To my mind, making "under God" a required part of the Pledge would be prohibiting free exercise of religion, but so would omitting it. So, make it optional. Case closed, can everyone be happy now?
This is something I find somewhat interesting in that congress specifically made a law to add "under God" to the pledge of allegiance. Is there any other act of congress that anyone can mention that specifically refers to God? Doesn't this fit the literal reading of the first amendment that "Congress shall make no law respecting an establishment of religion". I think the ruling screwed up slightly in that the real ruling should have been that the act to add the words to the pledge was unconstitutional, and thereby revert the pledge to pre-195x wording.
I say that because, once you know wha your doing, you start doing things piece meal instead of all at once. If I ever decided to replace everything in my system all at once, I would get some white box guy to build it for me to my specs.
As surprising as it might seem, letting that local shop order components for you and assemble it may actually be cheaper than buying the components yourself. The days of 50% markup over cost on PCs are long gone, now it's more like 2-5%. The shops can buy in volume and get better deals than you can. Come up with a spec on your PC and ask some of the local shops for a quote before buying the components yourself.
I agree. I woudl go a step further than specs and choose specific model parts for the PC. Plus, it is much harder to ignore some one face to face than over the phone.
My opinion is the best deal is to decide exactly what you want by doing your research on the web. Choose, motherboard, CPU, memory, video, sound, hard disk, case specs, etc... Then, find a local white box assembler that will build it with a warranty. You have a couple advantages:
1) It will probably cost less because the white box guy might add say $100 to his cost for profit. But, buying individual pieces might have much more than $100 in profit included. (Just example, don't know actual profit.)
2) If something doesn't work, you can walk in with the box, and harass the vendor personally, it is much harder to ignore an irate customer face to face than over the phone.
3) Warranty.
Do it yourself is really only advantageous if you can upgrade piece meal, and cannibalize your own system. For example: my next upgrade will be whatever Athlon and MB I can get for ~$150, 512MB DDR memory (appropriate for MB), and a GeForce4 card (4200 probably), everything else will carry over from my current system. Total cost will be a little over $400. Maybe less depending on memory prices and Geforce4200 prices.
I think pretty much any court would uphold that a NWN module is a derivative work and therefore up to Bioware to license. So, you can't charge for it based on the EULA. Big whoop! If you really think your modules are good enough to charge for and make a profit, go to Bioware and get a license to charge for it. Bioware will ask for a piece of the action, so they can make some money, and your off. I wouldn't be surprised if there are a few D20 publishers out there watching sales of NWN and distribution of user mods to see if it is worth the effort to make and sell modules. At which point they go to Bioware and negotiate a deal.
At 100% efficiency, a solar plant would have a collector surface area of 14.82e6 square meters to equal Plant Votgle's power output (that's a square a little less than four kilometers on a side).
Not bad, but the nice thing about solar is it can allow distributed energy generation. Using more realistic numbers at 30% efficiency. And, a middle of the road average of 4000Whr/sq m per day in the US. And, an average home usage of 10,000whr/day. It would take 10,000/(.3*4000)= 8.3 sqm of solar panels to power a single home. Call it a square 10ft on a side.
See here to provide power for the entire US. A square 100 miles by 129 miles. Sounds like a lot, but there are some very desolate deserts in the southwest that could easily fit that. Of course you woudln't want it all in one place, so ideally, you cover every rooftop, and then put whatever extra is needed in the most god forsaken desolate piece of desert in the southwest you can find.
Add some wind, which will tend to smooth out the valleys of solar. Some nuclear, just in case. And, Water to hydrogen electrolysis and fuel cells to store energy for nights and peaks.
I'll take the cheaper option, foreign films and quality shows anyday...
Do you have any idea how much film costs?!?
I looked up the cost of 35mm film at 10 minutes roll costs $200 and that doesn't include developing. that is about $.33 per second. I will call it $.20 per second to account for bulk discount.
Digital on the other hand has higher up front costs for equipment, but lower marginal costs. The Sony HDW-F900 (used in Attack of the Clones) has a resolution of 1920x1080 = 2M x 24bit = 6M per frame uncompressed. The cassettes are about $60/50 minutes. Or $.02 per second. Which puts film cost at anywhere from 10 times that of digital for media.
There is also the advantage to the diretcor beign able to view each take right away and decide whether additional angles are needed or takes or whether the scene is good and they can move on.
The company also complained that the Pentagon is funding research on making free software more secure, which in effect subsidizes Microsoft's open-source competitors, Stenbit said.
Is it just me or is this quote complete and total crap. They could just as easily have said.
The open source community complained that the Pentagon is paying Microsoft for their software, which in effect subsidizes open-source's proprietary competitors, Stenbit said.
I hope that the Pentagon can see through how self serving and hypocritical that complaint is.
The arguments against severability are that Congress wouldn't have passed just the future extension, all the lobbying was for the extension of existing works, and the Court shouldn't be in the business of rewriting flawed legislation.
That is not entirely accurate. The Supreme Court can remove words from a law, but cannot add words. The problem is the way the law is written the retroactive portion cannot be removed by just striking out a few words without eliminating also the extensionto new copyrights. Therefore, the whole thing would have to be struck down. If Congress had put the retroactive and future provisions into separate clauses then the Supreme Court could have struck one down without affecting the other. This is regardless of Congresses actual intent.
I wasn't present, but a similar problem occurred at my company at a trade show with a brand new product. The system was shipped uncrated and was just tied down. The dock at the trade show was not even with the truck bed, and multiple casters were sheared off. they got the system to the booth eventually, but after that we always crated systems for that show, so the fork lift coudl move it to the booth.
A couple posters have already linked to the source paper which is actually not terribly difficult to read. But, here is the gist of the analysis.
Essentially, they didn't detect two earthquakes. They went through the data looking for groups of at least seven detecting stations that detected seismic events that were not attributed to earthquakes using a point source model. They then attempted to fit the seismic data from all seven stations to a linear model where each station would not have necessarily detected the entry or exit point, but actually would have received the first signal from the point of closest approach(POCA). Some of the stations even had directional capability, so instead of having just a 7 data point fit to a linear event, they had a 9 point or more fit. So, the same results would actually require at multiple earthquakes which would have been corroborated as earthquakes by other stations.
On top of all that the seismograph signals really don't look much like earthquakes. Firstly, magnitude 4 - 4.5 eathquakes just don't last 30 seconds. Secondly, You don't get 15 seconds of earthquake plus 15 seconds of lesser earthquake consistently over 30 seconds from a 4 - 4.5 earthquake. But, you can get the second from a strangelet whose POCA to the station was 15 seconds from exit, and upon exit the station just receiving echos from prior to POCA.
So in order to make this hypothesis fly, you'd need evidence of some unexplained explosions, or else you'd need some statistical analysis that it's unlikely two given quakes are so closely timed.
How about the fact that the strangelet events look nothing like earthquakes on the seismograph?
I quote from the Daily Telegraph article:
According to Prof Herrin, the two events agree with predictions for strangelet impacts, which are expected to occur about once a year. He added, however, that finding more would be difficult, as seismic databases now automatically remove all signals not linked to earthquakes. He said: "To find more events we need to get at the data before that happens."
I think the suggestion is that the strangelets contain enough up and down quarks and (presumably) electrons to make the aggregate close to zero charge.
Would charge really matter that much? Considering the strong force (which would hold these together) is 3(more?) orders of magnitude stronger than the electromagnetic force. Protons are +e and don't fly apart.
There's a certain pull to it. Everyone has felt like an outsider at some point (to greater or lesser degrees), everyone has felt powerful in some area that no one else it - and everyone has felt the tug of conflicting interests.
I think there is more to Spider-man's succes than the above. You what you described of the outsider and "geek hero" going on, but that won't translate to 114 million and multiple viewings in a single weekend.
I think the real strength of the movie is there is something in it for everyone, and somehow it pulls this off without tripping all over itself. There is the romance between Peter and MJ, just enough to keep the women watching and wondering what will happne. There is the action. The dealing with the consequences of his power. His realationships with the people around him, and how his double life affects those relationships. Spiderman pulled these elements together without seeming preachy or blatantly sucking up to the people who like those different kinds of movies.
So your solar electricity plant has a large ecological impact (dozens or hundreds of acres required) and can't produce electricity as reliably as a stinky old coal plant. Plus you have to clean the mirrors/solar cells.
40000 square miles to produce all electricity needed in the US at 15% efficiency. At ~3.5 million square miles for the entire country that is 1% of the land surface. Sounds like a lot, but how many square miles of roof exist in the US.
And, you are right it doesn't produce electricity as reliably. But, that is why you don't put all the solar cells in one place, and use the grid to transport electricity from high solar flux area to low solar flux areas. You also throw in some wind, hydro, nuclear, and a little natural gas, just in case. And, have enough excess capacity to hydrolize hydrogen during the day for night production, cloudy days, cars, and transport to places impractical for grid transport (Alaska?).
Won't happen today, probably not in the next 10 year. But, when mass produced photovoltaics drop in price and start hitting efficiencies of around 30%, I think you will see more roofs covered with them.
We would probably have to find the numbers on coal because the internal combustion engine is very infefficient, so they might come in pretty close. The number we need is energy per unit of pollution.
Even if they were located at the equator, solar power would still cost a lot. Same with wind. Solar and wind power are not serious contenders to solve our energy woes. They cost too much.
I would qualify this by saying that solar is not a SHORT term contender. But, long term could handle much of the power needs.
As I recall, photovoltaic efficiency right now is somewhere around 15% (at best) of total solar energy hitting the cell. Pushing that towards 30-40% at near current cost would eliminate the cost issue. Or, inversely reducing cost of the cell at current efficiencies would get the same effect.
Another interesting item is the time factor. Currently, home photovoltaics panels are warrantied for 20 years. At that lifetime the cost is around 15-20 cents/kwh at retail, but do we really have the data to say that 20 years is the limit. Due to the lack of moving parts 40 years may be realistic. With that lifetime you again halve the cost per kwh.
Basically, all this says is that short term, solar doesn't cut it for mass production of power. Long term through reducing manufacturing costs, increasing efficiency, and improving durability, solar could supply long term energy needs.
A much bigger problem is where the hell do we get enough energy to feed 10 billion people?
Amd has 2 High Volume CPU fabs. And, FAB25 in Ausitn is being converted to Flash, so it is going to 1 this year. Onteh other hand withthe transition to 130nm processes FAB30 in Dresden can make 50 million processors a year. Compare that to current sales of 8million a quarter or 32 million a year. AMD has plenty of capacity in Dresden without spending a ton of money getting 12" equipment. Then, there is the UMC foundry deal which is probably worth up to 20 million more processors.
So, let's say up to 70 million CPUs/year starting next year, IF they can sell them. At 8 million this quarter they are claiming ~20% market share. Which implies total market of 40million/quarter or 160 million/year. So, let's assume modest growth fothe CPU market to 180 million/year. AMD either has to grow their market share to 27% to sell out Fab30, or 38% to sell out all available capacity. Assuming more modest market share growth to 25% AMD then needs a CPU market of 200 million units to sell out Fab30 and 280million units to sell out total capacity.
What is the point? The point is from a business perspective AMD is being smart. Increasing CPU speed through die shrinks to increase market share is good, and also results in increased capacity. What AMD is finding is at this poitn they can increase capacity sufficiently to support sales through die shrinks and the foundry agreement with UMC. In anticipation of needing more CPUs in 2005 they are building a 12" Fab in Singapore.
How could Intel manufacture faster memory than the companies that are SPECIALIZED in producing state-of-the-art memory chips?
Because the interconnect is at chip level. Once you go off chip latencies go up. Intel doesn't have to make a faster SRAM, they just need the on-chip interconnect which means they have to make the SRAM.
On top of that, everyone knows how to make SRAM anyways, there are really no new big secrets there. It is so simple they use it to test new fab processes.
The lowest end CPU you can find now-a-days is like 800Mhz, unless you go to auctions...
Yes, and that 800Mhz CPU with motherboard runs around $150 at Fry's in SoCal. Actually, for $179 I believe you can get an Athlon XP 1600+ with motherboard.
Let's compare that to the lower end hardware. The Wednesday Fry's ad has a K6-2 500 with motherboard for $79.
The low end seems to be the inverse of the high end. At the high end a small increase in speed results in a large increase in price. At the low end a large decrease in speed results in a small decrease in price. Try to find a P1-120 with motherboard for less than $79, and tell me would you rather pay a little extra for the K6-2 500 which is an VIA MVP4 chipset so you have onboard audio and video.
Regardless of the issues of taxation generally. My problem is why can't anyone get the math right. You get your withholding setup to pay the federal government jus the right amount over the course of the year, and whammo, you haven't paid the state (At least in California) enough by a significant amount. I am sure everyone uses the same tables, so why is it that the calculation for federal taxes comes out right, and the state taxes comes out wrong? It just seems incredibly stupid.
It's a "as long as it's the other guy that dies" philosophy that truly is disturbing.
I don't think this point can be emphasized enough. This attitude permeates the people who buy large vehicles because they are "safe". They are only safe to the few people in the vehicle. They are a hazard to every other person on the rode. But, of course that doesn't matter to the self serving bastards driving the biggest vehicles they can find, and putting their teenage children into the same sorts of vehicles so they can kill people.
And, of course none of them will admit that it is a kill or be killed mentality that makes them buy sucha vehicle.
Slashdot Mirror
To my mind, making "under God" a required part of the Pledge would be prohibiting free exercise of religion, but so would omitting it. So, make it optional. Case closed, can everyone be happy now?
This is something I find somewhat interesting in that congress specifically made a law to add "under God" to the pledge of allegiance. Is there any other act of congress that anyone can mention that specifically refers to God? Doesn't this fit the literal reading of the first amendment that "Congress shall make no law respecting an establishment of religion". I think the ruling screwed up slightly in that the real ruling should have been that the act to add the words to the pledge was unconstitutional, and thereby revert the pledge to pre-195x wording.
Just because it is harder to ignore some one face to face doesn't mean some a**h***s can't. :-)
I say that because, once you know wha your doing, you start doing things piece meal instead of all at once. If I ever decided to replace everything in my system all at once, I would get some white box guy to build it for me to my specs.
Don
As surprising as it might seem, letting that local shop order components for you and assemble it may actually be cheaper than buying the components yourself. The days of 50% markup over cost on PCs are long gone, now it's more like 2-5%. The shops can buy in volume and get better deals than you can. Come up with a spec on your PC and ask some of the local shops for a quote before buying the components yourself.
I agree. I woudl go a step further than specs and choose specific model parts for the PC. Plus, it is much harder to ignore some one face to face than over the phone.
My opinion is the best deal is to decide exactly what you want by doing your research on the web. Choose, motherboard, CPU, memory, video, sound, hard disk, case specs, etc... Then, find a local white box assembler that will build it with a warranty. You have a couple advantages:
1) It will probably cost less because the white box guy might add say $100 to his cost for profit. But, buying individual pieces might have much more than $100 in profit included. (Just example, don't know actual profit.)
2) If something doesn't work, you can walk in with the box, and harass the vendor personally, it is much harder to ignore an irate customer face to face than over the phone.
3) Warranty.
Do it yourself is really only advantageous if you can upgrade piece meal, and cannibalize your own system. For example: my next upgrade will be whatever Athlon and MB I can get for ~$150, 512MB DDR memory (appropriate for MB), and a GeForce4 card (4200 probably), everything else will carry over from my current system. Total cost will be a little over $400. Maybe less depending on memory prices and Geforce4200 prices.
Dastardly
I think pretty much any court would uphold that a NWN module is a derivative work and therefore up to Bioware to license. So, you can't charge for it based on the EULA. Big whoop! If you really think your modules are good enough to charge for and make a profit, go to Bioware and get a license to charge for it. Bioware will ask for a piece of the action, so they can make some money, and your off. I wouldn't be surprised if there are a few D20 publishers out there watching sales of NWN and distribution of user mods to see if it is worth the effort to make and sell modules. At which point they go to Bioware and negotiate a deal.
Dastardly
At 100% efficiency, a solar plant would have a collector surface area of 14.82e6 square meters to equal Plant Votgle's power output (that's a square a little less than four kilometers on a side).
Not bad, but the nice thing about solar is it can allow distributed energy generation. Using more realistic numbers at 30% efficiency. And, a middle of the road average of 4000Whr/sq m per day in the US. And, an average home usage of 10,000whr/day. It would take 10,000/(.3*4000)= 8.3 sqm of solar panels to power a single home. Call it a square 10ft on a side.
See here to provide power for the entire US. A square 100 miles by 129 miles. Sounds like a lot, but there are some very desolate deserts in the southwest that could easily fit that. Of course you woudln't want it all in one place, so ideally, you cover every rooftop, and then put whatever extra is needed in the most god forsaken desolate piece of desert in the southwest you can find.
Add some wind, which will tend to smooth out the valleys of solar. Some nuclear, just in case. And, Water to hydrogen electrolysis and fuel cells to store energy for nights and peaks.
Dastardly
I'll take the cheaper option, foreign films and quality shows anyday...
Do you have any idea how much film costs?!?
I looked up the cost of 35mm film at 10 minutes roll costs $200 and that doesn't include developing. that is about $.33 per second. I will call it $.20 per second to account for bulk discount.
Digital on the other hand has higher up front costs for equipment, but lower marginal costs. The Sony HDW-F900 (used in Attack of the Clones) has a resolution of 1920x1080 = 2M x 24bit = 6M per frame uncompressed. The cassettes are about $60/50 minutes. Or $.02 per second. Which puts film cost at anywhere from 10 times that of digital for media.
There is also the advantage to the diretcor beign able to view each take right away and decide whether additional angles are needed or takes or whether the scene is good and they can move on.
Dastardly
The company also complained that the Pentagon is funding research on making free software more secure, which in effect subsidizes Microsoft's open-source competitors, Stenbit said.
Is it just me or is this quote complete and total crap. They could just as easily have said.
The open source community complained that the Pentagon is paying Microsoft for their software, which in effect subsidizes open-source's proprietary competitors, Stenbit said.
I hope that the Pentagon can see through how self serving and hypocritical that complaint is.
Dastardly
The arguments against severability are that Congress wouldn't have passed just the future extension, all the lobbying was for the extension of existing works, and the Court shouldn't be in the business of rewriting flawed legislation.
That is not entirely accurate. The Supreme Court can remove words from a law, but cannot add words. The problem is the way the law is written the retroactive portion cannot be removed by just striking out a few words without eliminating also the extensionto new copyrights. Therefore, the whole thing would have to be struck down. If Congress had put the retroactive and future provisions into separate clauses then the Supreme Court could have struck one down without affecting the other. This is regardless of Congresses actual intent.
Dastardly
:P
Bite Me!
Go Sharks!
I wasn't present, but a similar problem occurred at my company at a trade show with a brand new product. The system was shipped uncrated and was just tied down. The dock at the trade show was not even with the truck bed, and multiple casters were sheared off. they got the system to the booth eventually, but after that we always crated systems for that show, so the fork lift coudl move it to the booth.
A couple posters have already linked to the source paper which is actually not terribly difficult to read. But, here is the gist of the analysis.
Essentially, they didn't detect two earthquakes. They went through the data looking for groups of at least seven detecting stations that detected seismic events that were not attributed to earthquakes using a point source model. They then attempted to fit the seismic data from all seven stations to a linear model where each station would not have necessarily detected the entry or exit point, but actually would have received the first signal from the point of closest approach(POCA). Some of the stations even had directional capability, so instead of having just a 7 data point fit to a linear event, they had a 9 point or more fit. So, the same results would actually require at multiple earthquakes which would have been corroborated as earthquakes by other stations.
On top of all that the seismograph signals really don't look much like earthquakes. Firstly, magnitude 4 - 4.5 eathquakes just don't last 30 seconds. Secondly, You don't get 15 seconds of earthquake plus 15 seconds of lesser earthquake consistently over 30 seconds from a 4 - 4.5 earthquake. But, you can get the second from a strangelet whose POCA to the station was 15 seconds from exit, and upon exit the station just receiving echos from prior to POCA.
Dastardly
So in order to make this hypothesis fly, you'd need evidence of some unexplained explosions, or else you'd need some statistical analysis that it's unlikely two given quakes are so closely timed.
How about the fact that the strangelet events look nothing like earthquakes on the seismograph?
I quote from the Daily Telegraph article:
According to Prof Herrin, the two events agree with predictions for strangelet impacts, which are expected to occur about once a year. He added, however, that finding more would be difficult, as seismic databases now automatically remove all signals not linked to earthquakes. He said: "To find more events we need to get at the data before that happens."
I think the suggestion is that the strangelets contain enough up and down quarks and (presumably) electrons to make the aggregate close to zero charge.
Would charge really matter that much? Considering the strong force (which would hold these together) is 3(more?) orders of magnitude stronger than the electromagnetic force. Protons are +e and don't fly apart.
Dastardly
There's a certain pull to it. Everyone has felt like an outsider at some point (to greater or lesser degrees), everyone has felt powerful in some area that no one else it - and everyone has felt the tug of conflicting interests.
I think there is more to Spider-man's succes than the above. You what you described of the outsider and "geek hero" going on, but that won't translate to 114 million and multiple viewings in a single weekend.
I think the real strength of the movie is there is something in it for everyone, and somehow it pulls this off without tripping all over itself. There is the romance between Peter and MJ, just enough to keep the women watching and wondering what will happne. There is the action. The dealing with the consequences of his power. His realationships with the people around him, and how his double life affects those relationships. Spiderman pulled these elements together without seeming preachy or blatantly sucking up to the people who like those different kinds of movies.
Dastardly
So your solar electricity plant has a large ecological impact (dozens or hundreds of acres required) and can't produce electricity as reliably as a stinky old coal plant. Plus you have to clean the mirrors/solar cells.
40000 square miles to produce all electricity needed in the US at 15% efficiency. At ~3.5 million square miles for the entire country that is 1% of the land surface. Sounds like a lot, but how many square miles of roof exist in the US.
And, you are right it doesn't produce electricity as reliably. But, that is why you don't put all the solar cells in one place, and use the grid to transport electricity from high solar flux area to low solar flux areas. You also throw in some wind, hydro, nuclear, and a little natural gas, just in case. And, have enough excess capacity to hydrolize hydrogen during the day for night production, cloudy days, cars, and transport to places impractical for grid transport (Alaska?).
Won't happen today, probably not in the next 10 year. But, when mass produced photovoltaics drop in price and start hitting efficiencies of around 30%, I think you will see more roofs covered with them.
We would probably have to find the numbers on coal because the internal combustion engine is very infefficient, so they might come in pretty close. The number we need is energy per unit of pollution.
a) prevent the next release causing more glabal warming and b) to get cheap clean fuel..why not mine it..?
You get Natural Gas from this. i.e methane.
So,
a) You cause global warming by recovering the nartural gas and burning it. Just slower.
b) Still get CO2, so not completly clean.
Dastardly
Even if they were located at the equator, solar power would still cost a lot. Same with wind. Solar and wind power are not serious contenders to solve our energy woes. They cost too much.
I would qualify this by saying that solar is not a SHORT term contender. But, long term could handle much of the power needs.
As I recall, photovoltaic efficiency right now is somewhere around 15% (at best) of total solar energy hitting the cell. Pushing that towards 30-40% at near current cost would eliminate the cost issue. Or, inversely reducing cost of the cell at current efficiencies would get the same effect.
Another interesting item is the time factor. Currently, home photovoltaics panels are warrantied for 20 years. At that lifetime the cost is around 15-20 cents/kwh at retail, but do we really have the data to say that 20 years is the limit. Due to the lack of moving parts 40 years may be realistic. With that lifetime you again halve the cost per kwh.
Basically, all this says is that short term, solar doesn't cut it for mass production of power. Long term through reducing manufacturing costs, increasing efficiency, and improving durability, solar could supply long term energy needs.
A much bigger problem is where the hell do we get enough energy to feed 10 billion people?
Amd has 2 High Volume CPU fabs. And, FAB25 in Ausitn is being converted to Flash, so it is going to 1 this year. Onteh other hand withthe transition to 130nm processes FAB30 in Dresden can make 50 million processors a year. Compare that to current sales of 8million a quarter or 32 million a year. AMD has plenty of capacity in Dresden without spending a ton of money getting 12" equipment. Then, there is the UMC foundry deal which is probably worth up to 20 million more processors.
So, let's say up to 70 million CPUs/year starting next year, IF they can sell them. At 8 million this quarter they are claiming ~20% market share. Which implies total market of 40million/quarter or 160 million/year. So, let's assume modest growth fothe CPU market to 180 million/year. AMD either has to grow their market share to 27% to sell out Fab30, or 38% to sell out all available capacity.
Assuming more modest market share growth to 25% AMD then needs a CPU market of 200 million units to sell out Fab30 and 280million units to sell out total capacity.
What is the point? The point is from a business perspective AMD is being smart. Increasing CPU speed through die shrinks to increase market share is good, and also results in increased capacity. What AMD is finding is at this poitn they can increase capacity sufficiently to support sales through die shrinks and the foundry agreement with UMC. In anticipation of needing more CPUs in 2005 they are building a 12" Fab in Singapore.
How could Intel manufacture faster memory than the companies that are SPECIALIZED in producing state-of-the-art memory chips?
Because the interconnect is at chip level. Once you go off chip latencies go up. Intel doesn't have to make a faster SRAM, they just need the on-chip interconnect which means they have to make the SRAM.
On top of that, everyone knows how to make SRAM anyways, there are really no new big secrets there. It is so simple they use it to test new fab processes.
The lowest end CPU you can find now-a-days is like 800Mhz, unless you go to auctions...
Yes, and that 800Mhz CPU with motherboard runs around $150 at Fry's in SoCal. Actually, for $179 I believe you can get an Athlon XP 1600+ with motherboard.
Let's compare that to the lower end hardware. The Wednesday Fry's ad has a K6-2 500 with motherboard for $79.
The low end seems to be the inverse of the high end. At the high end a small increase in speed results in a large increase in price. At the low end a large decrease in speed results in a small decrease in price. Try to find a P1-120 with motherboard for less than $79, and tell me would you rather pay a little extra for the K6-2 500 which is an VIA MVP4 chipset so you have onboard audio and video.
Dastardly
Regardless of the issues of taxation generally. My problem is why can't anyone get the math right. You get your withholding setup to pay the federal government jus the right amount over the course of the year, and whammo, you haven't paid the state (At least in California) enough by a significant amount. I am sure everyone uses the same tables, so why is it that the calculation for federal taxes comes out right, and the state taxes comes out wrong? It just seems incredibly stupid.
It's a "as long as it's the other guy that dies" philosophy that truly is disturbing.
I don't think this point can be emphasized enough. This attitude permeates the people who buy large vehicles because they are "safe". They are only safe to the few people in the vehicle. They are a hazard to every other person on the rode. But, of course that doesn't matter to the self serving bastards driving the biggest vehicles they can find, and putting their teenage children into the same sorts of vehicles so they can kill people.
And, of course none of them will admit that it is a kill or be killed mentality that makes them buy sucha vehicle.
Dastardly