I'll call your bluff, because less than a year ago, when they did some special investigations at, of all places, a major airport in the US, they found that a very significant portion of the workers (with access to restricted areas in the airport nonetheless) were people who gave the airport authorities false ssn's
My "bluff"? Who is "they" anyway, and what are these "special investigations"? What is "significant"? And what is "false SSNs"?
That's conveniently vague counter evidence you've used against me. And furthermore, does it support the claim that illegals are paying $9B per year? Nope. Ok then...
Illegal immigrants pay roughly $9 billion a year into the Social Security fund. Please post a reference... I am interested in understanding how they came up with this number.
Employers cannot screen whether or not a Social Security number is real Employers submit taxes which are tagged to SSNs every year. Most employers submit these taxes quarterly. The IRS conducts audits and screens for invalid SSNs. So, no the employers cannot screen, but the IRS does this for them.
Most of the people who are illegal aliens are not being "payed" (nice spelling) under the table I have seen many estimates of the number of illegal aliens in USA ranging from 5 to 12 million. The INS estimated 5 million in 1996.
I don't know how many of these are actually working, but assuming every man, woman, and child among them work (highly unlikely), and assuming all of them average poverty level ($12K, or $48/day, again highly unlikely) and all are using false SSNs that don't collide with other taxpayers (extremely unlikely), then at 6.8% they would be contributing $816 each, or about $4.1B. Assuming you include the employer's matching contribution, it is only $8.2B.
Because every assumption you made is at best highly unlikely, I consider this number a complete fabrication by the left.
Don't forget that there is nothing forcing anyone to "upgrade" to the v3 license. You are free to continue to license your code per the 1.0 or 2.0 terms.
The only thing that may/will get affected are new revisions of official GNU projects which will undoubtedly adopt the updated licenses.
yeah, I was more responding to the grandparent than the parent. My apologies to the parent...:)
RTFA, you big freaking slashdot lemming
on
Drafting GPL3
·
· Score: 2, Insightful
As for (4), who cares except for RMS what literary works he's published? More hubris.
First of all, this isn't about literary works that he's published. If you read the article, you would know that he is simply making the statement, "As the original concept creator and GPL author, and the largest reason it is around today, I am concerned about its future and want to make sure its meaning is intact".
Second of all, millions of people care about this one literary work that he's published, which is the GPL. The GPL is the foundation of all that is sacred for thousands of programmers and developers who want to give their hard work and accomplishments to the rest of the world without having their precious creation raped and pillaged by corporate greed. Furthermore, it is a revolutionary idea that through RMSs significant work and sacrifice has become the cornerstone of a vast amount of the Free software available today. For RMS to be concerned with its future direction is understandable if not commendable.
Dont worry... they don't "intersect": There won't be a collision in the middle, causing a traffic pile-up, with bodies strewn alongside, drawing emergency vehicles, and rubberneckers to clog up the opposing traffic.
These "intersecting" flow vectors will simply end up contributing to a single resulting flux that is somewhat top-left to bottom-right overall.
Thank you for clarifying that. He is definitely anti-software-restriction more than anything else.
In a way, he might even appreciate Microsoft for helping inspire so many to develop free apps (for both linux and win32).:)
I have a lot of respect for the man and what he has done. When I was in my early 20s about 10 years ago, he sent me an email which I still have saved somewhere. I was truly in awe that he was humble and generous enough to respond to my newbie post.
Raging anti-Microsoft zealots might though, but that's a segment of the population I think we can do without
Perhaps Mr. Torvalds is no anti-Microsoft zealot. But you can't say the same about many of the early leaders of the Free software world.
In particular, Richard Stallman, who is largely responsible for gcc, emacs, binary tools, the FSF foundation, the GPL license. He is even (arguably) largely responsible for the success of Linux in general. Yet he is incredibly anti-Microsoft and very well criticized. Granted he is not one of the zealots that are linux-illiterate, and so he will not be converting to osx any time soon.
But to answer your assertion above, no, I don't think the Linux community could have flourished without that segment of the population.
Perhaps you are new to the free software movement, and so I forgive you.
That was closer to what I meant:) sorry, that was an ambiguous statement that made sense to me at the time.
What I meant was the intellectual property value is not in the ink, it is in the printer. There is no intellectual property being violated when a customer puts counterfeit ink in the printer, whereas one cannot necessarily say the same about a counterfeit disc being played in a Nintindo console.
I am not sure the patent on the authentication chip had anything to do with it whatsoever.
Don't forget these other important facts:
* Nintendo is using their DRM technology to protect copyrighted software. Lexmark is using it to lock out competitors from using ink, which is not copyrightable. * A gaming console is a mechanism for playing games. The value of a game is contained on the copyrighted media. The game console checks for violations prior to playing the game. A printer is not simply a mechanism for using ink. The value of the printer is not in the ink, and there is no reason for the printer to check for copyright violations prior to using ink.
Console? You mean like xbox? What does Apple have to do with consoles?
Intel exited the console market because there is no margin in it. Microsoft and Sony are selling their consoles at a loss, why would they allow the silicon to get a disproportionate margin?
LOL... I'm not sure what you mean by "walking all over them" unless you mean a currently decreasing, less than 20% market share in both laptops and desktops, whining about how Intel violated the "rules of dual-core" by sneaking 2 pieces of silicon into the same package, or as you mentioned, failing to penetrate the AMD and now Apple markets, or by failing to capitalize of the potential lead they had in 64-bit extensions, no-execute, or multi-core, or by failing to lower their costs or scale up volumes sufficiently to achieve design win, or maybe you mean that AMD has about 1/8 the revenue and has operated at a loss recently more often than not.
It's almost like they are *trying* to lose. Contrary to what you might think, the goal of a business is to deliver what people want and make money doing it. OTOH, I can see why you and the rest of AMD would want Intel to convert to another market. LOL.
Intel manufacturing PPCs is far more probable than Apple jumping to x86
But...
Why would Intel go this route. They have higher margins than any high-volume silicon foundry in the world. There are very few other companies that come close to that revenue. This approach wouldn't add much of anything to their bottom line. I don't think Intel would go for that.
It has to be either (a) Apple spooking IBM (a la Dell and the annual AMD bluff), or (b) Intel increasing their x86 proliferation
If indeed (b), then Apple will certainly provide virtual layer to emulate the IBM platform. Perhaps they are ready to move forward on this and can make a free H/W decision. It would also follow that Apple can run Windows apps as well, like Wine, or at least full-blown windows itself.
LOL... none of that is even relevant. All the technical superiority gets them nowhere if they are pricing themselves out of the market.
Yes, as I have said before, Intel probably implemented a trivial SMP configuration. Not that I would call that a hack by any means, but you call it what you want. Regardless, this got them to market first and reduced development costs. And a largely-independent decision was to package two die in the same package. This again saved a huge amount of cost. So they deliver a slightly-lower performing part at half the cost, I don't think the market is in tune with "technical superiority" enough to pay 2x. AMD knows this, and is hoping that the absurd Intel-is-cheating-you-by-sneaking-2-pieces-of-sili con-into-your-dual-core-processor argument will buy them back some Average Joe credibility.
It is ironic that time and again AMD brings the "technically superior" solution to market, and time and again Intel delivers the most marketable solution. I guess AMD is content with that because some things never change.
Separate die do not preclude the benefits of a single plane. They could be placed side-by-side if heat dissipation is critical and adequate heat dissipation vehicle/substrate cannot be packaged adjacent to the stacked die. But... I am definitely not an expert on packaging technology.
But your question is definitely a valid one. There are countless architectural reasons why AMD would perform better. They could have better queueing mechanisms, faster IPC (inter-process communication), faster interconnect, differing arbitration, better OS support, better preemption support, better snoop support, etc.
From what I understand, AMD spent more time architecting their multi-core interconnect. The concensus here (on/.) is that Intel did a very straightforward interconnect. I am convinced the performance difference has nothing to do with the packaging technology. I wish one of fan sites would do a good writeup on the actual architectural differences between the two.
Perhaps what Intel delivered is as trivial as a typical multi-processor implementation. Perhaps the only difference is that the two cores reside in the same package with exactly the same connectivity as two independent processors. This may be what AMD is trying to say with their "doubt" article above. But, I wish AMD lots of luck in describing the difference to the average Joe out there. Instead they simply say:
It is a point to point protocol where both devices provide multiple agents on SEPARATE CHANNELS to maintain the control/data protocol. All of these agents you mention are not simply "wired together", but rather are connected point to point.
So, you cannot place multiple graphics devices on an AGP PORT therefore it is a point to point port indeed. Some MCH controllers support multiple AGP devices, and pinout multiple AGP PORTs, however most do not.
Well I care if they are on the same die or not. Being on the same die means faster communication between them False. it also means less heat False. which means a smaller heatsink, less noise and less electricity consumed. False.
Not only are these assmptions not inherently coupled with the # die integrated in the package, but... separating the cores out to multiple die actually can reduce latencies and enable better connectivity due to stacking, and furthermore has a drastic impact on yield, test time, and ultimately cost.
That caught you by surprise? You mean you actually figured out you had something called a "video card", that it was actually inside the box you have sitting on your desk, AND, you have opened successfully up your case, AND you identified which card was the video card somehow, AND you are considering putting a new one in its place, AND you have gone through the trouble to determine what is lacking about your card and which card is the best upgrade for it? All that trouble, AND you didn't think to read the NAME of the video card to notice whether it said "PCI-Express" or "AGP"?
Man, that is one big coincidence... what are the odds of all that happening to the same unfortunate soul? Well, what the hell... blame it on Intel, right?
I know this sounds like a flame, but I ave your best interests at heart... You should totally not be opening up your PC, much less believing the anal drivel AMD resorts to. What, are there rules to dual-core? And AMD gets to set these rules? Even if this means their processor will cost twice as much and also return lower margins? Yeah, they might get a few percent performance on a few benchmarks, but at twice the cost, who wants that?
Dual-core simply means you have two tightly-coupled cores. Generally this implies they are in the same package at least, and there is low-bandwidth interconnect, some means for IPC, probably a shared chipset, and some amount of shared memory hierarchy, perhaps a shared L2 or L3 cache.
The reason most people think it is synonymous with dual-cores-on-the-same-die is simply because it has historically not been simple or cheap enough to integrate multiple die in the same package. Intel has the capability to do it, but most others cannot.
As your parent post mentioned, this gives Intel a significant advantage in price (due to yield and test time and complexity improvements) as well as potential package size (depending on whether the die are stacked or side-by-side). And there are heat ramifications b/c stacked results in less surface area and higher heat density. Aside from Intel's alreadly higher heat dissipation.
Simply because AMD does not have the manufacturing capability, they are trying to FUD Intel's advantage here. What a ridiculously cheap and desparate move -- LOL! Believe me, if AMD could integrate multiple die as cheaply, they would do it too.
This is one of the reasons why AMD dual-core costs about twice what an Intel one costs. Although AMD wll run benchmarks faster, all the consumer will see is "a dual-core Pentium 4, cheaper than a dual-core AMD... cool!"
Here's my prediction. In 10 years, everything will run on x86. Yes, even PDAs and cell phones. The risk to Intel is that more competitors start building x86 processors (AMD is too small). Maybe Samsung or some other company with the capital requirements and semiconductor expertise necessary to build out fabs.
Good observation... Samsung could completely swallow up AMD, TMTA, Via Centaur, etc., and with them all their IP, patents, and other cross-licensing leverage, giving them an easy entry into the ludicrous world of x86. Combine this with the Korea's very "favorable" business conditions toward Samsung, Samsung's insane ability to down-pressure costs, complete silicon-to-product vertical solutions, existing memory and flash business, and they would be a force to reckon with.
As funny as that would be, Dell is not scared of this. In fact, if the deal went through (and I don't think it will - at least in the PC business) it would be in all three's interests to have Dell involved.
Apple SW + Intel CPU/chipset + Dell mobo and sales... the combined hype and marketing force alone would topple mountains, not to mention put a dent in Microsoft's ego and even profits.
Oh yeah, and throw a little Google in there somewhere too... desktop search or something... lol
That is incredibly difficult to believe for many reasons. Let me count the ways....
1) 99% of today's software runs on x86. Nobody wants to dump all of their software and migrate to a new ISA. Why else would CPU manufacturers continue to support legacy x86 even at the cost of up to 10% of their area and power budget just to decode x86 CISC instructions to RISC u-ops 2) There is an enormous business infrastructure built around this. It would all but KILL microsoft's reputation in the business world 3) The vast majority of Microsoft's revenue is on x86 software. It would be suicide for them to "theoretically begin to endorse" a new PC architecture that a) does not have the capacity and credibility to supply the world with enough PCs (ever wondered why Dell doesn't source from AMD???), b) does not have enough software to satisfy the demand, c) doesn't even have an owner stepping up to the plate (didn't IBM just sell off its PC business to Lenovo), d) can't provide the lowest prices 4) Nature abhores a vacuum. Any number of software vendors would love to get a crack at the x86 market that Microsoft vacated. Again, this would be suicide for Microsoft. 5) Intel has plenty of internal software, drivers, development tools, etc, not to mention an absolutely enormous amount of open-source win32 software and linux software 6) Every consumer service provider and hardware vendor in the world supports WIN32 on x86. Microsoft would be starting a platform from scratch that nobody would buy because the market does not sell anything for it: a) IO devices, b) broadband/VoIP/VPN, c) all the software and games they are used to 7) Even IBM would be a fool to think they could survive without x86 platforms to install their software and services onto 8) Not to mention that what you are saying is far from original -- the market has been saying for YEARS and YEARS that Intel is doomed due to a narrow focus. And yet just last quarter, Intel reported record revenue and profit. Their stock is taking off as investors expect great future growth. More than ever before, even more than during the dot.com hayday. The continue to beat down AMD to lower market share. They have pommelled TMTA and Via into oblivion. Please tell me how lack of diversity has been hurting them. Their margins are still in the 60% range, they went through the entire dot-bomb without posting a single quarterly loss (unlike ANY other large tech company I can think of), they are the first to 30cm wafer production, first to 65nm geometry in volume 9) And they do diversify. They have revolutionized the laptop platform, all but taken over the high-performance (i.e. high-margin) PDA and portable computing marketplace. They are a market leader in NOR flash (again beating out AMD to the point where they must sell their flash devision, not the same definition of "do very well" you must be thinking of). http://www.intel.com/pressroom/archive/releases/20 050302net_a.htm http://news.com.com/AMD+expects+flash+memory+to+hu rt+revenue/2100-1006_3-5521587.html They are pushing WiMAX to the market as a viable competitor to both cellular technology and cable/dsl broadband, and they are the first to bring wimax silicon to the market http://www.intel.com/ca/pressroom/2005/0418.htm They have a single-chip cellular GPRS baseband and high-performance application processor for entering the phone/PDA market. http://www.intel.com/design/pca/prodbref/252336.ht m 10) Time and time again, Intel has proven its marketing and execution genious, bringing to market products that are not necessarily the most academically superior, but certainly
Slashdot Mirror
I'll call your bluff, because less than a year ago, when they did some special investigations at, of all places, a major airport in the US, they found that a very significant portion of the workers (with access to restricted areas in the airport nonetheless) were people who gave the airport authorities false ssn's
My "bluff"? Who is "they" anyway, and what are these "special investigations"? What is "significant"? And what is "false SSNs"?
That's conveniently vague counter evidence you've used against me. And furthermore, does it support the claim that illegals are paying $9B per year? Nope. Ok then...
Illegal immigrants pay roughly $9 billion a year into the Social Security fund.
Please post a reference... I am interested in understanding how they came up with this number.
Employers cannot screen whether or not a Social Security number is real
Employers submit taxes which are tagged to SSNs every year. Most employers submit these taxes quarterly. The IRS conducts audits and screens for invalid SSNs. So, no the employers cannot screen, but the IRS does this for them.
Most of the people who are illegal aliens are not being "payed" (nice spelling) under the table
I have seen many estimates of the number of illegal aliens in USA ranging from 5 to 12 million. The INS estimated 5 million in 1996.
I don't know how many of these are actually working, but assuming every man, woman, and child among them work (highly unlikely), and assuming all of them average poverty level ($12K, or $48/day, again highly unlikely) and all are using false SSNs that don't collide with other taxpayers (extremely unlikely), then at 6.8% they would be contributing $816 each, or about $4.1B. Assuming you include the employer's matching contribution, it is only $8.2B.
Because every assumption you made is at best highly unlikely, I consider this number a complete fabrication by the left.
The end.
Don't forget that there is nothing forcing anyone to "upgrade" to the v3 license. You are free to continue to license your code per the 1.0 or 2.0 terms.
The only thing that may/will get affected are new revisions of official GNU projects which will undoubtedly adopt the updated licenses.
yeah, I was more responding to the grandparent than the parent. My apologies to the parent... :)
As for (4), who cares except for RMS what literary works he's published? More hubris.
First of all, this isn't about literary works that he's published. If you read the article, you would know that he is simply making the statement, "As the original concept creator and GPL author, and the largest reason it is around today, I am concerned about its future and want to make sure its meaning is intact".
Second of all, millions of people care about this one literary work that he's published, which is the GPL. The GPL is the foundation of all that is sacred for thousands of programmers and developers who want to give their hard work and accomplishments to the rest of the world without having their precious creation raped and pillaged by corporate greed. Furthermore, it is a revolutionary idea that through RMSs significant work and sacrifice has become the cornerstone of a vast amount of the Free software available today. For RMS to be concerned with its future direction is understandable if not commendable.
Dont worry... they don't "intersect": There won't be a collision in the middle, causing a traffic pile-up, with bodies strewn alongside, drawing emergency vehicles, and rubberneckers to clog up the opposing traffic.
These "intersecting" flow vectors will simply end up contributing to a single resulting flux that is somewhat top-left to bottom-right overall.
Thank you for clarifying that. He is definitely anti-software-restriction more than anything else.
:)
In a way, he might even appreciate Microsoft for helping inspire so many to develop free apps (for both linux and win32).
I have a lot of respect for the man and what he has done. When I was in my early 20s about 10 years ago, he sent me an email which I still have saved somewhere. I was truly in awe that he was humble and generous enough to respond to my newbie post.
Raging anti-Microsoft zealots might though, but that's a segment of the population I think we can do without
Perhaps Mr. Torvalds is no anti-Microsoft zealot. But you can't say the same about many of the early leaders of the Free software world.
In particular, Richard Stallman, who is largely responsible for gcc, emacs, binary tools, the FSF foundation, the GPL license. He is even (arguably) largely responsible for the success of Linux in general. Yet he is incredibly anti-Microsoft and very well criticized. Granted he is not one of the zealots that are linux-illiterate, and so he will not be converting to osx any time soon.
But to answer your assertion above, no, I don't think the Linux community could have flourished without that segment of the population.
Perhaps you are new to the free software movement, and so I forgive you.
That was closer to what I meant :) sorry, that was an ambiguous statement that made sense to me at the time.
What I meant was the intellectual property value is not in the ink, it is in the printer. There is no intellectual property being violated when a customer puts counterfeit ink in the printer, whereas one cannot necessarily say the same about a counterfeit disc being played in a Nintindo console.
I am not sure the patent on the authentication chip had anything to do with it whatsoever.
Don't forget these other important facts:
* Nintendo is using their DRM technology to protect copyrighted software. Lexmark is using it to lock out competitors from using ink, which is not copyrightable.
* A gaming console is a mechanism for playing games. The value of a game is contained on the copyrighted media. The game console checks for violations prior to playing the game. A printer is not simply a mechanism for using ink. The value of the printer is not in the ink, and there is no reason for the printer to check for copyright violations prior to using ink.
Console? You mean like xbox? What does Apple have to do with consoles?
Intel exited the console market because there is no margin in it. Microsoft and Sony are selling their consoles at a loss, why would they allow the silicon to get a disproportionate margin?
LOL... I'm not sure what you mean by "walking all over them" unless you mean a currently decreasing, less than 20% market share in both laptops and desktops, whining about how Intel violated the "rules of dual-core" by sneaking 2 pieces of silicon into the same package, or as you mentioned, failing to penetrate the AMD and now Apple markets, or by failing to capitalize of the potential lead they had in 64-bit extensions, no-execute, or multi-core, or by failing to lower their costs or scale up volumes sufficiently to achieve design win, or maybe you mean that AMD has about 1/8 the revenue and has operated at a loss recently more often than not.
It's almost like they are *trying* to lose. Contrary to what you might think, the goal of a business is to deliver what people want and make money doing it. OTOH, I can see why you and the rest of AMD would want Intel to convert to another market. LOL.
Sources shall remain nameless, but they are from Transmeta :)
Ah, then those rumors will remain as credible as the rest of Tansmeta's track record.
Transmeta was all about rumors. Oh, and gulping down VC and swindling investors out of hundreds of millions of dollars.
Intel manufacturing PPCs is far more probable than Apple jumping to x86
But...
Why would Intel go this route. They have higher margins than any high-volume silicon foundry in the world. There are very few other companies that come close to that revenue. This approach wouldn't add much of anything to their bottom line. I don't think Intel would go for that.
It has to be either (a) Apple spooking IBM (a la Dell and the annual AMD bluff), or (b) Intel increasing their x86 proliferation
If indeed (b), then Apple will certainly provide virtual layer to emulate the IBM platform. Perhaps they are ready to move forward on this and can make a free H/W decision. It would also follow that Apple can run Windows apps as well, like Wine, or at least full-blown windows itself.
Intel knows a thing or about RISC chips.
And don't forget that Pentiums are actually RISC engines running the decoded x86 (CISC) ISA
LOL... none of that is even relevant. All the technical superiority gets them nowhere if they are pricing themselves out of the market.
i con-into-your-dual-core-processor argument will buy them back some Average Joe credibility.
Yes, as I have said before, Intel probably implemented a trivial SMP configuration. Not that I would call that a hack by any means, but you call it what you want. Regardless, this got them to market first and reduced development costs. And a largely-independent decision was to package two die in the same package. This again saved a huge amount of cost. So they deliver a slightly-lower performing part at half the cost, I don't think the market is in tune with "technical superiority" enough to pay 2x. AMD knows this, and is hoping that the absurd Intel-is-cheating-you-by-sneaking-2-pieces-of-sil
It is ironic that time and again AMD brings the "technically superior" solution to market, and time and again Intel delivers the most marketable solution. I guess AMD is content with that because some things never change.
Separate die do not preclude the benefits of a single plane. They could be placed side-by-side if heat dissipation is critical and adequate heat dissipation vehicle/substrate cannot be packaged adjacent to the stacked die. But... I am definitely not an expert on packaging technology.
/.) is that Intel did a very straightforward interconnect. I am convinced the performance difference has nothing to do with the packaging technology. I wish one of fan sites would do a good writeup on the actual architectural differences between the two.
But your question is definitely a valid one. There are countless architectural reasons why AMD would perform better. They could have better queueing mechanisms, faster IPC (inter-process communication), faster interconnect, differing arbitration, better OS support, better preemption support, better snoop support, etc.
From what I understand, AMD spent more time architecting their multi-core interconnect. The concensus here (on
Perhaps what Intel delivered is as trivial as a typical multi-processor implementation. Perhaps the only difference is that the two cores reside in the same package with exactly the same connectivity as two independent processors. This may be what AMD is trying to say with their "doubt" article above. But, I wish AMD lots of luck in describing the difference to the average Joe out there. Instead they simply say:
"2 pieces of silicon != dual-core"
And that just makes them look desparate to me.
Yes, the name says it all.
It is a point to point protocol where both devices provide multiple agents on SEPARATE CHANNELS to maintain the control/data protocol. All of these agents you mention are not simply "wired together", but rather are connected point to point.
So, you cannot place multiple graphics devices on an AGP PORT therefore it is a point to point port indeed. Some MCH controllers support multiple AGP devices, and pinout multiple AGP PORTs, however most do not.
Well I care if they are on the same die or not. Being on the same die means faster communication between them
False.
it also means less heat
False.
which means a smaller heatsink, less noise and less electricity consumed.
False.
Not only are these assmptions not inherently coupled with the # die integrated in the package, but... separating the cores out to multiple die actually can reduce latencies and enable better connectivity due to stacking, and furthermore has a drastic impact on yield, test time, and ultimately cost.
That caught you by surprise? You mean you actually figured out you had something called a "video card", that it was actually inside the box you have sitting on your desk, AND, you have opened successfully up your case, AND you identified which card was the video card somehow, AND you are considering putting a new one in its place, AND you have gone through the trouble to determine what is lacking about your card and which card is the best upgrade for it? All that trouble, AND you didn't think to read the NAME of the video card to notice whether it said "PCI-Express" or "AGP"?
Man, that is one big coincidence... what are the odds of all that happening to the same unfortunate soul? Well, what the hell... blame it on Intel, right?
I know this sounds like a flame, but I ave your best interests at heart... You should totally not be opening up your PC, much less believing the anal drivel AMD resorts to. What, are there rules to dual-core? And AMD gets to set these rules? Even if this means their processor will cost twice as much and also return lower margins? Yeah, they might get a few percent performance on a few benchmarks, but at twice the cost, who wants that?
Dual-core simply means you have two tightly-coupled cores. Generally this implies they are in the same package at least, and there is low-bandwidth interconnect, some means for IPC, probably a shared chipset, and some amount of shared memory hierarchy, perhaps a shared L2 or L3 cache.
The reason most people think it is synonymous with dual-cores-on-the-same-die is simply because it has historically not been simple or cheap enough to integrate multiple die in the same package. Intel has the capability to do it, but most others cannot.
As your parent post mentioned, this gives Intel a significant advantage in price (due to yield and test time and complexity improvements) as well as potential package size (depending on whether the die are stacked or side-by-side). And there are heat ramifications b/c stacked results in less surface area and higher heat density. Aside from Intel's alreadly higher heat dissipation.
Simply because AMD does not have the manufacturing capability, they are trying to FUD Intel's advantage here. What a ridiculously cheap and desparate move -- LOL! Believe me, if AMD could integrate multiple die as cheaply, they would do it too.
This is one of the reasons why AMD dual-core costs about twice what an Intel one costs. Although AMD wll run benchmarks faster, all the consumer will see is "a dual-core Pentium 4, cheaper than a dual-core AMD... cool!"
From your source, the woman was seeking $20,000 to cover her medical bills, but McD only offered her a measly $800.
And, the woman received $160,000 in compensatory damages plus $480,000 (i.e. 3x compensatory) in punitive damages.
Here's my prediction. In 10 years, everything will run on x86. Yes, even PDAs and cell phones. The risk to Intel is that more competitors start building x86 processors (AMD is too small). Maybe Samsung or some other company with the capital requirements and semiconductor expertise necessary to build out fabs.
Good observation... Samsung could completely swallow up AMD, TMTA, Via Centaur, etc., and with them all their IP, patents, and other cross-licensing leverage, giving them an easy entry into the ludicrous world of x86. Combine this with the Korea's very "favorable" business conditions toward Samsung, Samsung's insane ability to down-pressure costs, complete silicon-to-product vertical solutions, existing memory and flash business, and they would be a force to reckon with.
As funny as that would be, Dell is not scared of this. In fact, if the deal went through (and I don't think it will - at least in the PC business) it would be in all three's interests to have Dell involved.
Apple SW + Intel CPU/chipset + Dell mobo and sales... the combined hype and marketing force alone would topple mountains, not to mention put a dent in Microsoft's ego and even profits.
Oh yeah, and throw a little Google in there somewhere too... desktop search or something... lol
That is incredibly difficult to believe for many reasons. Let me count the ways....
1) 99% of today's software runs on x86. Nobody wants to dump all of their software and migrate to a new ISA. Why else would CPU manufacturers continue to support legacy x86 even at the cost of up to 10% of their area and power budget just to decode x86 CISC instructions to RISC u-ops
2) There is an enormous business infrastructure built around this. It would all but KILL microsoft's reputation in the business world
3) The vast majority of Microsoft's revenue is on x86 software. It would be suicide for them to "theoretically begin to endorse" a new PC architecture that a) does not have the capacity and credibility to supply the world with enough PCs (ever wondered why Dell doesn't source from AMD???), b) does not have enough software to satisfy the demand, c) doesn't even have an owner stepping up to the plate (didn't IBM just sell off its PC business to Lenovo), d) can't provide the lowest prices
4) Nature abhores a vacuum. Any number of software vendors would love to get a crack at the x86 market that Microsoft vacated. Again, this would be suicide for Microsoft.
5) Intel has plenty of internal software, drivers, development tools, etc, not to mention an absolutely enormous amount of open-source win32 software and linux software
6) Every consumer service provider and hardware vendor in the world supports WIN32 on x86. Microsoft would be starting a platform from scratch that nobody would buy because the market does not sell anything for it: a) IO devices, b) broadband/VoIP/VPN, c) all the software and games they are used to
7) Even IBM would be a fool to think they could survive without x86 platforms to install their software and services onto
8) Not to mention that what you are saying is far from original -- the market has been saying for YEARS and YEARS that Intel is doomed due to a narrow focus. And yet just last quarter, Intel reported record revenue and profit. Their stock is taking off as investors expect great future growth. More than ever before, even more than during the dot.com hayday. The continue to beat down AMD to lower market share. They have pommelled TMTA and Via into oblivion. Please tell me how lack of diversity has been hurting them. Their margins are still in the 60% range, they went through the entire dot-bomb without posting a single quarterly loss (unlike ANY other large tech company I can think of), they are the first to 30cm wafer production, first to 65nm geometry in volume
9) And they do diversify. They have revolutionized the laptop platform, all but taken over the high-performance (i.e. high-margin) PDA and portable computing marketplace. They are a market leader in NOR flash (again beating out AMD to the point where they must sell their flash devision, not the same definition of "do very well" you must be thinking of). http://www.intel.com/pressroom/archive/releases/20 050302net_a.htm
http://news.com.com/AMD+expects+flash+memory+to+hu rt+revenue/2100-1006_3-5521587.html
They are pushing WiMAX to the market as a viable competitor to both cellular technology and cable/dsl broadband, and they are the first to bring wimax silicon to the market http://www.intel.com/ca/pressroom/2005/0418.htm They have a single-chip cellular GPRS baseband and high-performance application processor for entering the phone/PDA market. http://www.intel.com/design/pca/prodbref/252336.ht m
10) Time and time again, Intel has proven its marketing and execution genious, bringing to market products that are not necessarily the most academically superior, but certainly
AMD made a great presentation for WinHEC ex