Not for the data that the baloons give you. The key information that the baloon gives you is the wind, temperature, and humidity at different altitudes. Unfortunately, satellites are not currently capable of providing this for different altitudes. The data is important for weather forecasters, and particularly important for aviation weather forecasts.
Please explain how antibiotics affect virus particles in any way at all. For bonus points, attach an annotated diagram of a bacteriophage. Write on both sides of the paper. Either way up boy, I'm not bothered.
For extra bonus points, draw a diagram of missing the point.
The F-111 was in service into the 90s. I believe the EF-111 jamming plane is still in service. In fact, the F-111 proved itself a pretty good plane in Desert Storm.
The Yak-141 never made it into service. If it had, it would have been the first supersonic VTOL airplane in service. The JSF will take that crown now. The -141 would have revolutionized the capability of the Kiev VTOL carriers, which run Yak-38s, which are pieces of junk. They have low payloads, bad reliability, and poor electronics. The -38 isn't really a match for much of anything in the sky. It could probably be used for shooting down incoming attackers if they were unescorted, but it has such short legs that you're probably better off just firing off some cruise missiles than sending out the planes.
There are several pumped storage plants in the U.S. where you use electricity to pump water up a hill at night, and then let it run through hydroelectric turbines during the day. This works particularly well with Nuclear power where you can't really shut down the plant at night. The full-cycle efficiency is about 60-70%.
To reply to the first part of your message, yes, you're never going to assembly optimize large parts of your program (I assume that's what you mean by "dynamic" instructions). And it's true that if the program doesn't have a long running time, it probably doesn't matter if you optimize it. However, there are still situations where this kind of thing is very valuable. The last time I seriously got my hands dirty with asm was with a computer vision application that ran on 64 Pentium IIs. The objective was to reconstruct the shape of an object in real-time. This program spent over 90% of its time in a loop which was about ten lines of C++ (C really as that part didn't use any objects). Proper assembly optimization allowed a 400% improvement in overall speed. This was partially due to being able to effectively use MMX, which is very difficult for the compiler, because one has to organize the data in a very specific way to take advantage of MMX.
Unfortunately now there isn't a good reference for the P4. When I was doing P2 programming, I used the Intel manuals and Agner Fog's site. From these sources it is possible to know the pipeline cold. You can look at code and see almost exactly if you're going to run out of decode bandwidth, whether a branch is easily predictable, whether you're using all the functional units you can, etc. GCC doesn't know any of this. The Intel compiler knows some of this, but it is limited.
IA64 is a very different assembly language, and perhaps the compiler will be harder to beat. In an SIMD situation, I belive the human has the upper hand, as there are many things the compiler can't do because it can't prove that it will be equivalent. IA-64 is MIMD, so that's not so much of an issue. Ultimately, though the basic issue remains that a compiler has relatively little knowledge and ingenuity, but lots of patience. Compilers have been getting better but processors have been getting more complex. IA-64 changes that equation for sure. However, it's far from clear that IA-64 is going to win in the marketplace anyway, so we will see.
I've read the IA64 manual, I understand, and I've seen the code that comes out of the Intel compiler for IA64. If you're clever and you have enough time you can beat it. Ultimately the issue is that the compiler can only act at the level of description of the code. The compiler is never going to do something like reorganize your data structures to take advantage of a particular instruction. There are many things that the compiler can't assume in the general case that will speed your program up. If you know that it will work in your specific case, you can do it yourself. For example, C guarantees left-to-right evaluation. This may not matter in a lot of cases, but the compiler is required to honor it unless it can prove that it doesn't make any difference.
The compiler most certainly can be beaten, even more so today than in the past. I haven't done much asm programming on RISC machines, but on x86, the stuff the compiler puts out is generally garbage. If you're using GCC, it's trivial to beat, as it doesn't know how to deal with the extreme lack of registers on the x86. GCC is constantly going to memory when, with some rearangement, it's possible to keep many more things in registers. The Intel compiler is much better, but it still isn't hard to beat.
Furthermore, with the SIMD stuff in the newer x86 processors (MMX, SSE, SSE2), an asm programmer can get huge speedups which the compiler just doesn't know how to exploit. The Intel compiler will use these features in some instances, but far from optimally. Mind you, you have to know the processor well, and for the big wins, you have to optimize for a specific processor, but if you're doing computationally intensive stuff, the gains can be huge.
Paradise was great. I played in '95 and '96, and I always liked it much better than Bronco. Unfortunately after '96 or so, one would never find a full server.
We are burying a huge energy resource -- Plenty of good fissile Pu and U-235. If we used breeder reactors and reprocessing, we wouldn't have to bury this stuff, and we wouldn't have to dig new Uranium up.
From my understanding of the Feist decision, it is quite possible that this list is not even copyrightable. Unfortunately it would take a lot of money to make such an argument, as it would establish new case law. IANAL.
The message sent by vixie says that a contract will be drawn up, and outlines some of its features. It is not a contract in itself. Unfortunately, it is very easy to not worry about such things when a business relationship is starting up because everything is fine and you don't want to make the other party feel you don't trust them. Unfortunately if things go bad, all you have left is PGP signed vague emails instead of signed and notarised contracts. IANAL.
You are not responsible for what others may do after you take a prinicpled stand. If there is something that you feel is morally wrong and you refuse to do it you have met your obligation. If someone else is willing to do it, that is their decision, as everyone has to act according to what they feel is best. Just because you believe something is wrong doesn't mean that everyone does, and your ethical decisions are not binding on the entire world.
That's not true. You have a very simplistic understanding of the stock market. Without getting into all of the reasons why a company might want to hang onto a lot of cash, a company can buy back stock. This has the same effect as paying a dividend. One way of thinking about is, suppose there are ten shareholders, each holding 10 shares of stock. If each share is trading at $10 each, the company decides to use an extra $100 in cash to buy back ten shares instead of paying a $100 dividend, each shareholder can sell one share if its stock for $10. The shareholders then each get $10 from selling their stock, and they still each own 10% of the outstanding shares.
The big advantage of this is taxes. Instead of paying ordinary income taxes on the dividend, they pay taxes on the capital gain. Capital gains rates are lower than income taxes, so this is a major advantage with the same economic effect. There are other advantages to buybacks though, because you don't have to sell. Instead you can increase your ownership in the company without engaging in a separate transaction. Normally taking your dividends and reinvesting it in the company has tax consequences. A buyback does not.
ASCI Red I believe originally had Pentium Pro 200s in it. All of the processors were replaced in it a few years back with Pentium II 300 Overdrive chips, which is a drop-in replacement for the PPro. Incidentally, the PII Overdrive has a totally different cache structure than the regular PII, I believe it has a 256k full speed L2 cache.
Your resetting of marketing preferences without authorization has shown me that your company has no integrity. I have never given you my phone number, address, or credit card number, and I never will in the future. Your disregard for even the known explicit preferences of your users simply demonstrates that you have no respect for your customers, and that you are a dishonest group of people. Wouldn't you like the work of thousands of people to stand for more than earnings per share?
I'm well aware that there are encodings from turing machines to bits. My point is that, given the definition of a Turing machine, there is no one single canonical encoding that is objectively correct. Similarly, there are an infinite number of correct ways to encode any type of graph. Just because your textbook defines one way does not mean that it is the only way.
Slashdot Mirror
Clue: Life is better when you mellow out a bit.
Not for the data that the baloons give you. The key information that the baloon gives you is the wind, temperature, and humidity at different altitudes. Unfortunately, satellites are not currently capable of providing this for different altitudes. The data is important for weather forecasters, and particularly important for aviation weather forecasts.
Please explain how antibiotics affect virus particles in any way at all. For bonus points, attach an annotated diagram of a bacteriophage. Write on both sides of the paper. Either way up boy, I'm not bothered.
For extra bonus points, draw a diagram of missing the point.
For cracking codes, the NSA has computers that are far faster than this.
How do you explain qualia then?
The F-111 was in service into the 90s. I believe the EF-111 jamming plane is still in service. In fact, the F-111 proved itself a pretty good plane in Desert Storm.
The Yak-141 never made it into service. If it had, it would have been the first supersonic VTOL airplane in service. The JSF will take that crown now. The -141 would have revolutionized the capability of the Kiev VTOL carriers, which run Yak-38s, which are pieces of junk. They have low payloads, bad reliability, and poor electronics. The -38 isn't really a match for much of anything in the sky. It could probably be used for shooting down incoming attackers if they were unescorted, but it has such short legs that you're probably better off just firing off some cruise missiles than sending out the planes.
There are several pumped storage plants in the U.S. where you use electricity to pump water up a hill at night, and then let it run through hydroelectric turbines during the day. This works particularly well with Nuclear power where you can't really shut down the plant at night. The full-cycle efficiency is about 60-70%.
To reply to the first part of your message, yes, you're never going to assembly optimize large parts of your program (I assume that's what you mean by "dynamic" instructions). And it's true that if the program doesn't have a long running time, it probably doesn't matter if you optimize it. However, there are still situations where this kind of thing is very valuable. The last time I seriously got my hands dirty with asm was with a computer vision application that ran on 64 Pentium IIs. The objective was to reconstruct the shape of an object in real-time. This program spent over 90% of its time in a loop which was about ten lines of C++ (C really as that part didn't use any objects). Proper assembly optimization allowed a 400% improvement in overall speed. This was partially due to being able to effectively use MMX, which is very difficult for the compiler, because one has to organize the data in a very specific way to take advantage of MMX.
Unfortunately now there isn't a good reference for the P4. When I was doing P2 programming, I used the Intel manuals and Agner Fog's site. From these sources it is possible to know the pipeline cold. You can look at code and see almost exactly if you're going to run out of decode bandwidth, whether a branch is easily predictable, whether you're using all the functional units you can, etc. GCC doesn't know any of this. The Intel compiler knows some of this, but it is limited.
IA64 is a very different assembly language, and perhaps the compiler will be harder to beat. In an SIMD situation, I belive the human has the upper hand, as there are many things the compiler can't do because it can't prove that it will be equivalent. IA-64 is MIMD, so that's not so much of an issue. Ultimately, though the basic issue remains that a compiler has relatively little knowledge and ingenuity, but lots of patience. Compilers have been getting better but processors have been getting more complex. IA-64 changes that equation for sure. However, it's far from clear that IA-64 is going to win in the marketplace anyway, so we will see.
I've read the IA64 manual, I understand, and I've seen the code that comes out of the Intel compiler for IA64. If you're clever and you have enough time you can beat it. Ultimately the issue is that the compiler can only act at the level of description of the code. The compiler is never going to do something like reorganize your data structures to take advantage of a particular instruction. There are many things that the compiler can't assume in the general case that will speed your program up. If you know that it will work in your specific case, you can do it yourself. For example, C guarantees left-to-right evaluation. This may not matter in a lot of cases, but the compiler is required to honor it unless it can prove that it doesn't make any difference.
The compiler most certainly can be beaten, even more so today than in the past. I haven't done much asm programming on RISC machines, but on x86, the stuff the compiler puts out is generally garbage. If you're using GCC, it's trivial to beat, as it doesn't know how to deal with the extreme lack of registers on the x86. GCC is constantly going to memory when, with some rearangement, it's possible to keep many more things in registers. The Intel compiler is much better, but it still isn't hard to beat.
Furthermore, with the SIMD stuff in the newer x86 processors (MMX, SSE, SSE2), an asm programmer can get huge speedups which the compiler just doesn't know how to exploit. The Intel compiler will use these features in some instances, but far from optimally. Mind you, you have to know the processor well, and for the big wins, you have to optimize for a specific processor, but if you're doing computationally intensive stuff, the gains can be huge.
You might want to check your math yourself.
Paradise was great. I played in '95 and '96, and I always liked it much better than Bronco. Unfortunately after '96 or so, one would never find a full server.
We are burying a huge energy resource -- Plenty of good fissile Pu and U-235. If we used breeder reactors and reprocessing, we wouldn't have to bury this stuff, and we wouldn't have to dig new Uranium up.
Just make sure it has a toggle switch for amber/green screen emulation. Can't quite get the same effect without the old-school monochrome.
From my understanding of the Feist decision, it is quite possible that this list is not even copyrightable. Unfortunately it would take a lot of money to make such an argument, as it would establish new case law. IANAL.
The message sent by vixie says that a contract will be drawn up, and outlines some of its features. It is not a contract in itself. Unfortunately, it is very easy to not worry about such things when a business relationship is starting up because everything is fine and you don't want to make the other party feel you don't trust them. Unfortunately if things go bad, all you have left is PGP signed vague emails instead of signed and notarised contracts. IANAL.
You are not responsible for what others may do after you take a prinicpled stand. If there is something that you feel is morally wrong and you refuse to do it you have met your obligation. If someone else is willing to do it, that is their decision, as everyone has to act according to what they feel is best. Just because you believe something is wrong doesn't mean that everyone does, and your ethical decisions are not binding on the entire world.
It specifically forbids giving poisions.
That's not true. You have a very simplistic understanding of the stock market. Without getting into all of the reasons why a company might want to hang onto a lot of cash, a company can buy back stock. This has the same effect as paying a dividend. One way of thinking about is, suppose there are ten shareholders, each holding 10 shares of stock. If each share is trading at $10 each, the company decides to use an extra $100 in cash to buy back ten shares instead of paying a $100 dividend, each shareholder can sell one share if its stock for $10. The shareholders then each get $10 from selling their stock, and they still each own 10% of the outstanding shares.
The big advantage of this is taxes. Instead of paying ordinary income taxes on the dividend, they pay taxes on the capital gain. Capital gains rates are lower than income taxes, so this is a major advantage with the same economic effect. There are other advantages to buybacks though, because you don't have to sell. Instead you can increase your ownership in the company without engaging in a separate transaction. Normally taking your dividends and reinvesting it in the company has tax consequences. A buyback does not.
Touche.
That was carefully sterilized.
ASCI Red I believe originally had Pentium Pro 200s in it. All of the processors were replaced in it a few years back with Pentium II 300 Overdrive chips, which is a drop-in replacement for the PPro. Incidentally, the PII Overdrive has a totally different cache structure than the regular PII, I believe it has a 256k full speed L2 cache.
Your resetting of marketing preferences without authorization has shown me that your company has no integrity. I have never given you my phone number, address, or credit card number, and I never will in the future. Your disregard for even the known explicit preferences of your users simply demonstrates that you have no respect for your customers, and that you are a dishonest group of people. Wouldn't you like the work of thousands of people to stand for more than earnings per share?
I'm well aware that there are encodings from turing machines to bits. My point is that, given the definition of a Turing machine, there is no one single canonical encoding that is objectively correct. Similarly, there are an infinite number of correct ways to encode any type of graph. Just because your textbook defines one way does not mean that it is the only way.