AMD offers a very compelling server CPU, but it is unclear if they can supply Dell as reliably as Intel. AMD has concentrated risk in one or two fabs while Intel has several fabs.
Also, it is more complex/expensive/disruptive to maintain 2 different motherboard/chipset combinations. 64 bits is not a good reason since Intel has x86-64 support now. AMD performance is better but not enought to offset the operational risk. The AMD solution doesn't enable a platform that Intel could not also provide (albeit at lower performance).
I think the executives at SGI, HP etc... saw the writing on the wall.
High end CPU development is expensive and becomes even more expensive with each new process generation. Having everyone (MIPS, SPARC, PA-RISC, Alpha) develop their own architectures with little difference in performance between them was not cost effective. Capital investment in high end CPU development did not return enough dollars to recoup the costs.
x86 on the other hand, had plenty of PC revenue and profits to justify incredible investments in R&D. Intel was in a clearly superior financial position to take on ever more risk, complexity and R&D costs with respect to CPU development.
The executives probably crunched the numbers and realized that they could no longer afford to be in the game and "outsource" their CPU development to Intel.
Much in the same way that they "outsource" their DRAM development to the DRAM semiconductor companies.
It just doesn't make economic sense to fund a CPU design which takes a few billion dollars these days and several years of development --- with such low volumes as found in the server and workstation markets. Especially when that market is even further decimated by 3 or 4 competing RISC architectures.
Another nail in the RISC coffin is that the ISA choice - RISC vs. CISC, matters less and less with each process generation and with new architectural innovations like pre-decoded instruction caches, trace caches and store forwarding buffers. The fixed overhead of x86 gets dimished as transitor budgets grow with each generation.
>> Hey, if our brains do a nice enough job using lots of parallel instruction "dumb" processors, why are we so obssessed with ultra fast only-a-few-instructions-at-a-time single processors?
Because it's extremely difficult to do parallel programming.
The entire world has been trained to think about programming in a linear single thread way. This will change because hardware people can no longer make single threads faster w/o sacrificing tremendous amounts of power. Programming just became that much harder...
>> Because it's the right thing to do! Maybe some people live on farms and grow crops to feed our fat asses - does that mean they shouldn't be have good roads and good schools? Because that's what would happen if it was privatized: there's no profit in giving these people proper utilities because it's not densly populated enough.
The right thing to do is let the market (supply/demand) decide where resources go. The market is the aggregate wishes of the entire population as expressed through how things are priced.
Clearly we need food to survive. Don't you think that people would put up capital to support farming? If roads are required to supply food to a public that demands it, you bet your ass they'd get built because there is an economic incentive to do so. Otherwise we'd starve right? If you want to take the "feed our asses" to the extreme - Why isn't there a government program to provide 3 meals a day to every citizen?
>> Who said I think it was free? Don't put words into my mouth, thanks. I actually said "I'm sure it wasn't cheap." Not sure how that implies that I think it was free?
Guilty - you didn't say free. I'm sure it's not cheap, but private schools would be less expensive than the public schools that we currently have. This is due to competitive forces that would make the schools more efficient otherwise they would risk insolvency.
>> And you think things would be better in a completely privatized corporation based system of public services and government?
I advocate a minimalist state - which protects us militarily and protects our inalienable rights. Everything else should be left up to the public (ie - free market). The reasons for this are too lengthy to go into in a slashdot comment, but I would recommend reading Nozick's "Anarchy, State and Utopia" for a clear and logical analysis of this political philosophy.
>> I don't just pay into public services I don't use because I'm feeling nice. One day, I might use those services. Or my kids might, or my parents, or my friends.
And you're free to allocate your money to those ends as you see fit. Just don't foricbly take money from others to do so.
>> And shit man, maybe I think the poor family down the road should be able to feed their family, and maybe I think the farmer out in the boondox should have a paved road and clean water to drink. Maybe I think we should all do a little bit to help out our fellow citizens and our taxes to a lot to help. If a few bucks a month out of my pocket is going to be able to do all that, then I say it's worth it.
Perhaps that family should move? If they aren't contributing to society (as measured by their economic output - ie - they aren't supplying what the public demands), then maybe they should do something that does satisfy the public. Perhaps, instead of giving money to the government, that money could be more efficiently invested in the free market to create jobs in more productive areas - areas that the public demands (through their buying habits).
>> I will say this again though, just because I don't think you're getting this: I don't think all our tax money is being used wisely. I think that a lot of things need to change. I think our current system works - we need to get rid of the abuse and corruption in our government to make things right. Privatizing things won't cure this either (see: Enron, Worldcom, etc..)
Tax money very clearly isn't being used wisely. That's because the incentives are setup such that it is guaranteed that the resource allocation will be inefficient (monopolies suppress the price signal, lack of competition allows waste, command and control resource allocation cannot efficiently allocate capital due to a lack of information and coordination). How do you get rid of corruption and abuse when there is little countermeasure to prevent it? The politician's only interest is in getting re-elected - the costs be damned.
>> The real drivers seem to be low levels of corruption and proper law enforcement. It isn't particularly related to how often people go to the polls.
Good points.
I would also add, vigorous protection of property rights and a good banking system, which allows capital to flow more smoothly.
Freeing capital to flow to where demand is and dictated by supply in the most efficient manner possible is a good predictor of economic success.
Give the people what they want and let it happen through the coordination of the price signal (supply/demand) instead of through centralized government planning.
>> This isn't true, really. The same would happen with what we have in Internet access. Only the densely populated areas (where you can get more tolls) would get roads, and everyone else would get dirt.
There is a cost and a benefit to everything. Why should we build roads where there is little benefit to offset the costs? You're implying that people should subsidize rural development with no discernable benefit to themselves? Why?
>> If it were a "pay as you go, only if you use it" program, it would be far too expensive to put your kids through 12 years of school. If it were privitized, we'd still have to have tax money pay for it..
Why would it be more expensive? How much money goes into the tax system today to fund schools? If that money were not sent to the government and sent to private schools instead - at the discretion of the citizen (not some government bureaucrat), then becuase of competitive forces, the costs would be lower. The only reason that you think schooling is "free" is because you don't directly see the costs - that big chunk of change that comes out of your paycheck or the thousands of dollars of sales tax that you pay when you buy a car or the thousands that you pay for the "privelage" to live on your own property through property taxes. Take all of that out, and re-invest in private schools and see if you come out ahead or not...
>> No, this is the essence of greed and anarchy. Let the rich get richer and the poor stay where they are, right? Because that's what you do if you don't pay anything into public services unless you use them yourself. It's selfish.
Perhaps you don't quite understand capitalism. Greed and selfishness are the hallmarks of capitalism. Do you go to work for free? No, you trade your services for money. Do you redistribute all of your disposable income to poor people? NO? Well you're a greedy bastard. AKA - capitalist.
We live in a world of finite resources that we all compete over - many are unconscious of this. The question is - how to we allocate those resources? Should a small group of elite power-brokers in washington decide? Or should it be the public that allocates resources through supply and demand (ie - prices)?
Look, everyone should be free to do what they want with their money. If you want to spend more than your daily usage of public utility, then you are free to do so. Just don't steal money from others who are not willing to do so.
>> Yes, but if you were awake in economics class, you'll remember that there are some services which are better allocated by governments. For example, how would we decide who has to pay to defend the country were it invaded? Does everyone have the station mercenaries on their front yard?
Of course there is a role for government. You can't have anarchy. In fact, my favorite book on policatal philosophy is "anarchy, state and utopia" by Nozick which gives a very elegant and logical justification for a minimalist state that provides defense and protection of certain inalienable rights.
>> Roads are the same way. Our interstates are mostly not toll roads, and as a result, we can can ship products from point a to point b cheaper
How do you know that it is cheaper? The government has a monopoly on interstates and therefore the critical "price" signal is suppressed. Therefore, necessarily, capital is allocated to built roads with incorrect information. If priviate corps. built interstates and charged tolls, the price you pay would likely be LESS than the taxes you pay currently to fund roads. Which gets back to my original point of alternative uses - The money saved could be invested in the economy in more productive ways.
>> No one advocates PhDs for everyone because as a society, we only need so many PhDs around. However, for a modern society to function, its citizenry must have some basic education. We all benefit from the increase in productivity. We all benefit when the people around us are able to make educated decisions.
If education is important (I'm not disagreeing that it isn't)...then, the free market will allocate capital towards it. People will pay for what they need. Everyone needs a home for society to function, does that mean the government should give us all houses? Everyone needs food for a society to function, does that mean that the government should be handing out free meals to everyone?
There is a line that can be drawn and if you think through it, you'll see that MANY government services are not necessary. The money saved by eliminating these government programs could be put to more efficient uses. Again, I'd recommend looking though the book I mentioned earlier.
The 8MB of cache on the R12000 is OFF-DIE. It is connected to the CPU using a bus that is in fact SLOWER than the bus to DRAM on modern PCs (MIPS L2 cache BW is ~2GB/sec). So great, you have an 8MB cache but it's 2x slower than the 512MB of DDR DRAM that you can find in a PC! I could say that my PC has 512MB of cache...
Secondly, the 2MB of cache on the P4 is ON-DIE. Let's do some math...that's one 128-bit load per clock at 3.73GHz which is a shitload of bandwidth (59.68 Gbytes/sec). This is of course, no accident - you better design a memory system capable of feeding a 3.73GHz execution core otherwise it will starve.
>> A lot of people are of the incorrect opinion that "If I don't use it, why should I pay for it?" It's not as simple as that.
In fact, it is that simple. Look into capitalism sometime. Capital is never efficiently allocated by governments.
Capital has ALTERNATIVE uses... Think of the progress we'd have made today if all the wasted capital that is stolen from citizens by the government to fund wasteful programs were actually put towards productive uses... We'd have flying cars fueled by salt water by now so we wouldn't need roads:)
But seriously, roads could have been built by private corporations and paid for by people that use them via tolls. Since roads are integral for business enterprise, there would be a very strong incentive for capital flow there in order to build them.
Schools are the same way. A school would likely be more efficiently run by a private enterprise due to competitive forces. Therefore, it would be cheaper than government-run schools. If citizens value education, they will allocate capital for it.
If the government is so concerned with the education of its citizens, why is the public school system widely and correctly criticised as inadequate? Why does the government stop at some arbitrary point such as high school for "free" education? Why not mandate that everyone is entitled to a "free" Phd? Clearly, people would be better off with higher eduction - let's legislate it! Give me a break!
>> But you can't tell this to some people, they apparently don't have the capability to think past their own $5 in their pocket.
The $5 that they have is theirs to do with as they wish. That's the essence of freedom. When the government forcibly takes a chuck of that from you because "they know better" - it's the opposite of freedom.
>> 3. The majority shareholders are mostly a small group of already wealthy people.
False. Most majority shareholders are mutual fund and pension funds primarily funded by middle class people.
>> 2. Corporations do best for the shareholders.
Therefore, corporations do what's good for the mutual funds and pension funds which benefits many average investors (ie - the majority).
>> 5. Fire redundant pawns. Feed jobs overseas.
If they were redundant - why should they have jobs? That's an inefficient use of capital. If someone overseas can do the job for lower labor costs, that SHOULD be done because wage capital is allocated more efficiently. Inefficient use of capital destroys wealth.
>> 8. Most people and the local economy lose.
False. Most people are better off by redirecting inefficient resources towards more productive ends. Costs are lowered and therefore people can buy more than they could before with the same amount of money -- more wealth is created.
Insightful indeed... Why does slashdot hate capitalism?
Non-commercial Use Only Google Desktop Search is made available to you for your non-commercial use only. If you want to make commercial use of Google Desktop Search, including but not limited to selling or distributing Google Desktop Search for payment, you must enter into an agreement with Google or obtain Google's written permission in advance.
Therefore, using it in a commercial sense, ie - at you place of work will probably require giving google some money
So you're saying that laptops are as loud or louder than desktops for x86 but not macs? Why do you suppose that is?
You give 2 reasons
>> 1) The smaller package means that its harder to dissapate head into the surrounding air. >> 2) The processor is physically closer to you, because its on your lap rather than across the desk or on your floor.
How are these parameters different for PCs vs. Macs?
Here is the reason. The x86 notebook chips are far higher performance. Pentium-M runs circles around a G4. When Apple puts a G5 in a notebook, come back and tell me how loud your notebook fan is. This gets back to my original point of trading performance for power. Computer makers will manufacture and sell whatever generates the most profit. Considering that x86 has about 95% of the total notebook market it seems that the market wants performance over noise at the moment. The nanosecond that this demand changes, you'll see the quieter machines.
You never know what the future will bring. Perhaps this additional computing power will be put to good use someday.
$1000 worth of computers today is way more powerful than $1000 worth of computers 5 years ago -- significantly outpacing inflation. Performance is given to you for basically free!
If people really wanted a $120 computer to browse the web and write email, WEBTV would have been much more successful of a product. If there is a demand for such a product, you bet your ass somebody is working on it - there's just too much money at stake.
How much performance are you willing to give up in order to reduce cooling requirements?
I would suggest getting a notebook. They're reasonably quiet.
Small form factor solutions are in their infancy. They are generally more expensive than standard desktop solutions, but as demand picks up, economies of scale will take over and reduce costs as well as decibels of noise.
>> Huh, where did i even mention the speed of of electrons along wires? Im simply stating that wires will never be able to deliver enough data for the processor to be able to function at a correct regime.
I'm saying that there is not really a compelling need to go optical. Electrical does just fine with intelligent caching.
>> We might get some return to efficient coding being the norm, instead of writing systems anyhow and throwing more/faster hardware at it until it runs acceptably (Microsoft; its you I'm looking at!)
Efficient coding is only useful if there is a return on your investment for efficiency. Exponentially increasing hardware capability over time at the same cost point makes this tradeoff obvious. The article is saying the hardware capability will still increase, but the programmer will have to learn to use concurrency to exploit it. This implies a fundamental shift in the single-thread world that we have lived in for a long time now.
>> Your (and your business') desktop machine might _not_ become obsolete in no more than 2 years, and mmight continue in useful service as something more sensible than a whole PC doing the job of a router...
The only reason a computer becomes obsolete is that something better comes along to replace it. Why else would someone spend money to replace something that's just as good. There must be value there. So, you're saying you want to stop or slow down progress??? Nobody is holding a gun to your head to upgrade. Keep your old computers and the rest of the world will move on and pay for increased value.
>> Processor designers might spend more time (i know they already spend some) on innovating new ideas, rather than solving the problems with just ramping up clock speeds.
Processor designers spend ALL their time innovating on new ideas. How do you think each and every new chip comes out faster than the last one? If we don't make a better chip, investors won't give us money to build new ones. It's all we do all damn day - build a better mousetrap. Innovation is part of the job. And "just ramping up clock speed" is pretty damn difficult thank you very much.
>> I find that software designers often do not take resource limits seriously. Programming is tedious, hard work.
The reason they don't take resource limits seriously is because they're getting resources for damn near free. Hardware capability increases exponentially at the same cost point over time. Why waste your VALUABLE engineering time optimizing a problems that the hardware solves for you?
This article is claiming that the gravy train of free resources is over and programmer will have to think again.
1) 75% idle time is nonsense. Where did you get that number? With SPECfp on an Athlon or P4 it's more like 20-30% idle. Just look at how spec scores scale with frequency to figure out the memory-idle time.
2) Increasing switching speed with optical technology increases bandwidth but does nothing for latency since nothing travels faster than the speed of light and electrons flowing along a wire can acheive close to 80% of the speed of light already. To reduce latency, what we need are smarter architectures and programmers that can prefetch the data into lower latency caches ahead of time.
Why not start your own company to execute your brilliant ideas?
As of very recently, Google's market capitalization is about 52 Billion dollars.
52 BILLION!!! This, for a company that brings in a meagre 0.22 Billion plus change in profits.
Let me ask you something. If I gave you 52 billion dollars, could you replicate Google's business model? Of course you could, many times over. In fact the technology part of the business could be replicated with "only" 2-3 billion at most and I'm being very generous. The hard part is replicating the VERY good brand. However, 52 billion buys a ton of advertising to grow your brand - much more than you need. For example - Intel spent just 0.3 Billion on advertising to brand "Centrino". Now, even my grandma knows what Centrino is. You give me 50+ billion and you bet your ass my search engine would become a household name.
Therefore, the stock is overvalued. Working for Google at this point in time is retarded. The value has been monetized already by the brilliant pioneers 4-5+ years ago. In other words, if you want to be rich by working for Google, it's too late.
Well...why not build your own empire? Don't have the resources? That's what VC firms are for... Maybe your idea is so good that Google will have to pay big bucks to acquire it.
This post is predicated on the assumption that you care about money:)
Re:Solution - no conditionals or templates!
on
Programming Puzzles
·
· Score: 1
Nice solution!
Re:Solution - no conditionals or templates!
on
Programming Puzzles
·
· Score: 1
(condition == 0) is a conditional. (num >= max) is also a conditional.
Anything that compiles into a branch is a conditional.
Slashdot Mirror
>> Should read 'embrace AMD's architecture'
Yes, embrace AMD's architecture, which is in itself a trivial extension of INTEL's original x86 which AMD has copied for 2 decades.
AMD offers a very compelling server CPU, but it is unclear if they can supply Dell as reliably as Intel. AMD has concentrated risk in one or two fabs while Intel has several fabs.
Also, it is more complex/expensive/disruptive to maintain 2 different motherboard/chipset combinations. 64 bits is not a good reason since Intel has x86-64 support now. AMD performance is better but not enought to offset the operational risk. The AMD solution doesn't enable a platform that Intel could not also provide (albeit at lower performance).
I think the executives at SGI, HP etc... saw the writing on the wall.
High end CPU development is expensive and becomes even more expensive with each new process generation. Having everyone (MIPS, SPARC, PA-RISC, Alpha) develop their own architectures with little difference in performance between them was not cost effective. Capital investment in high end CPU development did not return enough dollars to recoup the costs.
x86 on the other hand, had plenty of PC revenue and profits to justify incredible investments in R&D. Intel was in a clearly superior financial position to take on ever more risk, complexity and R&D costs with respect to CPU development.
The executives probably crunched the numbers and realized that they could no longer afford to be in the game and "outsource" their CPU development to Intel.
Much in the same way that they "outsource" their DRAM development to the DRAM semiconductor companies.
It just doesn't make economic sense to fund a CPU design which takes a few billion dollars these days and several years of development --- with such low volumes as found in the server and workstation markets. Especially when that market is even further decimated by 3 or 4 competing RISC architectures.
Another nail in the RISC coffin is that the ISA choice - RISC vs. CISC, matters less and less with each process generation and with new architectural innovations like pre-decoded instruction caches, trace caches and store forwarding buffers. The fixed overhead of x86 gets dimished as transitor budgets grow with each generation.
>> Hey, if our brains do a nice enough job using lots of parallel instruction "dumb" processors, why are we so obssessed with ultra fast only-a-few-instructions-at-a-time single processors?
Because it's extremely difficult to do parallel programming.
The entire world has been trained to think about programming in a linear single thread way. This will change because hardware people can no longer make single threads faster w/o sacrificing tremendous amounts of power. Programming just became that much harder...
>> Because it's the right thing to do! Maybe some people live on farms and grow crops to feed our fat asses - does that mean they shouldn't be have good roads and good schools? Because that's what would happen if it was privatized: there's no profit in giving these people proper utilities because it's not densly populated enough.
The right thing to do is let the market (supply/demand) decide where resources go. The market is the aggregate wishes of the entire population as expressed through how things are priced.
Clearly we need food to survive. Don't you think that people would put up capital to support farming? If roads are required to supply food to a public that demands it, you bet your ass they'd get built because there is an economic incentive to do so. Otherwise we'd starve right? If you want to take the "feed our asses" to the extreme - Why isn't there a government program to provide 3 meals a day to every citizen?
>> Who said I think it was free? Don't put words into my mouth, thanks. I actually said "I'm sure it wasn't cheap." Not sure how that implies that I think it was free?
Guilty - you didn't say free. I'm sure it's not cheap, but private schools would be less expensive than the public schools that we currently have. This is due to competitive forces that would make the schools more efficient otherwise they would risk insolvency.
>> And you think things would be better in a completely privatized corporation based system of public services and government?
I advocate a minimalist state - which protects us militarily and protects our inalienable rights. Everything else should be left up to the public (ie - free market). The reasons for this are too lengthy to go into in a slashdot comment, but I would recommend reading Nozick's "Anarchy, State and Utopia" for a clear and logical analysis of this political philosophy.
>> I don't just pay into public services I don't use because I'm feeling nice. One day, I might use those services. Or my kids might, or my parents, or my friends.
And you're free to allocate your money to those ends as you see fit. Just don't foricbly take money from others to do so.
>> And shit man, maybe I think the poor family down the road should be able to feed their family, and maybe I think the farmer out in the boondox should have a paved road and clean water to drink. Maybe I think we should all do a little bit to help out our fellow citizens and our taxes to a lot to help. If a few bucks a month out of my pocket is going to be able to do all that, then I say it's worth it.
Perhaps that family should move? If they aren't contributing to society (as measured by their economic output - ie - they aren't supplying what the public demands), then maybe they should do something that does satisfy the public. Perhaps, instead of giving money to the government, that money could be more efficiently invested in the free market to create jobs in more productive areas - areas that the public demands (through their buying habits).
>> I will say this again though, just because I don't think you're getting this: I don't think all our tax money is being used wisely. I think that a lot of things need to change. I think our current system works - we need to get rid of the abuse and corruption in our government to make things right. Privatizing things won't cure this either (see: Enron, Worldcom, etc..)
Tax money very clearly isn't being used wisely. That's because the incentives are setup such that it is guaranteed that the resource allocation will be inefficient (monopolies suppress the price signal, lack of competition allows waste, command and control resource allocation cannot efficiently allocate capital due to a lack of information and coordination). How do you get rid of corruption and abuse when there is little countermeasure to prevent it? The politician's only interest is in getting re-elected - the costs be damned.
Privatizing will help - corr
>> The real drivers seem to be low levels of corruption and proper law enforcement. It isn't particularly related to how often people go to the polls.
Good points.
I would also add, vigorous protection of property rights and a good banking system, which allows capital to flow more smoothly.
Freeing capital to flow to where demand is and dictated by supply in the most efficient manner possible is a good predictor of economic success.
Give the people what they want and let it happen through the coordination of the price signal (supply/demand) instead of through centralized government planning.
>> This isn't true, really. The same would happen with what we have in Internet access. Only the densely populated areas (where you can get more tolls) would get roads, and everyone else would get dirt.
There is a cost and a benefit to everything. Why should we build roads where there is little benefit to offset the costs? You're implying that people should subsidize rural development with no discernable benefit to themselves? Why?
>> If it were a "pay as you go, only if you use it" program, it would be far too expensive to put your kids through 12 years of school. If it were privitized, we'd still have to have tax money pay for it..
Why would it be more expensive? How much money goes into the tax system today to fund schools? If that money were not sent to the government and sent to private schools instead - at the discretion of the citizen (not some government bureaucrat), then becuase of competitive forces, the costs would be lower. The only reason that you think schooling is "free" is because you don't directly see the costs - that big chunk of change that comes out of your paycheck or the thousands of dollars of sales tax that you pay when you buy a car or the thousands that you pay for the "privelage" to live on your own property through property taxes. Take all of that out, and re-invest in private schools and see if you come out ahead or not...
>> No, this is the essence of greed and anarchy. Let the rich get richer and the poor stay where they are, right? Because that's what you do if you don't pay anything into public services unless you use them yourself. It's selfish.
Perhaps you don't quite understand capitalism. Greed and selfishness are the hallmarks of capitalism. Do you go to work for free? No, you trade your services for money. Do you redistribute all of your disposable income to poor people? NO? Well you're a greedy bastard. AKA - capitalist.
We live in a world of finite resources that we all compete over - many are unconscious of this. The question is - how to we allocate those resources? Should a small group of elite power-brokers in washington decide? Or should it be the public that allocates resources through supply and demand (ie - prices)?
Look, everyone should be free to do what they want with their money. If you want to spend more than your daily usage of public utility, then you are free to do so. Just don't steal money from others who are not willing to do so.
>> Yes, but if you were awake in economics class, you'll remember that there are some services which are better allocated by governments. For example, how would we decide who has to pay to defend the country were it invaded? Does everyone have the station mercenaries on their front yard?
Of course there is a role for government. You can't have anarchy. In fact, my favorite book on policatal philosophy is "anarchy, state and utopia" by Nozick which gives a very elegant and logical justification for a minimalist state that provides defense and protection of certain inalienable rights.
>> Roads are the same way. Our interstates are mostly not toll roads, and as a result, we can can ship products from point a to point b cheaper
How do you know that it is cheaper? The government has a monopoly on interstates and therefore the critical "price" signal is suppressed. Therefore, necessarily, capital is allocated to built roads with incorrect information. If priviate corps. built interstates and charged tolls, the price you pay would likely be LESS than the taxes you pay currently to fund roads. Which gets back to my original point of alternative uses - The money saved could be invested in the economy in more productive ways.
>> No one advocates PhDs for everyone because as a society, we only need so many PhDs around. However, for a modern society to function, its citizenry must have some basic education. We all benefit from the increase in productivity. We all benefit when the people around us are able to make educated decisions.
If education is important (I'm not disagreeing that it isn't)...then, the free market will allocate capital towards it. People will pay for what they need. Everyone needs a home for society to function, does that mean the government should give us all houses? Everyone needs food for a society to function, does that mean that the government should be handing out free meals to everyone?
There is a line that can be drawn and if you think through it, you'll see that MANY government services are not necessary. The money saved by eliminating these government programs could be put to more efficient uses. Again, I'd recommend looking though the book I mentioned earlier.
Where do I begin?
The 8MB of cache on the R12000 is OFF-DIE. It is connected to the CPU using a bus that is in fact SLOWER than the bus to DRAM on modern PCs (MIPS L2 cache BW is ~2GB/sec). So great, you have an 8MB cache but it's 2x slower than the 512MB of DDR DRAM that you can find in a PC! I could say that my PC has 512MB of cache...
Secondly, the 2MB of cache on the P4 is ON-DIE. Let's do some math...that's one 128-bit load per clock at 3.73GHz which is a shitload of bandwidth (59.68 Gbytes/sec). This is of course, no accident - you better design a memory system capable of feeding a 3.73GHz execution core otherwise it will starve.
>> A lot of people are of the incorrect opinion that "If I don't use it, why should I pay for it?" It's not as simple as that.
:)
In fact, it is that simple. Look into capitalism sometime. Capital is never efficiently allocated by governments.
Capital has ALTERNATIVE uses... Think of the progress we'd have made today if all the wasted capital that is stolen from citizens by the government to fund wasteful programs were actually put towards productive uses... We'd have flying cars fueled by salt water by now so we wouldn't need roads
But seriously, roads could have been built by private corporations and paid for by people that use them via tolls. Since roads are integral for business enterprise, there would be a very strong incentive for capital flow there in order to build them.
Schools are the same way. A school would likely be more efficiently run by a private enterprise due to competitive forces. Therefore, it would be cheaper than government-run schools. If citizens value education, they will allocate capital for it.
If the government is so concerned with the education of its citizens, why is the public school system widely and correctly criticised as inadequate? Why does the government stop at some arbitrary point such as high school for "free" education? Why not mandate that everyone is entitled to a "free" Phd? Clearly, people would be better off with higher eduction - let's legislate it! Give me a break!
>> But you can't tell this to some people, they apparently don't have the capability to think past their own $5 in their pocket.
The $5 that they have is theirs to do with as they wish. That's the essence of freedom. When the government forcibly takes a chuck of that from you because "they know better" - it's the opposite of freedom.
>> 3. The majority shareholders are mostly a small group of already wealthy people.
False. Most majority shareholders are mutual fund and pension funds primarily funded by middle class people.
>> 2. Corporations do best for the shareholders.
Therefore, corporations do what's good for the mutual funds and pension funds which benefits many average investors (ie - the majority).
>> 5. Fire redundant pawns. Feed jobs overseas.
If they were redundant - why should they have jobs? That's an inefficient use of capital. If someone overseas can do the job for lower labor costs, that SHOULD be done because wage capital is allocated more efficiently. Inefficient use of capital destroys wealth.
>> 8. Most people and the local economy lose.
False. Most people are better off by redirecting inefficient resources towards more productive ends. Costs are lowered and therefore people can buy more than they could before with the same amount of money -- more wealth is created.
Insightful indeed...
Why does slashdot hate capitalism?
From the Terms and conditions:
Non-commercial Use Only
Google Desktop Search is made available to you for your non-commercial use only. If you want to make commercial use of Google Desktop Search, including but not limited to selling or distributing Google Desktop Search for payment, you must enter into an agreement with Google or obtain Google's written permission in advance.
Therefore, using it in a commercial sense, ie - at you place of work will probably require giving google some money
So you're saying that laptops are as loud or louder than desktops for x86 but not macs? Why do you suppose that is?
You give 2 reasons
>> 1) The smaller package means that its harder to dissapate head into the surrounding air.
>> 2) The processor is physically closer to you, because its on your lap rather than across the desk or on your floor.
How are these parameters different for PCs vs. Macs?
Here is the reason. The x86 notebook chips are far higher performance. Pentium-M runs circles around a G4. When Apple puts a G5 in a notebook, come back and tell me how loud your notebook fan is. This gets back to my original point of trading performance for power. Computer makers will manufacture and sell whatever generates the most profit. Considering that x86 has about 95% of the total notebook market it seems that the market wants performance over noise at the moment. The nanosecond that this demand changes, you'll see the quieter machines.
Yes, and 640K should be enough for everybody.
You never know what the future will bring. Perhaps this additional computing power will be put to good use someday.
$1000 worth of computers today is way more powerful than $1000 worth of computers 5 years ago -- significantly outpacing inflation. Performance is given to you for basically free!
If people really wanted a $120 computer to browse the web and write email, WEBTV would have been much more successful of a product. If there is a demand for such a product, you bet your ass somebody is working on it - there's just too much money at stake.
Power consumption = Performance
How much performance are you willing to give up in order to reduce cooling requirements?
I would suggest getting a notebook. They're reasonably quiet.
Small form factor solutions are in their infancy. They are generally more expensive than standard desktop solutions, but as demand picks up, economies of scale will take over and reduce costs as well as decibels of noise.
>> Huh, where did i even mention the speed of of electrons along wires? Im simply stating that wires will never be able to deliver enough data for the processor to be able to function at a correct regime.
I'm saying that there is not really a compelling need to go optical. Electrical does just fine with intelligent caching.
Oh my...where to begin.
>> We might get some return to efficient coding being the norm, instead of writing systems anyhow and throwing more/faster hardware at it until it runs acceptably (Microsoft; its you I'm looking at!)
Efficient coding is only useful if there is a return on your investment for efficiency. Exponentially increasing hardware capability over time at the same cost point makes this tradeoff obvious. The article is saying the hardware capability will still increase, but the programmer will have to learn to use concurrency to exploit it. This implies a fundamental shift in the single-thread world that we have lived in for a long time now.
>> Your (and your business') desktop machine might _not_ become obsolete in no more than 2 years, and mmight continue in useful service as something more sensible than a whole PC doing the job of a router...
The only reason a computer becomes obsolete is that something better comes along to replace it. Why else would someone spend money to replace something that's just as good. There must be value there. So, you're saying you want to stop or slow down progress??? Nobody is holding a gun to your head to upgrade. Keep your old computers and the rest of the world will move on and pay for increased value.
>> Processor designers might spend more time (i know they already spend some) on innovating new ideas, rather than solving the problems with just ramping up clock speeds.
Processor designers spend ALL their time innovating on new ideas. How do you think each and every new chip comes out faster than the last one? If we don't make a better chip, investors won't give us money to build new ones. It's all we do all damn day - build a better mousetrap. Innovation is part of the job. And "just ramping up clock speed" is pretty damn difficult thank you very much.
>> I find that software designers often do not take resource limits seriously. Programming is tedious, hard work.
The reason they don't take resource limits seriously is because they're getting resources for damn near free. Hardware capability increases exponentially at the same cost point over time. Why waste your VALUABLE engineering time optimizing a problems that the hardware solves for you?
This article is claiming that the gravy train of free resources is over and programmer will have to think again.
A few problems with your post.
1) 75% idle time is nonsense. Where did you get that number? With SPECfp on an Athlon or P4 it's more like 20-30% idle. Just look at how spec scores scale with frequency to figure out the memory-idle time.
2) Increasing switching speed with optical technology increases bandwidth but does nothing for latency since nothing travels faster than the speed of light and electrons flowing along a wire can acheive close to 80% of the speed of light already. To reduce latency, what we need are smarter architectures and programmers that can prefetch the data into lower latency caches ahead of time.
How about more reliable computer systems where NOTHING is really deleted. Documents that you are editing get saved and time-stamped every second.
Employ various forms of redundancy to improve reliability.
Why not start your own company to execute your brilliant ideas?
:)
As of very recently, Google's market capitalization is about 52 Billion dollars.
52 BILLION!!! This, for a company that brings in a meagre 0.22 Billion plus change in profits.
Let me ask you something. If I gave you 52 billion dollars, could you replicate Google's business model? Of course you could, many times over. In fact the technology part of the business could be replicated with "only" 2-3 billion at most and I'm being very generous. The hard part is replicating the VERY good brand. However, 52 billion buys a ton of advertising to grow your brand - much more than you need. For example - Intel spent just 0.3 Billion on advertising to brand "Centrino". Now, even my grandma knows what Centrino is. You give me 50+ billion and you bet your ass my search engine would become a household name.
Therefore, the stock is overvalued. Working for Google at this point in time is retarded. The value has been monetized already by the brilliant pioneers 4-5+ years ago. In other words, if you want to be rich by working for Google, it's too late.
Well...why not build your own empire? Don't have the resources? That's what VC firms are for... Maybe your idea is so good that Google will have to pay big bucks to acquire it.
This post is predicated on the assumption that you care about money
Nice solution!
(condition == 0) is a conditional.
(num >= max) is also a conditional.
Anything that compiles into a branch is a conditional.
Let's say that everyone lives for an average of 1000 years. Given current interest rates, everyone would be rich by the time they are 100 years old.
How can everyone be rich? That's a classic inflation scenario --- too much money chasing too few goods. Interest rates would have to adjust.
The ? : operator is a conditional.