Regardless of the facts or reality or the economics or even the technical plausibility. What Intel typically does is make older models too expensive or unavailable so man'f'turs are forced to use new chips and motherboards.
I for one am completely sick of newer faster hotter! Take some of the effort that goes into this and build me a reliable half-speed PC for $99 bucks. This is all starting to sound like the dopes who go out and buy a Ferrari and brag about how great it is - - at least in 2nd gear - - because there aren't any roads that they can go over ~80mph.
1 Criminalize Napster 2 Guaranty that artists see no money per record company contracts 3 Guaranty corporate profits a-la monopoly regulation 4 Uninvent the Atomic bomb
First off CD prices are up and unit sales are up about 8%.In all of the chazzerei about Napster I have never seen an artist nor the RIAA demonstrate, actually demonstrate how an artist is harmed by Napster when compared to no Napster. The simple fact is that the record industry pushes all of the risk to the artist in the form of advanced money to produce albums and videos which is then recoverable from the artist to the record company as money taken from the artists' take of the revenue. A second dip. To say nothing of the fact that most recording artists are extremely lucky to get 10% of each sale. The average is about half that or 5%. So the record company gets at least 90% of the dollars that come back to them from gross sales minus distribution costs, which in this case is whatever profit the retailer networks take out. From that 10% remaining artists typically give up another third or half in givebacks to the record company.
So look at it this way (and take an extreme postion): The sum total of all Napster has increased sales 8% of which translates to about $0.13/unit per artist from which they give back between 6 and 9 cents, leaving them with 4 to 7 cents/unit more money. Without Napster they would be no better off than before.
Now now I know all of you statisticians out there will say that Napster is not the sole reason unit sales are up 8%. OK pick some positive number less than 8. On the other hand you are arguing that sales otherwise would be 10 or 11 or 12% higher instead of 8 then the RIAA has to demonstrate that. How does one do that with a straight face?
"well we at the RIAA projected our sales to be 120% higher than the previous year therefore Napster ows us 2.2 times of the total revenue for all record companies in 1999"
"well we at the RIAA demand that all living souls purchase 15 CD's/year and anyone who doesn't is obviously a criminal"
...ya see it's pretty hard to make a case for that. It's one thing to demand to be treated as a monopoly, or more correctly, an oligopoly. IT's quite another to demand legal remedy to punish consumers who don't happen to believe that. And by the way if you want to be an oligopoly then by all means go to the Federal gov't and DEMAND restrictive regulation be placed on yourselves as any other protected utility. They will regulate things like your ROI, what you can invest in, how much you charge, etc.
What can we expect next? An RIAA surcharge on only the cheap seats at concerts because 'obviously' those people are avoiding their sacred duty to enrich the the record companies?
What utter fucking nonsense. The problem with security holes is: a)too many people know about them, b)they would go away if we just all pretended they didn't exist and c)the fault lies completely with evil little hackers and not with the people who wrote and can fix the problem.
Hey pundit-guy where do you think these problems get exposed from? From the vendor? Or from the results of dedicated efforts to break something.
Does anyone honestly believe that any commercial vendor would fix something no one knew about? Do you think any IT empty suit would tell their galley slaves to row faster if that person's boss could never find out what problems they are exposed to?
What do we call this? The rhythm method for IT security, that's what.
2 different arguments I'm afraid. and BTW hindsight is a poor argument. Anyway...perhaps we can look at the problem this way. There is a individual and we want to promote the best the healthiest, the smartest, etc. for that individual. A personal ubermensch as it were. Couples or single parents have a wide lattitude of what they consider 'advantaged', or better. For some it would be healthier or longer living. For others it would be smarts. For others still it would be some of the left brain activities like ability to paint or sing or write or do television commercials. For others yet it would be pure physical attractiveness. I can imagine some parents wanting their children not too too smart in lieu of being better socialized . Fine. That's the individual.
Populations have different criteria for wanting to engineer the person. Perhaps you friendly local government wants to reduce antisocial behavior or learning disabilities or the tendency to smoke rock. And the hell with everything else. Just not their problem whether the the resulting person is fat stupid and short lived or not. Populations really have no interest in individual performance whether that's mandated by science, gov't or evolution. It's simply not important. What is important is that those traits that drag down the population on average be eliminated or reduced. That is, eliminate the sickest or most violent 10% and the average performance of everybody else starts to look pretty damn good. The population as a whole does not benefit if you think your offspring have a better chance when they are right handed, left brained and a have a good golf swing. But the individual can benefit from unique tweaking if that person's parents made the right choices. <Let's hope the ability to make good choices about one's children is a trait that must be left in the pool>.
Before I ran out and bought a bunch of CD's I checked out some tracks for myself. This is obvious on it's face. Strip away all the blather and what I want is to listen to these tracks on a CD Player, in my car, @ home, whatever.
In the late 80's and early 90's we ran networks of AS/400's that had better uptime than our commercial MVS environment. Exactly 0 unplanned IPL's in 8.5 years. Only downtime was OS upgrades and tape drive replacements. Rock solid the most reliable stuff I've ever seen ever better than 4381's. But remember some of that came with operational discipline, planning and rigor we borrowed from the MVS world and its 30 years of refinement.
For gamers this is great. For everyone else running real world business applications OC is a tragic waste of time & money. If your job depends on keeping things running OC is too risky. If you have a few lab machines - fine; experiment. But in the REAL WORLD you're better off spending the money on faster disk or even SSD . Rebuilding my DB2 indicies just isn't going to be appreciably faster with big bad CPU's when disk performance is the throttle. For a coupla hundred bucks I'd rather get another U2W SCSI adapter, some more disks and parallel the I/O.
C'mon - colors? Ok that's cool for the mass market aka Nokia phones. But what about just building a tougher hardened case? Shockproof, waterproof? Generally something that you don't have to baby as much? True the screen will always be the weak point but there must be some way to make a mass market device for young people that's designed to be used the way young people would treat it - aka abusively. Heck I have a Nokia 636 analog cell phone you can pound nails in with. I was hoping that this was what Sony would bring to the party - a consumer oriented tough unit you could treat like a Discman.
Ummm doesn't Nokia already sell a huge number of phones that are already regulated by EU radiation standards? Can someone comment on this? If this this true then the whole act is just window dressing. AFAIK EU regs are years ahead of the US anyway. They were years ahead of us in monitor emissions at any rate.
I'd be much more concerned if my office mate's 21" monitor was beaming x-rays at my head all day out the back of the tube which is where most of the errant radiation comes from.
Like how great it was before electricity, paved roads and plumbing. All of it is just a veiled complaint against people from somewhere else who happen to live here. It's not fashionable to say you hate Northerners so you make a dull joke about all the Yankees moving in. Yeah growth is a problem, the roads are overloaded, the school system has problems - - blah blah blah.
Actually the liklihood of being successfully sued for malpractice is rather small in the US. While they would have you believe that malprac insurance is this huge onerous burden that spikes the cost of healthcare it is in fact roughly 7% of the avg. practictioners' revenue. Ans that does not include practitioners who are employees of HMOs and therefore pay no direct malprac insurance premium at all, unless it's imputed into their benefits deduction. And the liklihood of successfully suing at all is small to none since the first step of the process is peer review. Peers as in other doctors who aren't any more likely to rat out each other than cops. Then if you're lucky you get in front of a judge who will grant an immediate continuance pending a conference to pressure you into a settlement. If that doesn't work then you go to trial and watch you money fly out the window as the process grinds on for a few months or years.
Gasp. Civilization is disintegrating around us!! A person is using foul language in an auction site. Darkness is falling. Wild starving dogs prowl the streets. Incidents of cannibalism have been reported. God is dead. <BR> <BR>
What may change is the brand
on
Endgame For SCO
·
· Score: 1
SCO owns a great deal of technology that they licence to others. They own SVR4. They have a great deal of talent and most shops that use them are fairly happy. What may change is an SCO branded distribution on Unix. What should change is the less than efficient supoport path. We may see that replaced by SCOLinux or whatever they call it.It will be a baseline Linux distro with a bunch of SCO developed technology on top of it. Question is can they keep the company afloat until then?
I gave away a P200 a P201, two P72's and G72. The display on my notebook is fine. Half my office is filled with the three glass monitors, P70 and G70 that are left. It's nice that they're 17+" each and I'm sure the people @ home would want one but...
a pencil. No really it was. Then I got a slide rule. Eventually - - aeons later I got unrestricted use of a PDP-8, paper tape and all. Then the big time - - a Univac 1180/82E. Does anybody remember coding in octal?
Take all the money you spend on liquid nitrogen heat exchangers, Peltier plates and go buy an SSD array. Since most of the performance problem for most applications in the real world is I/O eliminate all the mechanical parts altogether. If you want better game fps go buy a high end video card.
OTOH we should be able to sue vendors and recover if they sell us a grey market machine that's relabeled - just like if we bought a used car sold as new.
Part of the problem is the adjusted launch cost/lb or kilo or whatever. Just as important is the minimum launch weight. In economic terms it's the hurdle cost or the minimally sufficient investment to get pound 1 off the ground. Any investment less than that minimum hurdle is a sunk cost with a net cost/lb of zero. In common sense terms launch weights have to be at least a certain lower limit and the cost of getting that minimal weight off the ground is some value "X" where the marginal cost of getting each additional pound off the ground is a decreasing function until you reach some technically limiting upper bound. When commercial organizations talk about cost per launch weight they are talking about the marginal part of the curve and do include the left tail of the curve or the hurdle rate. The key to getting total launch costs down is to shift the left saddle point of the curve further to the left. That is, create a system where one could feasibly launch an object that weighs only a few ounces or less. But since there is no practical technology that allows one to launch anything meaningful that weighs a few ounces or grams (except for Tim Leary's ashes) the baseline saddle point stays resolutely where it is today. Until some clever people can build a launch payload that is very very small you'll still need a booster that get a several pounds into orbit.
The differences in officially supported like CPU design speeds in the Intel world are a matter of manufacturing efficiency more than anything else. Basically what happens is that they take a design spec and start to bake wafers. As each stepping comes out of the oven a bunch of performance, reliability and heat tests are applied to the batch. After a given % of each stepping batch fails the family tests at a given speed then the family is labeled with a bench speed. Better wafers pass higher speed tests and therefore higher official speeds are attached to that family. As the process becomes more refined; better tolerances, cleaner lithography, cleaner materials, etc. each family is better able to withstand higher speed tests and so are certified at those higher speeds. It takes a different physical architecture of the CPU to make those larger jumps in speed before they also begin to fail higher and higher speeds. So basically what it comes down to is that the difference between 700 and 733 or any other 2 fairly close speed gaps is that the manufacturing/baking process has been improved and the family is able to pass a higher QA standard. Normally the process works like - take a basic design, make some chips, sell them, invest some money in improving the process, speed up the chips and continue until the cost of investing in the process is greater than the benefit realized. At that point you make subtle changes in the physical design, apply whatever manufacturing process to that variation and continue in your quest for manufacturing process improvement.
Sony was slowly losing their entire case. In the past few months, one claim after another was tossed out by the court; eg. had no merit. Sony was facing a difficult task on the remaining claims. And oh by the way.....
Sony is huge player in the games title business (er... I mean content). They didn't see the logic of going to court and pissing off a bunch of game evangelists all for the sake of preserving a low margin aspect of their business - the HARDWARE end when they could just coopt and force more games through the channel. Games is where they make the real money cause you invest to get the first one and then each next copy takes a few bucks to make, distribute and sell. Consoles are expensive to make, hard to ship, get returned....blah blah blah.
And anyway does emu run as well as a console? or is a BIG company like Sony banking that slow performance will motivate folks to switch to a console. And what about PS2?
There are only 2 classes of things that this information will ever be used for: body image improvement eg, weight loss, body sculpting, and replacement therapies are too expensive from an insurance company's/HMO's perspective like Multiple Sclerosis or HIV. Everything else will create nearly zero interest from any investors since the payback will be too far off or too low. Remember we're a nation of fat lazies or eating disorders and nobody wants to pay for long term care
Yeah the Japanese machines are probably TCM based sysplex units a-la older IBM type ES9000 mainframes. I believe the largest off the shelf TCM machine is a 12-way. These processors are enormously fast, consume tremendous amounts of electricity and throw off vast quantities of heat which is why they're water cooled. If we compare the performance of TCM units vs. the latest CMOS mainframe class CPU's it still takes about 3 CMOS to match the raw performance of 1 TCM. Now moving down the scale, the IBM-like mainframe class CMOS CPU's themselves are built specifically for mainframe machines and have very very high performance baselines. How high? Hard to tell since IBM will not publish performance benchmarks for mainframe machines that can be compared to other types or brands. Instead they use an internally derived benchmark that uses a 'commonly' know basic performance figure based for example on some well known IBM class mainframe like a 9021-831 or something like that. Any other machine is evaluated as a factor or that. At any rate the latest mainframe class CMOS machines have complexes or the rough analog of SMP cages that contain at least one CPU (up to..I don't remember, you can check). Each complex or base machine model is then sysplex'd to other same-type machines up to 12 or 14 machines or even higher. This is what the Hitachi/Fujitsu machines do. They build an x-way complex and then sysplex all the complexes together. As a rough comparison a VERY large commercial sysplex is typically a 12-way with each complex containing 12-24 individual processors for a total of 144 to 288 discrete CPU chips. This honestly is the high end of the high end for standard (non custom built special purpose) mainframe class machines. Compare this to an IBM RS/6000 SP2 frame with say 8 nodes of 12 processors each and ganging 20 or 30 or more frames together across a second level backplane switch for a total assembly of at least a few thousand discrete CPU's to more or less the same work. At least in the commercial world. In the nuclear simulation world obviously you want the highest possible FP performance so a TCM based mainframe design is enhanced with additional or different vector processors compared to simply exploiting the general purpose FP performance of whatever RISC CPU you're using. Another reason why the numbers of CPU's in the two classes of machines is so different.
When 3 of the top 5 PC companies announce they plan to use a completely new processor, this is a good thing. Consider that even in the best case the companies that do intend to support Crusoe will roll them out slowly while they make adjustments necessary for all of the quirks that will appear for video, sound, modem and network adapters....and so on. If Compaq and Dell don't want to commit, so what? Go buy another brand and stop complaining.
ERPimplementations are about business process and business requirements. Buy any ERP 'toolset' and take the box home. Open it up. What you will find inside is 2 pieces of paper. One says "you now have the right to hire gobs of consultants" and the other says "think very very very hard about how you want to run your business". If it were just a matter of implementing a bunch of modules then 80% of the attempts would not fail. Problem is that customers all think that ERP kits are silver slugs you can pull off the shelf, install and you're up & running. Not true. Not true. And while all of the so called experts decry ERP vendors for lagging with web integration the truth is that most customers don't understand their own businesses or their own processes and didn't design any of that to begin with. What they have is a bunch of organic business functions that grew up over time independent of each other. The ERP kits are deployed and what you end up with almost everytime is a paved cowpath. Just an automated way of doing the same wrong thing faster.
Except for Medicare Part B just about any local/state/Fed agency forces you to deal w/ multiple forms. Just about any insurance company forces you to deal with forms.
Slashdot Mirror
Regardless of the facts or reality or the economics or even the technical plausibility. What Intel typically does is make older models too expensive or unavailable so man'f'turs are forced to use new chips and motherboards.
I for one am completely sick of newer faster hotter! Take some of the effort that goes into this and build me a reliable half-speed PC for $99 bucks. This is all starting to sound like the dopes who go out and buy a Ferrari and brag about how great it is - - at least in 2nd gear - - because there aren't any roads that they can go over ~80mph.
1 Criminalize Napster
2 Guaranty that artists see no money per record company contracts
3 Guaranty corporate profits a-la monopoly regulation
4 Uninvent the Atomic bomb
It's the goddamn economics. Get it straight.
First off CD prices are up and unit sales are up about 8%.In all of the chazzerei about Napster I have never seen an artist nor the RIAA demonstrate, actually demonstrate how an artist is harmed by Napster when compared to no Napster. The simple fact is that the record industry pushes all of the risk to the artist in the form of advanced money to produce albums and videos which is then recoverable from the artist to the record company as money taken from the artists' take of the revenue. A second dip. To say nothing of the fact that most recording artists are extremely lucky to get 10% of each sale. The average is about half that or 5%. So the record company gets at least 90% of the dollars that come back to them from gross sales minus distribution costs, which in this case is whatever profit the retailer networks take out. From that 10% remaining artists typically give up another third or half in givebacks to the record company.
So look at it this way (and take an extreme postion): The sum total of all Napster has increased sales 8% of which translates to about $0.13/unit per artist from which they give back between 6 and 9 cents, leaving them with 4 to 7 cents/unit more money. Without Napster they would be no better off than before.
Now now I know all of you statisticians out there will say that Napster is not the sole reason unit sales are up 8%. OK pick some positive number less than 8. On the other hand you are arguing that sales otherwise would be 10 or 11 or 12% higher instead of 8 then the RIAA has to demonstrate that. How does one do that with a straight face?
"well we at the RIAA projected our sales to be 120% higher than the previous year therefore Napster ows us 2.2 times of the total revenue for all record companies in 1999"
"well we at the RIAA demand that all living souls purchase 15 CD's/year and anyone who doesn't is obviously a criminal"
...ya see it's pretty hard to make a case for that. It's one thing to demand to be treated as a monopoly, or more correctly, an oligopoly. IT's quite another to demand legal remedy to punish consumers who don't happen to believe that. And by the way if you want to be an oligopoly then by all means go to the Federal gov't and DEMAND restrictive regulation be placed on yourselves as any other protected utility. They will regulate things like your ROI, what you can invest in, how much you charge, etc.
What can we expect next? An RIAA surcharge on only the cheap seats at concerts because 'obviously' those people are avoiding their sacred duty to enrich the the record companies?
What utter fucking nonsense. The problem with security holes is: a)too many people know about them, b)they would go away if we just all pretended they didn't exist and c)the fault lies completely with evil little hackers and not with the people who wrote and can fix the problem.
Hey pundit-guy where do you think these problems get exposed from? From the vendor? Or from the results of dedicated efforts to break something.
Does anyone honestly believe that any commercial vendor would fix something no one knew about? Do you think any IT empty suit would tell their galley slaves to row faster if that person's boss could never find out what problems they are exposed to?
What do we call this? The rhythm method for IT security, that's what.
2 different arguments I'm afraid. and BTW hindsight is a poor argument. Anyway...perhaps we can look at the problem this way. There is a individual and we want to promote the best the healthiest, the smartest, etc. for that individual. A personal ubermensch as it were. Couples or single parents have a wide lattitude of what they consider 'advantaged', or better. For some it would be healthier or longer living. For others it would be smarts. For others still it would be some of the left brain activities like ability to paint or sing or write or do television commercials. For others yet it would be pure physical attractiveness. I can imagine some parents wanting their children not too too smart in lieu of being better socialized . Fine. That's the individual.
Populations have different criteria for wanting to engineer the person. Perhaps you friendly local government wants to reduce antisocial behavior or learning disabilities or the tendency to smoke rock. And the hell with everything else. Just not their problem whether the the resulting person is fat stupid and short lived or not. Populations really have no interest in individual performance whether that's mandated by science, gov't or evolution. It's simply not important. What is important is that those traits that drag down the population on average be eliminated or reduced. That is, eliminate the sickest or most violent 10% and the average performance of everybody else starts to look pretty damn good. The population as a whole does not benefit if you think your offspring have a better chance when they are right handed, left brained and a have a good golf swing. But the individual can benefit from unique tweaking if that person's parents made the right choices. <Let's hope the ability to make good choices about one's children is a trait that must be left in the pool>.
Anyway that's all I can think of for now.
Before I ran out and bought a bunch of CD's I checked out some tracks for myself. This is obvious on it's face. Strip away all the blather and what I want is to listen to these tracks on a CD Player, in my car, @ home, whatever.
In the late 80's and early 90's we ran networks of AS/400's that had better uptime than our commercial MVS environment. Exactly 0 unplanned IPL's in 8.5 years. Only downtime was OS upgrades and tape drive replacements. Rock solid the most reliable stuff I've ever seen ever better than 4381's. But remember some of that came with operational discipline, planning and rigor we borrowed from the MVS world and its 30 years of refinement.
For gamers this is great. For everyone else running real world business applications OC is a tragic waste of time & money. If your job depends on keeping things running OC is too risky. If you have a few lab machines - fine; experiment. But in the REAL WORLD you're better off spending the money on faster disk or even SSD . Rebuilding my DB2 indicies just isn't going to be appreciably faster with big bad CPU's when disk performance is the throttle. For a coupla hundred bucks I'd rather get another U2W SCSI adapter, some more disks and parallel the I/O.
C'mon - colors? Ok that's cool for the mass market aka Nokia phones. But what about just building a tougher hardened case? Shockproof, waterproof? Generally something that you don't have to baby as much? True the screen will always be the weak point but there must be some way to make a mass market device for young people that's designed to be used the way young people would treat it - aka abusively. Heck I have a Nokia 636 analog cell phone you can pound nails in with. I was hoping that this was what Sony would bring to the party - a consumer oriented tough unit you could treat like a Discman.
Guess I just have unrealistic expectations.
Ummm doesn't Nokia already sell a huge number of phones that are already regulated by EU radiation standards? Can someone comment on this? If this this true then the whole act is just window dressing. AFAIK EU regs are years ahead of the US anyway. They were years ahead of us in monitor emissions at any rate.
I'd be much more concerned if my office mate's 21" monitor was beaming x-rays at my head all day out the back of the tube which is where most of the errant radiation comes from.
Like how great it was before electricity, paved roads and plumbing. All of it is just a veiled complaint against people from somewhere else who happen to live here. It's not fashionable to say you hate Northerners so you make a dull joke about all the Yankees moving in. Yeah growth is a problem, the roads are overloaded, the school system has problems - - blah blah blah.
Actually the liklihood of being successfully sued for malpractice is rather small in the US. While they would have you believe that malprac insurance is this huge onerous burden that spikes the cost of healthcare it is in fact roughly 7% of the avg. practictioners' revenue. Ans that does not include practitioners who are employees of HMOs and therefore pay no direct malprac insurance premium at all, unless it's imputed into their benefits deduction. And the liklihood of successfully suing at all is small to none since the first step of the process is peer review. Peers as in other doctors who aren't any more likely to rat out each other than cops. Then if you're lucky you get in front of a judge who will grant an immediate continuance pending a conference to pressure you into a settlement. If that doesn't work then you go to trial and watch you money fly out the window as the process grinds on for a few months or years.
Gasp. Civilization is disintegrating around us!! A person is using foul language in an auction site. Darkness is falling. Wild starving dogs prowl the streets. Incidents of cannibalism have been reported. God is dead.
<BR>
<BR>
SCO owns a great deal of technology that they licence to others. They own SVR4. They have a great deal of talent and most shops that use them are fairly happy. What may change is an SCO branded distribution on Unix. What should change is the less than efficient supoport path. We may see that replaced by SCOLinux or whatever they call it.It will be a baseline Linux distro with a bunch of SCO developed technology on top of it. Question is can they keep the company afloat until then?
I gave away a P200 a P201, two P72's and G72. The display on my notebook is fine. Half my office is filled with the three glass monitors, P70 and G70 that are left. It's nice that they're 17+" each and I'm sure the people @ home would want one but...
a pencil. No really it was. Then I got a slide rule. Eventually - - aeons later I got unrestricted use of a PDP-8, paper tape and all. Then the big time - - a Univac 1180/82E. Does anybody remember coding in octal?
Take all the money you spend on liquid nitrogen heat exchangers, Peltier plates and go buy an SSD array. Since most of the performance problem for most applications in the real world is I/O eliminate all the mechanical parts altogether. If you want better game fps go buy a high end video card.
OTOH we should be able to sue vendors and recover if they sell us a grey market machine that's relabeled - just like if we bought a used car sold as new.
Part of the problem is the adjusted launch cost/lb or kilo or whatever. Just as important is the minimum launch weight. In economic terms it's the hurdle cost or the minimally sufficient investment to get pound 1 off the ground. Any investment less than that minimum hurdle is a sunk cost with a net cost/lb of zero. In common sense terms launch weights have to be at least a certain lower limit and the cost of getting that minimal weight off the ground is some value "X" where the marginal cost of getting each additional pound off the ground is a decreasing function until you reach some technically limiting upper bound. When commercial organizations talk about cost per launch weight they are talking about the marginal part of the curve and do include the left tail of the curve or the hurdle rate. The key to getting total launch costs down is to shift the left saddle point of the curve further to the left. That is, create a system where one could feasibly launch an object that weighs only a few ounces or less. But since there is no practical technology that allows one to launch anything meaningful that weighs a few ounces or grams (except for Tim Leary's ashes) the baseline saddle point stays resolutely where it is today. Until some clever people can build a launch payload that is very very small you'll still need a booster that get a several pounds into orbit.
The differences in officially supported like CPU design speeds in the Intel world are a matter of manufacturing efficiency more than anything else. Basically what happens is that they take a design spec and start to bake wafers. As each stepping comes out of the oven a bunch of performance, reliability and heat tests are applied to the batch. After a given % of each stepping batch fails the family tests at a given speed then the family is labeled with a bench speed. Better wafers pass higher speed tests and therefore higher official speeds are attached to that family. As the process becomes more refined; better tolerances, cleaner lithography, cleaner materials, etc. each family is better able to withstand higher speed tests and so are certified at those higher speeds. It takes a different physical architecture of the CPU to make those larger jumps in speed before they also begin to fail higher and higher speeds. So basically what it comes down to is that the difference between 700 and 733 or any other 2 fairly close speed gaps is that the manufacturing/baking process has been improved and the family is able to pass a higher QA standard. Normally the process works like - take a basic design, make some chips, sell them, invest some money in improving the process, speed up the chips and continue until the cost of investing in the process is greater than the benefit realized. At that point you make subtle changes in the physical design, apply whatever manufacturing process to that variation and continue in your quest for manufacturing process improvement.
Sony was slowly losing their entire case. In the past few months, one claim after another was tossed out by the court; eg. had no merit. Sony was facing a difficult task on the remaining claims. And oh by the way .....
Sony is huge player in the games title business (er... I mean content). They didn't see the logic of going to court and pissing off a bunch of game evangelists all for the sake of preserving a low margin aspect of their business - the HARDWARE end when they could just coopt and force more games through the channel. Games is where they make the real money cause you invest to get the first one and then each next copy takes a few bucks to make, distribute and sell. Consoles are expensive to make, hard to ship, get returned....blah blah blah.
And anyway does emu run as well as a console? or is a BIG company like Sony banking that slow performance will motivate folks to switch to a console. And what about PS2?
There are only 2 classes of things that this information will ever be used for: body image improvement eg, weight loss, body sculpting, and replacement therapies are too expensive from an insurance company's/HMO's perspective like Multiple Sclerosis or HIV. Everything else will create nearly zero interest from any investors since the payback will be too far off or too low. Remember we're a nation of fat lazies or eating disorders and nobody wants to pay for long term care
Yeah the Japanese machines are probably TCM based sysplex units a-la older IBM type ES9000 mainframes. I believe the largest off the shelf TCM machine is a 12-way. These processors are enormously fast, consume tremendous amounts of electricity and throw off vast quantities of heat which is why they're water cooled. If we compare the performance of TCM units vs. the latest CMOS mainframe class CPU's it still takes about 3 CMOS to match the raw performance of 1 TCM. Now moving down the scale, the IBM-like mainframe class CMOS CPU's themselves are built specifically for mainframe machines and have very very high performance baselines. How high? Hard to tell since IBM will not publish performance benchmarks for mainframe machines that can be compared to other types or brands. Instead they use an internally derived benchmark that uses a 'commonly' know basic performance figure based for example on some well known IBM class mainframe like a 9021-831 or something like that. Any other machine is evaluated as a factor or that. At any rate the latest mainframe class CMOS machines have complexes or the rough analog of SMP cages that contain at least one CPU (up to..I don't remember, you can check). Each complex or base machine model is then sysplex'd to other same-type machines up to 12 or 14 machines or even higher. This is what the Hitachi/Fujitsu machines do. They build an x-way complex and then sysplex all the complexes together. As a rough comparison a VERY large commercial sysplex is typically a 12-way with each complex containing 12-24 individual processors for a total of 144 to 288 discrete CPU chips. This honestly is the high end of the high end for standard (non custom built special purpose) mainframe class machines. Compare this to an IBM RS/6000 SP2 frame with say 8 nodes of 12 processors each and ganging 20 or 30 or more frames together across a second level backplane switch for a total assembly of at least a few thousand discrete CPU's to more or less the same work. At least in the commercial world. In the nuclear simulation world obviously you want the highest possible FP performance so a TCM based mainframe design is enhanced with additional or different vector processors compared to simply exploiting the general purpose FP performance of whatever RISC CPU you're using. Another reason why the numbers of CPU's in the two classes of machines is so different.
When 3 of the top 5 PC companies announce they plan to use a completely new processor, this is a good thing. Consider that even in the best case the companies that do intend to support Crusoe will roll them out slowly while they make adjustments necessary for all of the quirks that will appear for video, sound, modem and network adapters....and so on. If Compaq and Dell don't want to commit, so what? Go buy another brand and stop complaining.
ERPimplementations are about business process and business requirements. Buy any ERP 'toolset' and take the box home. Open it up. What you will find inside is 2 pieces of paper. One says "you now have the right to hire gobs of consultants" and the other says "think very very very hard about how you want to run your business". If it were just a matter of implementing a bunch of modules then 80% of the attempts would not fail. Problem is that customers all think that ERP kits are silver slugs you can pull off the shelf, install and you're up & running. Not true. Not true. And while all of the so called experts decry ERP vendors for lagging with web integration the truth is that most customers don't understand their own businesses or their own processes and didn't design any of that to begin with. What they have is a bunch of organic business functions that grew up over time independent of each other. The ERP kits are deployed and what you end up with almost everytime is a paved cowpath. Just an automated way of doing the same wrong thing faster.
Except for Medicare Part B just about any local/state/Fed agency forces you to deal w/ multiple forms. Just about any insurance company forces you to deal with forms.