If that were really true, then if you are smart, you should be able to make a killing at the stock market.
Not so. Most people who play the lottery are stupid, too. But no amount of smarts is going to help you make a "killing" playing the lottery.
The stock market is similar to the lottery, except that the expected payback is usually somewhat above 1.0, instead of the 0.5 or so payback that most state lotteries yield. (I assume everyone here realizes that the an almost fraudulently advertised $100M jackpot figure does not represent acutal present value.)
Sure, some people do much better than average in the stock market, just as some people get lucky and win the lottery. However, much of that is dumb luck, and most of the rest is due to having inside connections and privileged information you can't access as a member of the general public.
Members of the public don't have enough information to reliably pick stocks, just as they can't predict which pingpong balls will pop out of the lotto machine. Thus, your best bet as an indivdual is to treat the stock market the way the state treats its lottery: diversify enough so that the individual gambles are irrelevant and depend on the overal odds to bring you revenue.
Microsoft-sponsored benchmarks prove that multiple WinTel Web servers perform better than a Linux mainframe acting as a Web server consolidator.
Well, that's not too surprising. I've read one or two articles in the past that pretty much raked mainframes over the coals w.r.t. their performance in a modern software environment. The basic point was: mainframes have huge capacity and incredible uptimes, but were designed for a batch processing environment where heavyweight processes crunch big sets of data in the middle of the night.
Serving up interactive web pages is not their strength. Paying the huge costs for a mainframe to do the job of a cluster of interchangeable stateless web servers doesn't make much sense.
However, none of this has much to do with what OS is running on the Mainframe or on the PC server boxes. The apples-to-apples comparison really should be Windows on a PC vs. Linux on a PC.
That's the cool part. We don't really do much of anything anymore.
We've somehow convinced the rest of the world to send us all their stuff in exchange for IOUs that will never be repaid. We're living high off the hog at others' expense. Thanks for all the stuff!
Any time you're working with data types which are a multiple of 64 bits, being 64 bit will mean you will only need to use half as many instructions to process the same amount of data. This is a "big deal".
X86 floating point has been using 1 clock and 1 instruction to operate on 80-bit chunks since the Pentium I. My argument was that outside the context of pointer arithmetic, 64-bit integers are almost useless. (Maybe after 2038 you'll need them for holding 'time_t' values.) Once numbers get that big, you are usually going to be using floating-point anyway, which works just fine today.
In 20 years of programming, I don't recall that I've ever had a need for numbers above 4 billion, but less than 2e19, but which don't support fractions (again, this is excluding pointer arithmetic). Sometimes I would have liked to do boolean operations on as many bits as possible, but IIRC, the MMX already lets you do that on 128 bits. As far as other operations on data types bigger than 64 bits, the other main one is bulk moves. However, that tends to be more determined by cache architecture than anything else. I seem to recall that the x86 already has tricks efficiently move entire cache blocks of data around.
I tend to agree with you on the registers. I have argued in the past that the X86 register scarcity wasn't a big deal because of register renaming and reordering. However, I've seen so many assertions that the new compiler-visible X86-64 registers were helping performance that I ended up accepting that maybe it was helping. Perhaps this new chip is just faster due to a bunch of unexciting little enhancements.
However, I still maintain that the 64-bit integer math doesn't help the average user in and of itself.
Wrong, he is on sabbtical from Transmeta, he is still officially an employee.
But let's apply the corporate press release decoder ring:
"He/she has made valuable contributions to the project and will be missed" -> He/she screwed up. Good riddance.
"Is leaving to spend more time with his/her family" -> Has been ousted
"Is leaving to pursue personal interests" -> Has been ousted
"Gone on sabbatical" -> Has left the company for good and will never, ever return. Further press release confirming official resignation to follow within 9 months.
Unfortunately, there's a lot of people out there who think accounting is a science.
I learned this point first hand decades ago when I had a summer job working for the head accountant of a manufacturing firm. Most of the job involved hand typing numbers from mainframe printouts into spreadsheets on a PC. (The kind of task Perl would eventually be invented to handle.)
However, part of the job was to adjust the magic "fudge factors" in the spreadsheets until the results matched their expectations. Each kind of product had quite a few fuzzy parameters like "overhead", "scrap percentage", and other strange acronyms I didn't understand. Historically, they got certain profit margins on different types of products. The parameters all needed to be tweaked until the profit margins looked right.
They had me, some college kid, making up numbers that affected millions of dollars of accounts and tax calculations. However, I don't think it was even possible to determine a single correct value for these numbers, so my choices were as good as any. After that I realized that most things in the real world are too complicated to accurately project onto the one-dimensional space represented by money.
Also note that x86-64 corrects some of the weaknesses of the x86 architecture, so x86-64 apps are automatically faster. Counter-strike was 30% faster, clock-for-clock.
That's an important point that people need to keep in mind. The 64-bitness in itself provides just about zero benefit to 99.9% of users.
Many people fail to realize that 64-bit in this context only means 64-bit pointers. Apps have been using 64-bit data for years: 80-bit floating point has been implemented in the X86 since the 1980s, and apps have been able to process 128 bits of fixed-point data at a time since the MMX was introduced in the late 1990s.
Even in a 64-bit CPU, most ints will probably be compiled at 32 bits to save memory space. There are very few situations where you need a 64-bit integer value in real world programs. You will not see any speedup due to suddenly being able to do arithmetic on bigger numbers.
The only "big deal" about 64-bit processors is the 64-bit addressing logic. This eliminates the 4GB limit on a contiguous virtual memory space. While this may be valuable for people running huge databases and scientific simulations, the individual user today has no need for this. The only big piles of data the average user may have today are videos, and algorithms to process video are highly streamable. There's no need to map an entire video into memory at one time.
64-bit pointers bring the disadvantage of consuming more valuable cache real-estate and bus bandwidth for every operation. These resources aren't free, and the extra cache and bandwidth added to support 64-bit pointers in a 64-bit CPU could have been utilized in a 32-bit processor to improve its performance as well.
AMD has indeed apparently produced a CPU that's faster on the average users' tasks, but as was pointed out, that's largely due to adding more register space to reduce the x86's notorious register pressure problem. AMD could have acheived the same effect by adding more registers and keeping the CPU 32 bits. I'm not saying that such a move would make any sense; it wouldn't (in particular, it would be a marketing disaster). I'm just pointing out that that's the only feature the vast majority of users will actually benefit from in the next 5 years with this CPU.
NASA's business model makes this kind of a rip off. In this space-aged version of "give away the razor, sell the blades", you don't actually get to run your own mission with this freebie software. Instead, all you get to do is watch as NASA calls all the shots for the two demo probes that are being sent to Mars this month. It's like they're handing out game consoles with no input controllers to go with them.
What they don't tell you up front is that if you actually want to run your own mission, you're going to have to pony up at least $100 million to buy a probe of your own. NASA stands to make a pretty penny if they sucker many people to buy into this scheme.
Copyright was never intended to be combined with trade secrets (proprietary file formats) to create an unassailable market position. Copyrights and patents were originally instituted to get secrets out in the open by encouraging publishing and revealing secret manufacturing processes to the public.
Interoperability or interchangeability is important for almost any market. When combined with trade secrets, copyrights can create a huge barrier to entry into a software market, because it makes interoperability very difficult to achieve. A free market in software is not impossible today, but it is an unstable state with the current system. A slight push tends to move any segment of the software market towards a single dominant vendor. The vast majority of Microsoft's revenue is made in markets which are not in a healthy balance.
If copyright law were changed so that software vendors were required to publicly release the file formats and protocols used by a piece of software in exchange for the exclusive monopoly, then the software market would be much more open and competetive.
Yeah. It sounds like your mind is pretty cluttered if you can't keep up with the minutia required to administrate a wireless network.
I can keep up with the minutia. I don't want to. I've got better things to do with my time.
You must not be running linux huh?....
OS is irrelevant. Linux, Windows, OpenBSD... I've used wireless adapters on all of them and they all have configuration and security problems to worry about.
I made my wireless network secure this year. After a couple of years of use, my wireless adapters are now sitting in the bottom of a drawer, and I tacked a Cat5 ethernet cable to my ceiling and walls to replace them.
No more worries about wireless security alerts, finicky configurations, key management, weird drivers, setting up VPNs within my own house, strange network freezeups or having to read articles to keep on top of it all.
To me, keeping my mind uncluttered and free from all that minutia is worth the ugliness of a few network cables.
I hope you don't keep those unmounted disks physically attached to the system. People have been lucky, because all worm writers to date have been kind-hearted enough not to zero out all of the systems' hard drives or flash their BIOSes. You have no guarantee that the next worm won't be written by a real asshole.
Probably even worse: a worm that quietly opens a back door to your system without you even knowing it. You could go run for months with your system totally compromised, and your backups would copy the same problem to your offline partitions.
This is great news for people with my religious beliefs. We believe that in order to survive in the afterlife, our physical bodies must be carefully preserved and remain intact, and we must be buried with all of the goods and servants we will need in the next world. If any of this is ever disturbed, our time in the afterlife will come to an end. It is of the utmost importance that our tombs never be violated or destroyed.
Naturally, this has been a great problem for my anscestors, with looters and archaeogists plundering our graves. First, we tried similar a similar long-term storage system in a huge man made stone polyhedron, but this was too conspicuous and attracted robbers. Then, we tried vaults hidden in a valley, but the robbers scoured the entire area, destroying the afterlife hopes for countless of my ancestors.
Maybe this system, at the center of a real mountain, perpetually guarded by corporate rent-a-cops, will finally ensure the endless afterlife that we strive for. I'm going to have them send me a brochure.
..who claim we're coming to the limits of silicon, and XXXX MHz is the highest that can be achieved. Technology will keep on advancing relentlessly, changing and adapting.
While technology could keep advancing for quite some time, that doesn't mean that advances will be economically feasible.
Take aircraft development, for example. The maximum speed advanced on a roughly exponential scale from 1903 through the mid 60s, culminating with an X-15 flight at around mach 6. Even today, researchers are tinkering around with models of aircraft faster than that. However, 99.99% of all passengers and cargo still move at the speed they did in 1960: about 500 mph.
Why is this? Because fuel consumption and noise problems make it uneconomical to go faster than a 707. For air travel, every day reality has become decoupled from technological possiblity.
Likewise, CPU performance will almost certainly hit a wall where the power consumption makes it impractical for the average user to run more MIPS. Processor technology will continue to advance, but only for applications where power consumption is no object.
The problem is that when you can no longer target CPUs at the mass market, the potential revenue shrinks, so investment money dries up, slowing the development cycle. (This is a big part of the reason why 40 years after the X-15 and SR-71 we haven't come up with anything faster.) This will be the factor that ends exponential silicon CPU performance increases, even if there is no fundamental physical roadblock to producing faster processors.
Heck even cuba doesnt have nuclear weapons. They all manage to stand up to US "bullying".
They did have them for a few weeks in 1962. That harrowing situation made the issue of our relationship with Cuba such a hot potato that we didn't do anything provocative until the end of the Cold War.
Since then, with communist expansion no longer a threat, we've just been waiting for Fidel to die of old age.
i see this alot in text book examples of writing C code. i never understood it other then the fact that using parens () forces the evaluation of what is in them before continuing with the rest of the statement.
From the C language FAQ:
20.19: Are the outer parentheses in return statements really optional?
A: Yes.
Long ago, in the early days of C, they were required, and just
enough people learned C then, and wrote code which is still in
circulation, that the notion that they might still be required
is widespread.
(As it happens, parentheses are optional with the sizeof
operator, too, under certain circumstances.)
References: K&R1 Sec. A18.3 p. 218; ISO Sec. 6.3.3, Sec. 6.6.6;
H&S Sec. 8.9 p. 254.
The Federal Interstate System and the federal monies to lesser highways were started for the National Defense primarily, with economic reasons secondary.
That's what it was sold as, but really everybody just wanted a shiny fun new toy to drive on. People probably just didn't want to admit that the federal government was again expanding its charter and getting into the highway building business, so they said "The federal government is making something for its military forces, just like the constitution says it should."
If you wanted a network of roads for primarily for military use, it would be far more cost effective to build roads that are mostly single lane with no shoulders. With traffic under central military command, there's no need even for bidirectional travel on the roads. In fact, it would be even more cost effective to just use railroads. How big of an expressway do you really need to supply an ICBM base?
Hitler built the autobahn under the same military pretext, but they went almost unused through the war because it turned out that they mostly used the rail system for moving military traffic.
the site would be 3 miles from the sea where deuterium is plentiful
Yes, we'd really like to avoid having to ship the dozens of grams of deuterium that this machine will burn over its lifetime all the way from the ocean to an inland site; or even worse, pay the water bills involved with extracting this deuterium from tapwater.
BUT even the Discovery Science channel tends to show the same documentaries and episodes of shows.
As far as I can tell, all they did to create that channel was splice together two weeks worth of low-budget documentary tapes onto one giant spool, then they put it in a closet and set it spinning. That same tape loop has playing for years now. Even the ads never seem to change.
Re:The unintended benefits of pollution
on
Global Dimming
·
· Score: 4, Funny
These pollution-created artificial clouds probably reduce global warming (the article mentions this effect and a correlated decrease in cloudiness and increase in temperatures in the 1990s).
The Venusians apparently made the same assumption about their climate, unfortunately.
Re:tomorrows weather, 20 and sunny.
on
Global Dimming
·
· Score: 1
Once they manage to consistently predict tomorrows weather successfully they may go onwards and claim they have a clue how the weather will be 100 years from now.
Predicting tomorrow's weather and predicting the global climate have nothing to do with each other. If I dump some milk into a cup of coffee, there is no way that I can accurately predict the pattern of swirls that I'll see. However, I can accurately predict the final color of the blended coffee based on how much milk I added.
It's the same for short-term weather vs. global climate. Maybe climate prediction isn't all that accurate today, but making it accurate will not depend at all on knowing how to do accurate short-term weather prediction.
Slashdot Mirror
Not so. Most people who play the lottery are stupid, too. But no amount of smarts is going to help you make a "killing" playing the lottery.
The stock market is similar to the lottery, except that the expected payback is usually somewhat above 1.0, instead of the 0.5 or so payback that most state lotteries yield. (I assume everyone here realizes that the an almost fraudulently advertised $100M jackpot figure does not represent acutal present value.)
Sure, some people do much better than average in the stock market, just as some people get lucky and win the lottery. However, much of that is dumb luck, and most of the rest is due to having inside connections and privileged information you can't access as a member of the general public.
Members of the public don't have enough information to reliably pick stocks, just as they can't predict which pingpong balls will pop out of the lotto machine. Thus, your best bet as an indivdual is to treat the stock market the way the state treats its lottery: diversify enough so that the individual gambles are irrelevant and depend on the overal odds to bring you revenue.
Well, that's not too surprising. I've read one or two articles in the past that pretty much raked mainframes over the coals w.r.t. their performance in a modern software environment. The basic point was: mainframes have huge capacity and incredible uptimes, but were designed for a batch processing environment where heavyweight processes crunch big sets of data in the middle of the night.
Serving up interactive web pages is not their strength. Paying the huge costs for a mainframe to do the job of a cluster of interchangeable stateless web servers doesn't make much sense.
However, none of this has much to do with what OS is running on the Mainframe or on the PC server boxes. The apples-to-apples comparison really should be Windows on a PC vs. Linux on a PC.
That's the cool part. We don't really do much of anything anymore.
We've somehow convinced the rest of the world to send us all their stuff in exchange for IOUs that will never be repaid. We're living high off the hog at others' expense. Thanks for all the stuff!
X86 floating point has been using 1 clock and 1 instruction to operate on 80-bit chunks since the Pentium I. My argument was that outside the context of pointer arithmetic, 64-bit integers are almost useless. (Maybe after 2038 you'll need them for holding 'time_t' values.) Once numbers get that big, you are usually going to be using floating-point anyway, which works just fine today.
In 20 years of programming, I don't recall that I've ever had a need for numbers above 4 billion, but less than 2e19, but which don't support fractions (again, this is excluding pointer arithmetic). Sometimes I would have liked to do boolean operations on as many bits as possible, but IIRC, the MMX already lets you do that on 128 bits. As far as other operations on data types bigger than 64 bits, the other main one is bulk moves. However, that tends to be more determined by cache architecture than anything else. I seem to recall that the x86 already has tricks efficiently move entire cache blocks of data around.
I tend to agree with you on the registers. I have argued in the past that the X86 register scarcity wasn't a big deal because of register renaming and reordering. However, I've seen so many assertions that the new compiler-visible X86-64 registers were helping performance that I ended up accepting that maybe it was helping. Perhaps this new chip is just faster due to a bunch of unexciting little enhancements.
However, I still maintain that the 64-bit integer math doesn't help the average user in and of itself.
But let's apply the corporate press release decoder ring:
"He/she has made valuable contributions to the project and will be missed" -> He/she screwed up. Good riddance.
"Is leaving to spend more time with his/her family" -> Has been ousted
"Is leaving to pursue personal interests" -> Has been ousted
"Gone on sabbatical" -> Has left the company for good and will never, ever return. Further press release confirming official resignation to follow within 9 months.
I learned this point first hand decades ago when I had a summer job working for the head accountant of a manufacturing firm. Most of the job involved hand typing numbers from mainframe printouts into spreadsheets on a PC. (The kind of task Perl would eventually be invented to handle.)
However, part of the job was to adjust the magic "fudge factors" in the spreadsheets until the results matched their expectations. Each kind of product had quite a few fuzzy parameters like "overhead", "scrap percentage", and other strange acronyms I didn't understand. Historically, they got certain profit margins on different types of products. The parameters all needed to be tweaked until the profit margins looked right.
They had me, some college kid, making up numbers that affected millions of dollars of accounts and tax calculations. However, I don't think it was even possible to determine a single correct value for these numbers, so my choices were as good as any. After that I realized that most things in the real world are too complicated to accurately project onto the one-dimensional space represented by money.
That's an important point that people need to keep in mind. The 64-bitness in itself provides just about zero benefit to 99.9% of users.
Many people fail to realize that 64-bit in this context only means 64-bit pointers. Apps have been using 64-bit data for years: 80-bit floating point has been implemented in the X86 since the 1980s, and apps have been able to process 128 bits of fixed-point data at a time since the MMX was introduced in the late 1990s.
Even in a 64-bit CPU, most ints will probably be compiled at 32 bits to save memory space. There are very few situations where you need a 64-bit integer value in real world programs. You will not see any speedup due to suddenly being able to do arithmetic on bigger numbers.
The only "big deal" about 64-bit processors is the 64-bit addressing logic. This eliminates the 4GB limit on a contiguous virtual memory space. While this may be valuable for people running huge databases and scientific simulations, the individual user today has no need for this. The only big piles of data the average user may have today are videos, and algorithms to process video are highly streamable. There's no need to map an entire video into memory at one time.
64-bit pointers bring the disadvantage of consuming more valuable cache real-estate and bus bandwidth for every operation. These resources aren't free, and the extra cache and bandwidth added to support 64-bit pointers in a 64-bit CPU could have been utilized in a 32-bit processor to improve its performance as well.
AMD has indeed apparently produced a CPU that's faster on the average users' tasks, but as was pointed out, that's largely due to adding more register space to reduce the x86's notorious register pressure problem. AMD could have acheived the same effect by adding more registers and keeping the CPU 32 bits. I'm not saying that such a move would make any sense; it wouldn't (in particular, it would be a marketing disaster). I'm just pointing out that that's the only feature the vast majority of users will actually benefit from in the next 5 years with this CPU.
What they don't tell you up front is that if you actually want to run your own mission, you're going to have to pony up at least $100 million to buy a probe of your own. NASA stands to make a pretty penny if they sucker many people to buy into this scheme.
Oh, wait...
Interoperability or interchangeability is important for almost any market. When combined with trade secrets, copyrights can create a huge barrier to entry into a software market, because it makes interoperability very difficult to achieve. A free market in software is not impossible today, but it is an unstable state with the current system. A slight push tends to move any segment of the software market towards a single dominant vendor. The vast majority of Microsoft's revenue is made in markets which are not in a healthy balance.
If copyright law were changed so that software vendors were required to publicly release the file formats and protocols used by a piece of software in exchange for the exclusive monopoly, then the software market would be much more open and competetive.
I use:
-= Slashdot as Presentation Software =-
[ <<Prev ] [ Next >> ]
I can keep up with the minutia. I don't want to. I've got better things to do with my time.
You must not be running linux huh? ....
OS is irrelevant. Linux, Windows, OpenBSD... I've used wireless adapters on all of them and they all have configuration and security problems to worry about.
No more worries about wireless security alerts, finicky configurations, key management, weird drivers, setting up VPNs within my own house, strange network freezeups or having to read articles to keep on top of it all.
To me, keeping my mind uncluttered and free from all that minutia is worth the ugliness of a few network cables.
"There is no problem in computer science that cannot be solved by adding another layer of indirection."
I hope you don't keep those unmounted disks physically attached to the system. People have been lucky, because all worm writers to date have been kind-hearted enough not to zero out all of the systems' hard drives or flash their BIOSes. You have no guarantee that the next worm won't be written by a real asshole.
Probably even worse: a worm that quietly opens a back door to your system without you even knowing it. You could go run for months with your system totally compromised, and your backups would copy the same problem to your offline partitions.
Naturally, this has been a great problem for my anscestors, with looters and archaeogists plundering our graves. First, we tried similar a similar long-term storage system in a huge man made stone polyhedron, but this was too conspicuous and attracted robbers. Then, we tried vaults hidden in a valley, but the robbers scoured the entire area, destroying the afterlife hopes for countless of my ancestors.
Maybe this system, at the center of a real mountain, perpetually guarded by corporate rent-a-cops, will finally ensure the endless afterlife that we strive for. I'm going to have them send me a brochure.
While technology could keep advancing for quite some time, that doesn't mean that advances will be economically feasible.
Take aircraft development, for example. The maximum speed advanced on a roughly exponential scale from 1903 through the mid 60s, culminating with an X-15 flight at around mach 6. Even today, researchers are tinkering around with models of aircraft faster than that. However, 99.99% of all passengers and cargo still move at the speed they did in 1960: about 500 mph. Why is this? Because fuel consumption and noise problems make it uneconomical to go faster than a 707. For air travel, every day reality has become decoupled from technological possiblity.
Likewise, CPU performance will almost certainly hit a wall where the power consumption makes it impractical for the average user to run more MIPS. Processor technology will continue to advance, but only for applications where power consumption is no object.
The problem is that when you can no longer target CPUs at the mass market, the potential revenue shrinks, so investment money dries up, slowing the development cycle. (This is a big part of the reason why 40 years after the X-15 and SR-71 we haven't come up with anything faster.) This will be the factor that ends exponential silicon CPU performance increases, even if there is no fundamental physical roadblock to producing faster processors.
They did have them for a few weeks in 1962. That harrowing situation made the issue of our relationship with Cuba such a hot potato that we didn't do anything provocative until the end of the Cold War.
Since then, with communist expansion no longer a threat, we've just been waiting for Fidel to die of old age.
From the C language FAQ:
That's what it was sold as, but really everybody just wanted a shiny fun new toy to drive on. People probably just didn't want to admit that the federal government was again expanding its charter and getting into the highway building business, so they said "The federal government is making something for its military forces, just like the constitution says it should."
If you wanted a network of roads for primarily for military use, it would be far more cost effective to build roads that are mostly single lane with no shoulders. With traffic under central military command, there's no need even for bidirectional travel on the roads. In fact, it would be even more cost effective to just use railroads. How big of an expressway do you really need to supply an ICBM base?
Hitler built the autobahn under the same military pretext, but they went almost unused through the war because it turned out that they mostly used the rail system for moving military traffic.
Yes, we'd really like to avoid having to ship the dozens of grams of deuterium that this machine will burn over its lifetime all the way from the ocean to an inland site; or even worse, pay the water bills involved with extracting this deuterium from tapwater.
If he had done that, then the only clever one-liner you could write would be:
As far as I can tell, all they did to create that channel was splice together two weeks worth of low-budget documentary tapes onto one giant spool, then they put it in a closet and set it spinning. That same tape loop has playing for years now. Even the ads never seem to change.
The Venusians apparently made the same assumption about their climate, unfortunately.
Predicting tomorrow's weather and predicting the global climate have nothing to do with each other. If I dump some milk into a cup of coffee, there is no way that I can accurately predict the pattern of swirls that I'll see. However, I can accurately predict the final color of the blended coffee based on how much milk I added.
It's the same for short-term weather vs. global climate. Maybe climate prediction isn't all that accurate today, but making it accurate will not depend at all on knowing how to do accurate short-term weather prediction.