If you want to get the special deals, but not be at risk, just pick a different name on your application (and fake address, etc). (I remember somewhere someone suggested that we all sign up as George Orwell.)
Strike 4: Not promising: "The new technology is seamless to the end-user because the compressed data can be uncompressed in nanoseconds when needed." Call me a pessimist, but memory right now is around 6ns for PC133. Now, assuming a very conservative 2ns to decode the data, that's 8ns, which is a 25% performance hit. How many admins do you know that would take a 25% hit on performance on their servers to save a couple hundred bucks?
You are forgetting how expensive running out of memory is. Yes, a 25% performance hit is significant. But the memory equation is the following:
Average Access Time = Hit Time + Miss Rate*Miss Penalty
They are lowering the Miss Rate and increasing the Hit Time. This should effectively lower the Average Access Time IF they have lowered the miss rate significantly, which they have IF you are running (for example) a gigantic database and it doesn't fit entirely in RAM.
Remember, the Miss Penalty from RAM to Hard Disk is ENORMOUS, measured in milliseconds, not nanoseconds, so, suppose you can lower your miss rate from.01% to.005%, and the miss penalty is 1 ms.
Hence (note 1ms = 1 X 10^6 ns):
Old Average Access Time = 6ns +.01% X 1ms = 106 ns/access
New Average Access Time = 8ns +.005% X 1ms = 58 ns/access
(Yes, I know, these numbers are cooked, but that doesn't mean it doesn't apply to someone's system.)
...but it seems to me that RAM is already designed to do one simple thing (okay, two things: peek and poke) and to do it absolutely as fast as possible
Not quite. DRAM is also designed to be cheap, using only 2 transistors per bit of storage. There are other kinds of RAM (such as those used in your L1 cache) that are designed to be as fast as possible (i.e. single clock access time), but use MANY MORE TRANSISTORS(I believe >8 per bit of storage), so are much more expensive. DRAM is cheap, but because it uses so few transistors, it has to be "refreshed" periodically. This is one of the reasons why it is so slow (~10 clock cycle access time), and is why all modern processors have at least one cache (and frequently more than one) between the processor and the DRAM. Of course, paging to disk (which happens whenever you are using more memory than you have) is EXTREMELY slow (~1000 clock cycles/access). This is why IBM's technology sounds very interesting. It will make memory access slower, but probably not much slower than it already is. It will also reduce the amount of paging to disk, by effectively doubling your RAM.
Perhaps you don't know quite so much about Java as you think you do!
Java can be compiled to machine code just fine, thank you. One such Static Compiler is called TowerJ!
Of course, the reason why most java you use isn't compiled in this way is because that makes it no longer cross-platform (IMO - "A BAD THING").
(TowerJ is normally used on the server side, by the way, where cross-platform isn't needed)
Besides, with the latest (read Java 1.3, still in alpha) Just-in-time compiling JVMs, performance is really not an issue anymore.
Further, JITs can be faster than static compilation for programs that run longer (like WORD) because parts of the program that run more frequently (for me, cut and paste) can be optimized more (i.e. loops unrolled more, etc). Of course, they could do this with static compilation, but most static compilers don't because the programs GET BIGGER.
You owe it to yourself to try a JVM with a JIT!
I recommend the new IBM 1.3 JVM for linux. It's really something special, and it makes Java fly! Get it at: http://www.alphaworks.ibm.com/tech/linuxjdk
In my tests, IBM's new 1.3 JVM ran at 85% of native code on SciMark 2.0 (a scientific benchmark for Java, but also written in C).
Don't believe me? try the benchmark yourself: http://math.nist.gov/scimark2/
And please use a JVM with a JIT installed! You owe it to yourself!
The other thing the judge decided today is pretty annoying if you ask me: (Quote from CNN.com)
However, he also agreed to postpone a series of restrictions on Microsoft's business practices, which had been set to go into effect in September, pending the outcome of the appeal.
The latest development in the case takes some of the pressure off of Microsoft, and legal experts said it is unlikely that the high court would decide if it will hear the appeal until its next term, which begins next fall. But of more immediate concern to Microsoft were a series of restrictions on its business practices, which Jackson said were designed to prevent it from punishing hardware and software companies working on competing products and favoring companies that exclude Microsoft competitors. The order also required Microsoft to disclose technical information about its operating systems to independent hardware and software companies so that those companies can design products that are compatible with Windows...
Microsoft had said the restrictions would have a "devastating" effect on its business because it would have to shares its software source code with competitors, redesign all of its operating systems, and comply with price controls.
It had been generally expected that Jackson would deny Microsoft's request for a stay of those restrictions. But in his order Tuesday, Jackson agreed to postpone his entire final judgment until after Microsoft has exhausted its appeals.
I was really excited about the behavioral remedies. Now they don't really matter much at all.
The problem that I've always seen is that humans have stopped evolving. At least according to this scheme. Sure, we keep attaining more knowlege, but for the most part the biological (and this includes mental) aspect of humanity is not changing (at least in no way that is caused by the strictest sense of evolution). This is because we no longer abide by "survival of the fittest".
Yes, what you are saying is accurate in the West, but I'm not so sure it applies in places where (for example) overcrowding has resulted in lower birthrates and the like. For example, in China (where multiple-child families are disallowed), it is possible that the ability to sire male children is selected for, since so many female embryos are aborted.
Further, in places where drought and famine are serious issues, it is conceivable that the ability to survive with less water/food is being selected for - i.e. perhaps the rate of twins is dropping.
The main problem with any of these kind of discussions is that evolution of any kind develops over millenia, not years and decades.
My audiologist friend commented to me that fan noise is a serious issue for her because digital hearing aids require the use of a personal (READ: Windows) computer to fit correctly, and as soon as the hearing aid is in, the user hears the "previously non-existant" fan noise, and believes it is generated by the hearing aid.
The key to lower heat output is lower power consumption. The Crusoe accomplishes this in the following ways:
1. By performing scheduling and the like in software, the Crusoe has far fewer transistors (read ~1/4) than a PIII or Athlon. Hence, lower power consumption.
2. In addition, Crusoe has a power management "system" that actually lowers the clock speed whenever less speed is needed. All chips should do that, but they don't. And it doesn't slow down memory access at all, since the bottleneck there is not the processor clock speed. This technology is referred to as LongRun. Here's a quote from Transmeta:
"In a mobile setting, most conventional x86 CPUs regulate their power consumption by rapidly alternating between running the processor at full speed and (in effect) turning the processor off. Different performance levels can be obtained by varying the on/off ratio (the "duty cycle"). However, with this approach, the processor may be shut off just when a time-critical application needs it. This can result in glitches, such as dropped frames during movie playback, that are perceptible (and annoying) to a user.
In contrast, the TM5400 can adjust its power consumption without turning itself off - instead, it can adjust its clock frequency on the fly. It does so extremely quickly, and without requiring an operating system reboot or other slow and involved OS or BIOS operation. As a result, software can continuously monitor the demands on the processor and dynamically pick just the right clock speed (and hence power consumption) needed to run the application - no more and no less - so no power is wasted. Since the switching happens so quickly, it is not noticeable to the user.
Finally, the Code Morphing software can also adjust the Crusoe processor's voltage on the fly (since at a lower operating frequency, a lower voltage can be used). Because power varies linearly with clock speed and by the square of the voltage, adjusting both can produce cubic reductions in power consumption, whereas conventional CPUs can adjust power only linearly. For example, assume an application program only requires 90 percent of the processor's speed. On a conventional processor, throttling back the processor speed by 10 percent cuts power by 10 percent, whereas under the same conditions, LongRun power management can reduce power by almost 30 percent (0.903 = 0.73) - a noticeable advantage!"
For more information I highly recommend everybody to read "The Technology Behind Crusoe(TM) Processors".
Available at: http://www.transmeta.com/crusoe/download/pdf/cru soetechwp.pdf
Ayn Rand called this the "Economy of Pull". There is no political solution. Break-up Microsoft or don't. It won't make any difference when the next giant company wants to use it's monopoly of widgets to create a monopoly of other widgets. What we need is a revolution... -Michael
Slashdot Mirror
If you want to get the special deals, but not be at risk, just pick a different name on your application (and fake address, etc). (I remember somewhere someone suggested that we all sign up as George Orwell.)
You are forgetting how expensive running out of memory is. Yes, a 25% performance hit is significant. But the memory equation is the following:
They are lowering the Miss Rate and increasing the Hit Time. This should effectively lower the Average Access Time IF they have lowered the miss rate significantly, which they have IF you are running (for example) a gigantic database and it doesn't fit entirely in RAM.
Remember, the Miss Penalty from RAM to Hard Disk is ENORMOUS, measured in milliseconds, not nanoseconds, so, suppose you can lower your miss rate from .01% to .005%, and the miss penalty is 1 ms.
Hence (note 1ms = 1 X 10^6 ns):
Perhaps you don't know quite so much about Java as you think you do!
Java can be compiled to machine code just fine, thank you. One such Static Compiler is called TowerJ!
Of course, the reason why most java you use isn't compiled in this way is because that makes it no longer cross-platform (IMO - "A BAD THING").
(TowerJ is normally used on the server side, by the way, where cross-platform isn't needed)
Besides, with the latest (read Java 1.3, still in alpha) Just-in-time compiling JVMs, performance is really not an issue anymore.
Further, JITs can be faster than static compilation for programs that run longer (like WORD) because parts of the program that run more frequently (for me, cut and paste) can be optimized more (i.e. loops unrolled more, etc). Of course, they could do this with static compilation, but most static compilers don't because the programs GET BIGGER.
You owe it to yourself to try a JVM with a JIT!
I recommend the new IBM 1.3 JVM for linux. It's really something special, and it makes Java fly!
Get it at:
http://www.alphaworks.ibm.com/tech/linuxjdk
In my tests, IBM's new 1.3 JVM ran at 85% of native code on SciMark 2.0 (a scientific benchmark for Java, but also written in C).
Don't believe me? try the benchmark yourself:
http://math.nist.gov/scimark2/
And please use a JVM with a JIT installed! You owe it to yourself!
-Michael Arick
This is kind of nitpicky, but I don't like their console output code...
//A namespace in the .NET
//I edited it somewhat...
Here is C# Hello World:
using System;
class Hello
{
static void Main() {
Console.WriteLine("Hello,world");
}
}
So instead of System.out.println, we have System.Console.WriteLine. Lovely, huh? Just like Java.
Unfortunately, they take this whole "Embrace and Extend" thing one step too far. Notice the following program:
using System;
class Test
{
static void Main() {
int v1 = 0;
int v2 = v1;
Console.WriteLine("Values: {0}, {1}", v1, v2);
}
}
This looks ugly, doesn't it? You have to number your printed variables and put the numbers in braces. And no, I don't know if nested braces work...
For reference, here's how it's done in Java:
System.out.println("Values: " + v1 + ", " + v2);
Genetic Programming and Genetic Algorithms really work. Here's the general idea behind genetic algorithms (and a specific example - curve fitting)
1)Express the problem and solution space in terms of a set of numbers.
ex: coefficients on x^i where i steps from 0 to 100.
2)Express a fitness function - this can be very difficult!
ex: testing 1000 different points, fitness = sum of standard deviations
3)Generate a random set of hypothetical solutions to the problem - it's best to generate 100-1000.
4) Test the fitness of each possible solution.
ex. just as stated in 2, sum the standard deviations.
5) Keep all the solutions so far (within reason) and add:
5a)Some mutations of some of them.
ex. change some of the coefficients a bit.
5b)Some crossovers of some of them.
ex. take some coefficients from solution X, and THE OTHER COEFFICIENTS from solution Y.
Note: mutation and crossover policies have to be well designed so as to stop local minimum issues.
6)Go back to 4) until the fitness of a solution is within some threshold of the ideal fitness
(in my example, that might be 10.000000 or something).
Check out the following resource for source code if you want to try it out yourself:
http://www.aic.nrl.navy.mil/galist/src/
I was really excited about the behavioral remedies . Now they don't really matter much at all.
Yes, what you are saying is accurate in the West, but I'm not so sure it applies in places where (for example) overcrowding has resulted in lower birthrates and the like. For example, in China (where multiple-child families are disallowed), it is possible that the ability to sire male children is selected for, since so many female embryos are aborted.
Further, in places where drought and famine are serious issues, it is conceivable that the ability to survive with less water/food is being selected for - i.e. perhaps the rate of twins is dropping.
The main problem with any of these kind of discussions is that evolution of any kind develops over millenia, not years and decades.
My audiologist friend commented to me that fan noise is a serious issue for her because digital hearing aids require the use of a personal (READ: Windows) computer to fit correctly, and as soon as the hearing aid is in, the user hears the "previously non-existant" fan noise, and believes it is generated by the hearing aid.
Needless to say, this is a serious problem.
The key to lower heat output is lower power consumption. The Crusoe accomplishes this in the following ways:
u soetechwp.pdf
1. By performing scheduling and the like in software, the Crusoe has far fewer transistors (read ~1/4) than a PIII or Athlon. Hence, lower power consumption.
2. In addition, Crusoe has a power management "system" that actually lowers the clock speed whenever less speed is needed. All chips should do that, but they don't. And it doesn't slow down memory access at all, since the bottleneck there is not the processor clock speed. This technology is referred to as LongRun. Here's a quote from Transmeta:
"In a mobile setting, most conventional x86 CPUs regulate their power consumption by rapidly alternating between running the processor at full speed and (in effect) turning the processor off. Different performance levels can be obtained by varying the on/off ratio (the "duty cycle"). However, with this approach, the processor may be shut off just when a time-critical application needs it. This can result in glitches, such as dropped frames during movie playback, that are perceptible (and annoying) to a user.
In contrast, the TM5400 can adjust its power consumption without turning itself off - instead, it can adjust its clock frequency on the fly. It does so extremely quickly, and without requiring an operating system reboot or other slow and involved OS or BIOS operation. As a result, software can continuously monitor the demands on
the processor and dynamically pick just the right clock speed (and hence power consumption) needed to run the application - no more and no less - so no power is wasted. Since the switching happens so quickly, it is not noticeable to the user.
Finally, the Code Morphing software can also adjust the Crusoe processor's voltage on the fly (since at a lower operating frequency, a lower voltage can be used). Because power varies linearly with clock speed and by the square of the voltage, adjusting both can produce cubic reductions in power consumption, whereas conventional CPUs can adjust power only linearly. For example, assume an application program only requires 90 percent of the processor's speed. On a conventional processor, throttling back the processor speed by 10 percent cuts power by 10 percent, whereas under the same conditions, LongRun power management can reduce power by almost 30 percent (0.903 = 0.73) - a noticeable advantage!"
For more information I highly recommend everybody to read "The Technology Behind Crusoe(TM) Processors".
Available at:
http://www.transmeta.com/crusoe/download/pdf/cr
Ayn Rand called this the "Economy of Pull". There is no political solution. Break-up Microsoft or don't. It won't make any difference when the next giant company wants to use it's monopoly of widgets to create a monopoly of other widgets. What we need is a revolution... -Michael