Second, you are mixing quantum theory with DeCartes' "clockwork universe". Basically, maybe we could figure everything out with math, but only if we knew initial conditions to infinite precision. Impossible.
Also, quantum theory isn't the end-all-be-all. We don't know how to use it to explain charge-parity-time (CP) violation yet, for example. There are theories, but not proven.
If you really think we can do everything with quantum mechanics, I dare you to solve some fluid-dynamics problems with an abacus.
Wait, they were all "pencil and paper" scientists as well. Damn, were they well-educated or what?
And Archimedes was a "stick and dirt" scientist, right?
Give me a break. The computer is a tool. A very powerful tool, in fact indispensible now, but a tool nontheless. I'm an Electrical Engineering researcher, and I spend a lot of time writing computer programs for my research in a variety of languages. Please don't call me a Computer Scientist, though, or I might just throw up. I use oscilloscopes a lot too. Does that make me an "oscilloscope scientist"?
Computer science is a well developed discipline in which very smart people devise new ways to solve problems. People in other fields, like me, use what computer scientists come up with. We are not computer scientists in our own right.
You shouldn't use the term "Tri-State" is this context. Tri-State, a copyright of National Semiconductor, means drivers that can be put into a high-impedence output mode and so be disconnected from a bus in a simple way. What you are referring to is called "Multiple Valued Logic" and has been researched forever. It has found it's way into a few products (ROMs most notably) but in general is more work than it's worth.
That's because your brain is an analog computer, not a digital one. As for "if the brain can do it, it must be possible", that is simply not true. We are still in the dark ages as to what the brain actually does and how it actually does it; and we won't be able to use any of our discoveries in information processing technology for the forseeable future.
And no, artificial neural networks are NOT analogous to how the brain operates. It is most useful to think of them purely as mathematical creations. They are orders of magnitude simpler than the networks found in the brain; and their operations are at BEST guesswork.
Now that's the real trick, isn't it? In modern integrated circuit design the interconnect uses up more area than the transistors. Even if you could do all the wiring in other layers (by the way, only VERY recently have ICs come out with lots of layers. One or two wiring layers was the standard for YEARS) you would still need lots of vias to move the signals between layers and down to the transistors.
And as for cost, I just checked MOSIS and if you needed a 6.5mm by 6.5mm square silicon chip fabbed, MOSIS would charge you about $70000 for a lot of 25! Kind of pricey for a 68k processor, don't you think?
There are a lot of reasons silicon is useful, and I'd be VERY suprised if people started printing chips out on their desks.
Corporate Research Director, circa 1960, discussing invention of Integrated Circuit : Can you imaging how delicate a processor of this nature would be? The tiniest speck of dust could potentially disrupt the entire process. If we're talking digital computing, the superposition of all transistor outputs can represent a near-infinite number of positions, correct? If this is the case, the slightest fluxuation in substrate impedence could cause cross-talk and lead to data-corruption.
Isn't it just amazing we can build GHz processors in sub 0.2 micron technology with 100s of millions of transistors on a single substrate, given the sorry state-of-the-art in 1960? I'm very skeptical of quantum computing (I'm even more skeptical of biological computing), but I think off-the-cuff dismissals of the technology is dangerous. People have a way of making such statements look really dumb 40 years on.
However, this is purely based on assumptions. I could be completely wrong. I still do math using my 10 fingers.
Try using your head, it works better. =)
Actually, people used to design chips without computers all the time. Computers are used to make the process more efficient. They are necessary now only because of the extreme complexity of current chips, not because of any inherent need of computers.
Basically, we would re-populate the world in a very similar way that we design a compiler. We would design a very simple computer using very simple gate-level chips that don't need computers to be designed. Then, we would build the computer (it would be as large as a mainframe). We could then use this computer to design a more powerful computer. And so on. This is like compiler design in that a very rudimentary compiler is implemented and then used to compile its own components.
I'm a professional integrated circuit designer, and I can tell you that there are a lot of useful circuits that I could design without a computer. For high performance stuff, of course I would need a powerful computer for extensive simulation, but designers could easily design simple analog and digital circuits without them.
As for the semiconductor processing needed to make these simple chips, a lot of manual fab equipment exists, mostly in university teaching and research labs. While we would have to put the sub-micron fab lines at Intel in storage, I can tell you that we could build simple 2 micron chips by hand at UC Santa Barbara TODAY.
In summary, this is not something to worry about. Because, presumably, human knowledge would not be erased, I think we could bootstrap outselves back to modern technology in about 10 years. It would take that long only because of the many generations of simpler computers used to design more powerful computers required.
You can even repair a broken pixel in a TFT Flat-Panel display by putting into an oven at 350degrees Fahrenheit. I don't recommend this, of course, because all the other components of your laptop will melt!
That's not irony. For something to be ironic it needs to imply the opposite of reality. For example, a big guy named "tiny" is ironic because in reality he is the opposite of tiny. Also, Pacific Bell DSL's customer service is ironic in that "service" is the exact opposite of what they provide.
Solid-State hard drives are so expensive because they use SRAM, not the DRAM you are referring to. SRAM, or Static RAM, is an entirely different memory technology from DRAM, or dynamic RAM. It's main selling point is that it is MUCH faster than DRAM. The problem with it is it is much less dense than DRAM and uses a lot more power. These things make it much more expensive than DRAM which is why you don't use it to expand the memory of a PC. In fact, SRAM is the kind of memory used in on-chip cache in microprocessors because of its extreme speed.
You are missing two things: speed and volatility.
1. SPEED: A Solid-State hard disk is made out of static RAM (SRAM) not the dynamic RAM (DRAM) that consitutes the user RAM in a PC. SRAM is what is used in on-chip cache and is MUCH faster than DRAM because it stores information actively and has physical amplifiers in each memory cell (usually SR-latches), rather than passively storing the information on a capacitor as in DRAM. Because of this it is also much more expensive and burns more power than DRAM. That is why these solid-state hard disks are so expensive.
2. VOLATILITY: When your computer crashes, or you shut it down, your RAM disk is GONE. This means you have to periodically write it to a physical hard-disk. With a solid-state hard disk, it looks to the computer just to be an amazingly fast hard drive, and no memory-management overhead is required. This is a big deal to large data warehouses and data mining operations.
The real selling point to the solid-state hard drive is the speed. Internal SRAM can operate upwards of 1 GHz, and although it can't communicate with the outside world at that speed of course, with advanced high-speed digital signaling technologies you can achieve latencies and throughput unheard of with regular hard-disks and even DRAM based RAM-disks.
No, it's not. The Solid-State Circuit Society requires all papers presented at ISSCC to be based on measurements of physical prototypes, not simulations. So, the chip has been fabricated, and it does work, or it wouldn't be at the conference.
No, the wafer is probably eight inches or more in diameter. The 21.7x21.3mm^2 refers to the size of the silicon die. Many of these will be fabricated simultaneously on one wafer, which is how semiconductor manufacturers get economies of scale.
The package size depends most strongly on the material used and the number of pins the chip requires (for I/O, power, ground, etc.)
Yeah, if you want gigantic, slow, hot, wasteful chips. If you want high-performace chips, like we're talking about here, you would at least have to design the datapath by hand. And have you ever synthesized memory, by chance? Probably not. I bet you would say "just get a memory compiler". What about clock generation and distribution? What about high speed digital signaling?
On the other hand, attitudes like yours are good for people like me who design chips for a living. You see, when people that think "designing chips is like programming" start trying to design chips, that is true job security for people who actually know what they're doing.
This kind of stuff (blatant lying with no consequences) happens all the time in business and politics. When are we going to demand people start telling the truth?
Metal? But metal is a conductor? The reason we use SEMIconductor is we can, after appropriate doping, select whether they conduct or insulate with a control voltage. How do you suppose we would do that with METAL?
I have to disagree with you strongly here. Gallium Arsenide is NOT far better than silicon.
1. It is impossible to integrate GaAs to anywhere near the integration levels of Si because GaAs has such a HIGH defect rate. Trying to do a Million Transistor chip in GaAs would be futile; you would have no yield at all.
2. GaAs has a large amount of "surface" charge and it does not have a high quality native oxide. This means MOSFETs are next to impossible to create in GaAs (or any III/V semiconductor).
3. While GaAs does have a very high electron mobility (which is why it is faster) it actually has a LOWER hole mobility than silicon. This means complementary structures such as CMOS are not useful in GaAs. This is a real problem because complementary structures have no static power dissipation.
4. GaAs, like all III/V semiconductors, is mechanically brittle. This is because it is formed as a superlattice of two different materials with slightly different lattice constants. This is a big reason why it is more expensive to make than silicon, it is just plain harder. More to the point, because it is a bulk material, silicon can actually repair it's own lattice when heated. This is a process called annealing and is used in chip manufacturing.
5. It is difficult to design high performance, low power analog circuits in GaAs. It is used for stand alone power amps in wireless devices, but mixed analog/digital systems are quite hard in III/V materials. If you think mixed analog/digital systems aren't the future, why do you suppose Intel is buying every analog company it can get its hands on?
In summary, I would submit silicon is dominant for two reasons. First, because it is the best. Second, because it is cheap. It would be more correct I think to say we use CMOS over bipolar silicon technology because it is cheap.
Actually, for CMOS, Power = C(Vdd^2)*f
with Power in Watts, C (Capacitance) in Farads, Vdd in Volts, and f in Hertz.
Basically, this equation says to lower power dissipation, we can lower Vdd (power supply), lower the capacitance, or reduce the clock frequency.
Then you obviously aren't thinking. If you don't use the same compiler, then how do you know you aren't simply testing how good your Itanium to X86 cross-compiler is and not how good the hardware is.
To put it another way, imagine that we wanted to see if Red Hat Linux or Windows ME was faster on a specific processor. If I did such a test, using the best Windows C complier on the market and, for Linux, using some weak complier I cobbled together in my undergrad compilers class, everyone on Slashdot would be flaming the hell out of me when Windows wiped the floor with Linux because it was such an unfair test.
In my hypothetical test, I would say the results reflected the difference in quality of the compliers, not the operating systems.
All 5 million people could be online at once, but they are not served by a single base-station. There would be one base-station for every 1000 or so people. So, we would need 5e6/1000 = 5000 base-stations for all 5 million people to be online at once and still only use 200 MHz of bandwidth.
Cell-phone networks work the same way. Each "cell" is served by a base-station that can handle maybe 100 simultaneous calls in its area. When a call is made from cell-phone to cell-phone, each cell phone is actually communicating with its nearest base-station and the two base-stations are communicating with each other over a fiber-optic link. Because base-stations dynamically assign bandwidth to individual cell-phones requesting a connection, when two cell phones are talking to each other they could be on totally different frequencies.
In summary, when a wireless network is partitioned using cells, the bandwidth requirement is dependent on the number of users per base-station, not on the total number of users. Therefore, increasing the number of users from, say, 10000 to 5 million, only requires additional base-stations, NOT additional spectrum.
Wireless communication is great for cell phones and GPS and a bunch of other things, but when you start talking monstrous bandwidth, you need cable. Say, for instance, that 5 million New Yorkers want internet connections of 2 Mb/s each, and that your wireless technology can pack 10 bits per Hertz (that's really tight packing!). 5 million x 2 million / 10 = 1x10^12 So to give those folks their internet access, you need 1 THz (terahertz) of electromagnetic spectrum. But the whole usable spectrum is only about 300 GHz, and the FCC probably wants some of it for little things like radio stations, air traffic control, military communication, etc. ;) Wireless connections won't work because there are too many people.
I don't agree with your example. You are assuming that all 5 million New Yorkers are simultaneously connected to the same base-station. Your example seems to "prove" even cell-phones are impossible. In reality, a network of base-stations, each connected to each other and the backbone by fiber, would be used to implement the last mile. If one base-station was used per 1000 customers, then only (1e3 * 2e6) / 10 = 2e8 MHz if required. So, using your assumption of 10 bits per Hertz, only 200 MHz of bandwidth is needed to implement the network city-wide. Agreed, it would be damn hard to find an economical A/D converter if we needed a high SNR, but it certainly IS possible.
That is a truly ridiculous argument for "an- eye-for-an-eye", mate. Do you think it is ethical for you to go down to the market for a bottle of Pepsi? Well, according to the Categorical Imperitive, no, because if EVERYONE when down to the market anarchy would ensue with riots and murders leading to pitched battles as the supplies of Pepsi dwindled, and could eventually lead to the downfall of Western Civilization. Give a break.
As for your, er, analysis of the Prisoner's dilemma, "tit for tat" maximizes only the two prisoner's COLLECTIVE expected utility. The best result for a given prisoner is to sell out the other prisoner given that the other prisoner doesn't talk. That's why it's called a dilemma, mate. Tit for Tat did NOT win, because if you believe the other prisoner is honest, you can screw him and do better for yourself than if you were honest.
I am a scientist and I don't believe an eye for an eye is OK or workable. We live in an obstensibly civilized society, and to forgive is divine.
cout However, I wouldn't dream of claiming that the program is "object-oriented" simply because C++ is perceived to be an "object-oriented" language.
You know what? cout is an object oriented function! You can send it strings, characters, integers, and floating point variables without having to indicate to the complier what you are sending, as you do when using printf() in C. Try a better example. In fact, since C++ is a superset of C, I would submit that THIS is a truly non-OOPed C++ program:
int main() {
printf("%s\n","hello world!");
}
Now, THAT would compile with g++ and doesn't use anything OOP.
This American was a student at the Singapore American School and was a classmate of a very good friend of mine. About 6 months or so before he got arrested (for graffiti, by the way) he secretly put two firearms into my friends bag on a flight from Malasia to Singapore (this guy and my friend were on the basketball team together). Thank God the guns were never found because my friend could have gotten life in prison or the death penalty. This guy, whose parents were on 60 minutes whining about how their "Good Kid" had made a mistake, deserved a LOT more than a caning. True, he is now scared for life, but he should have thought about the consequences of his actions and not that "He's an American, dammit!"
This guy was a grade A a**hole and he needed a serious attitude adjustment.