Haha... a couple of new fledgeling companies think they can surpass IBM. These companies will have to catch up to IBM first. When have these companies designed a desktop CPU. What's with Taiwan these days, they think they're the next Japan. See where Japan is now... that will be Taiwan in a few years.
Get back to us when they have developed a CPU like AMD, Intel, or IBMs current CPUs.
By the way, first there is nothing wrong with American education, and second IBM is an international company and has more than American employees.
yeah so what happened to IBM's 90nm leadership?
these companies have no knowledge of computer architecture. they design semiconductor processes. they are foundries. they have more knowledge about how to make chips. IBM is losing.
by the way, American education is very wrong. American "college level" multivariable calculus, linear algebra, discrete mathematics, even FET, CMOS semiconductor device physics is learned in high school or even junior high in other countries.
Fabbing huge CPU like the PPC 970, the P4 or the Athlon 64 is the most difficult thing to fab.
One of the more easy one is to fab memory.
IBM is really competitive for fabbing and he cheers his indepedance. It's better to not be entirely dependant of an other companie. I hope that IBM will continue to fab his owns chips.
UMC and TSMC have the capability to fab complex chips, its just that the customers aren't there yet. these are really advanced 9 layer 90nm processes and they are pretty mature by now
UMC and TSMC have the capability to fab complex chips, its just that the customers aren't there yet. these are really advanced 9 layer 90nm processes and they are pretty mature by now
Having the capability to fab complex chips is one thing, making them is an another.
And for the level of US teaching : if it was that bad, how do you explain the number of nobel prize in US. You do not need a tons of people knowing CMOS physics, you need some very skilled peoples.
Grid/Cell systems turn system performance into a commodity.
The IBM grid computing stuff is targeted at enterprise type computer systems and the Sony Cell stuff is targeted at the opposite end of the computing spectrum with consumer media devices like the PS3. Apple's products cover most of this range, with the majority being near the Sony PS3.
UMC and TSMC have the capability to fab complex chips, its just that the customers aren't there yet. these are really advanced 9 layer 90nm processes and they are pretty mature by now
Yeah that's why nVidia is moving their production to IBM's Fishkill plant. Joker.
these companies have no knowledge of computer architecture. they design semiconductor processes. they are foundries. they have more knowledge about how to make chips. IBM is losing.
by the way, American education is very wrong. American "college level" multivariable calculus, linear algebra, discrete mathematics, even FET, CMOS semiconductor device physics is learned in high school or even junior high in other countries.
We're not privvy to all of the details of what IBM's fabs currently are and are not capable of. There have been indications that they already have some 90nm parts in production, just not 50+ million transistor CPUs. Processors are apparently only about 6 months away. For the future they are already researching 65nm (and other), and the results of that cannot be judged until it arrives. The players in this market continually jostle for "the lead", you can't declare a winner or loser based on currently available information (except for Motorola, that is).
A single company isn't limited to having one body of knowledge. IBM's processor design knowledge and expertise does not negatively impact its fab technology -- if anything it probably enhances it. At the very least IBM is trying to leverage the combination to make them more attractive to fabless companies because they can offer IP and design aid to support their clients.
I'm not interested in a debate on American education, especially not here. I won't disagree with you that the US system has big weaknesses, although I don't think your metric for what makes a strong system is the right metric. That is irrelevant, however. IBM has been an international company for a long time and hires people from all backgrounds and all parts of the world. If the talent they need isn't local, they'll import it.
IBM has been an international company for a long time and hires people from all backgrounds and all parts of the world. If the talent they need isn't local, they'll import it.
The problem is that IBM tends to hire from top research universities within the United States, and those engineers are of lower quality than EE and MS engineers from National Taiwan University for example.
I beleaves that this tech is being used in IBM's plants for tele-comm chips only. So I wouldn't expect to seed this process in a desktop chip anytime soon.
Heh.
But it looks like IBM will be using in SSOI in the 90nm G5. This is the advantage IBM has. They have such strong research they can leverage all kinds of technologies including gerManium substrates in fabricating their chips. If TSMC and UMC's designs were all that "mature" they'd have clients now. Since they don't, it's likely that they will be forced to license IBM's 90nm tech (which is mature) to build their customer's chips because that's what their customers will demand.
The problem is that IBM tends to hire from top research universities within the United States, and those engineers are of lower quality than EE and MS engineers from National Taiwan University for example.
It's nice to be proud of your countrie, and Taiwan has a great economy. However i don't see the necessity to start a flamewar about the level of US engineers.
But it looks like IBM will be using in SSOI in the 90nm G5. This is the advantage IBM has. They have such strong research they can leverage all kinds of technologies including gerManium substrates in fabricating their chips. If TSMC and UMC's designs were all that "mature" they'd have clients now. Since they don't, it's likely that they will be forced to license IBM's 90nm tech (which is mature) to build their customer's chips because that's what their customers will demand.
SSOI is too expensive and does not offer that much advantage in performance.
UMC has been shipping 90nm products for customers like Xilinx since March. IBM's product has been plagued with problems and low yield.
The problem is that IBM tends to hire from top research universities within the United States, and those engineers are of lower quality than EE and MS engineers from National Taiwan University for example.
The reason they're the top universities is because they are very good. I can tell you from experience though IBM is definitely not limited to universities in the US and actively headhunt the best talent in any country. My experience with American engineers has never been that they are of dramatically lower expertise, in fact I've met a few that I consider at the top of their fields, just that most of them are crap to work in teams with.
Quote:
Originally posted by Nr9
UMC has been shipping 90nm products for customers like Xilinx since March. IBM's product has been plagued with problems and low yield.
This is engineering and whenever you do production work there are issues. It doesn't matter how complicated or simple the process is stuff goes wrong and often in some damn unexpected ways. UMC certainly isn't perfect though. UMC had substantial troubles with their 130 nm SOI fabbing for AMD and is not producing anything of the difficulty level as a modern microprocessor yet on their 90 nm processes. There is a good reason AMD abandoned UMC for a partnership with IBM for future fabbing technologies as well.
I know for a fact IBM is shipping less complex 90 nm products that they're producing on contract. Ironically one of the companies buying them is Xilinx.
The reason they're the top universities is because they are very good. I can tell you from experience though IBM is definitely not limited to universities in the US and actively headhunt the best talent in any country. My experience with American engineers has never been that they are of dramatically lower expertise, in fact I've met a few that I consider at the top of their fields, just that most of them are crap to work in teams with.
This is engineering and whenever you do production work there are issues. It doesn't matter how complicated or simple the process is stuff goes wrong and often in some damn unexpected ways. UMC certainly isn't perfect though. UMC had substantial troubles with their 130 nm SOI fabbing for AMD and is not producing anything of the difficulty level as a modern microprocessor yet on their 90 nm processes. There is a good reason AMD abandoned UMC for a partnership with IBM for future fabbing technologies as well.
I know for a fact IBM is shipping less complex 90 nm products that they're producing on contract. Ironically one of the companies buying them is Xilinx.
Look at the above linked article. UMC and TSMC are having 20% better yields than IBM.
Look at the above linked article. UMC and TSMC are having 20% better yields than IBM.
Heh. Is that what " 20% ?more good die? on conventional 200mm wafers" means? Or does it mean their 200mm yields are 20% higher than IBM's yields on 300mm wafers? Which could still leave IBM with higher quantities.
And what size dies are they talking about? Are TSMC and UMC producing the same size dies on their 200mm wafers as IBM is on their 300mm wafers?
Heh. Is that what " 20% ?more good die? on conventional 200mm wafers" means? Or does it mean their 200mm yields are 20% higher than IBM's yields on 300mm wafers? Which could still leave IBM with higher quantities.
And what size dies are they talking about? Are TSMC and UMC producing the same size dies on their 200mm wafers as IBM is on their 300mm wafers?
Me thinks I smell spin.
it is a widely acknowledge fact that fishkil has problems. go do a search on it. IBM executives even admit it. why are you so defensive of IBM?
it is a widely acknowledge fact that fishkil has problems.
No, it's a widely acknowledged fact that the East Fishkill had problems starting up. That was last quarter.
Quote:
go do a search on it. IBM executives even admit it. why are you so defensive of IBM?
I'm not defensive. I just don't care for trollish behavior. And making claims like the above, dissing the quality of American educational institutions and engineers is trolling.
And another thing. High school students in your country may well take higher level courses than our high school students do. But you know what? Every American is entitled to a high school education. Every single one. Unlike some countries that only allow kids who pass tough entrance exams to attend or only allow the privileged to be educated. Sure, some kids take high level courses, but the majority may never get beyond trade schools, or the equivalent of a Jr. High level.
Look at the above linked article. UMC and TSMC are having 20% better yields than IBM.
Eh. I'm guessing you aren't actually an engineer or else you'd realise such is engineering. Often you're presented with multiple routes to take and sometimes what looks good sometimes screws up.
Given IBM pioneered several of the most important, and difficult, technologies and processes currently used in semiconductor manufacturing and managed to bring them successfully online years ahead of competition to claim they don't know what they're doing is naïve.
Unlike UMC and TSMC IBM only charges by the working chip rather than by the wafer so the poor yields hurt their bottom line more, except in terms of production capacity and nobody has made any great secret that the Fishkill plant has had difficulties ramping up.
Education does not in any way have anything to do with ingenuity and creativity. There are plenty of people with little education that create amazing products and processes.
I know plenty of well educated scientists that couldn't solder two wires together or build a shack to live in, let alone change a tire on their car. They work real well sitting in an office working on one specific problem their entire lifetime. In many cases they don't even know where the problem they are working on fits in the scheme of things.
...and trolling is an interesting form of science and a waste of an education...
Comments
Originally posted by Leonard
Haha... a couple of new fledgeling companies think they can surpass IBM. These companies will have to catch up to IBM first. When have these companies designed a desktop CPU. What's with Taiwan these days, they think they're the next Japan. See where Japan is now... that will be Taiwan in a few years.
Get back to us when they have developed a CPU like AMD, Intel, or IBMs current CPUs.
By the way, first there is nothing wrong with American education, and second IBM is an international company and has more than American employees.
yeah so what happened to IBM's 90nm leadership?
these companies have no knowledge of computer architecture. they design semiconductor processes. they are foundries. they have more knowledge about how to make chips. IBM is losing.
by the way, American education is very wrong. American "college level" multivariable calculus, linear algebra, discrete mathematics, even FET, CMOS semiconductor device physics is learned in high school or even junior high in other countries.
Originally posted by Powerdoc
Fabbing huge CPU like the PPC 970, the P4 or the Athlon 64 is the most difficult thing to fab.
One of the more easy one is to fab memory.
IBM is really competitive for fabbing and he cheers his indepedance. It's better to not be entirely dependant of an other companie. I hope that IBM will continue to fab his owns chips.
UMC and TSMC have the capability to fab complex chips, its just that the customers aren't there yet. these are really advanced 9 layer 90nm processes and they are pretty mature by now
Originally posted by Nr9
UMC and TSMC have the capability to fab complex chips, its just that the customers aren't there yet. these are really advanced 9 layer 90nm processes and they are pretty mature by now
Having the capability to fab complex chips is one thing, making them is an another.
And for the level of US teaching : if it was that bad, how do you explain the number of nobel prize in US. You do not need a tons of people knowing CMOS physics, you need some very skilled peoples.
http://www-106.ibm.com/developerworks/grid/
http://www-3.ibm.com/chips/news/2001...y-toshiba.html
Grid/Cell systems turn system performance into a commodity.
The IBM grid computing stuff is targeted at enterprise type computer systems and the Sony Cell stuff is targeted at the opposite end of the computing spectrum with consumer media devices like the PS3. Apple's products cover most of this range, with the majority being near the Sony PS3.
Originally posted by Nr9
UMC and TSMC have the capability to fab complex chips, its just that the customers aren't there yet. these are really advanced 9 layer 90nm processes and they are pretty mature by now
Yeah that's why nVidia is moving their production to IBM's Fishkill plant. Joker.
Originally posted by Nr9
yeah so what happened to IBM's 90nm leadership?
these companies have no knowledge of computer architecture. they design semiconductor processes. they are foundries. they have more knowledge about how to make chips. IBM is losing.
by the way, American education is very wrong. American "college level" multivariable calculus, linear algebra, discrete mathematics, even FET, CMOS semiconductor device physics is learned in high school or even junior high in other countries.
We're not privvy to all of the details of what IBM's fabs currently are and are not capable of. There have been indications that they already have some 90nm parts in production, just not 50+ million transistor CPUs. Processors are apparently only about 6 months away. For the future they are already researching 65nm (and other), and the results of that cannot be judged until it arrives. The players in this market continually jostle for "the lead", you can't declare a winner or loser based on currently available information (except for Motorola, that is).
A single company isn't limited to having one body of knowledge. IBM's processor design knowledge and expertise does not negatively impact its fab technology -- if anything it probably enhances it. At the very least IBM is trying to leverage the combination to make them more attractive to fabless companies because they can offer IP and design aid to support their clients.
I'm not interested in a debate on American education, especially not here. I won't disagree with you that the US system has big weaknesses, although I don't think your metric for what makes a strong system is the right metric. That is irrelevant, however. IBM has been an international company for a long time and hires people from all backgrounds and all parts of the world. If the talent they need isn't local, they'll import it.
Originally posted by stingerman
Yeah that's why nVidia is moving their production to IBM's Fishkill plant. Joker.
http://www.xbitlabs.com/news/video/d...108042430.html
Originally posted by Programmer
IBM has been an international company for a long time and hires people from all backgrounds and all parts of the world. If the talent they need isn't local, they'll import it.
The problem is that IBM tends to hire from top research universities within the United States, and those engineers are of lower quality than EE and MS engineers from National Taiwan University for example.
Originally posted by Nr9
http://www.xbitlabs.com/news/video/d...108042430.html
Has nVidia confirmed this report?
Originally posted by stingerman
Has nVidia confirmed this report?
both digitimes and xbitlabs are pretty damned reliable.
Originally posted by 709
I beleaves that this tech is being used in IBM's plants for tele-comm chips only. So I wouldn't expect to seed this process in a desktop chip anytime soon.
Heh.
But it looks like IBM will be using in SSOI in the 90nm G5. This is the advantage IBM has. They have such strong research they can leverage all kinds of technologies including gerManium substrates in fabricating their chips. If TSMC and UMC's designs were all that "mature" they'd have clients now. Since they don't, it's likely that they will be forced to license IBM's 90nm tech (which is mature) to build their customer's chips because that's what their customers will demand.
Originally posted by Nr9
The problem is that IBM tends to hire from top research universities within the United States, and those engineers are of lower quality than EE and MS engineers from National Taiwan University for example.
It's nice to be proud of your countrie, and Taiwan has a great economy. However i don't see the necessity to start a flamewar about the level of US engineers.
Do I see some jingoims here ?
Originally posted by Tomb of the Unknown
Heh.
But it looks like IBM will be using in SSOI in the 90nm G5. This is the advantage IBM has. They have such strong research they can leverage all kinds of technologies including gerManium substrates in fabricating their chips. If TSMC and UMC's designs were all that "mature" they'd have clients now. Since they don't, it's likely that they will be forced to license IBM's 90nm tech (which is mature) to build their customer's chips because that's what their customers will demand.
SSOI is too expensive and does not offer that much advantage in performance.
UMC has been shipping 90nm products for customers like Xilinx since March. IBM's product has been plagued with problems and low yield.
Originally posted by Nr9
The problem is that IBM tends to hire from top research universities within the United States, and those engineers are of lower quality than EE and MS engineers from National Taiwan University for example.
The reason they're the top universities is because they are very good. I can tell you from experience though IBM is definitely not limited to universities in the US and actively headhunt the best talent in any country. My experience with American engineers has never been that they are of dramatically lower expertise, in fact I've met a few that I consider at the top of their fields, just that most of them are crap to work in teams with.
Originally posted by Nr9
UMC has been shipping 90nm products for customers like Xilinx since March. IBM's product has been plagued with problems and low yield.
This is engineering and whenever you do production work there are issues. It doesn't matter how complicated or simple the process is stuff goes wrong and often in some damn unexpected ways. UMC certainly isn't perfect though. UMC had substantial troubles with their 130 nm SOI fabbing for AMD and is not producing anything of the difficulty level as a modern microprocessor yet on their 90 nm processes. There is a good reason AMD abandoned UMC for a partnership with IBM for future fabbing technologies as well.
I know for a fact IBM is shipping less complex 90 nm products that they're producing on contract. Ironically one of the companies buying them is Xilinx.
Originally posted by Telomar
The reason they're the top universities is because they are very good. I can tell you from experience though IBM is definitely not limited to universities in the US and actively headhunt the best talent in any country. My experience with American engineers has never been that they are of dramatically lower expertise, in fact I've met a few that I consider at the top of their fields, just that most of them are crap to work in teams with.
This is engineering and whenever you do production work there are issues. It doesn't matter how complicated or simple the process is stuff goes wrong and often in some damn unexpected ways. UMC certainly isn't perfect though. UMC had substantial troubles with their 130 nm SOI fabbing for AMD and is not producing anything of the difficulty level as a modern microprocessor yet on their 90 nm processes. There is a good reason AMD abandoned UMC for a partnership with IBM for future fabbing technologies as well.
I know for a fact IBM is shipping less complex 90 nm products that they're producing on contract. Ironically one of the companies buying them is Xilinx.
Look at the above linked article. UMC and TSMC are having 20% better yields than IBM.
Originally posted by Nr9
Look at the above linked article. UMC and TSMC are having 20% better yields than IBM.
Heh. Is that what " 20% ?more good die? on conventional 200mm wafers" means? Or does it mean their 200mm yields are 20% higher than IBM's yields on 300mm wafers? Which could still leave IBM with higher quantities.
And what size dies are they talking about? Are TSMC and UMC producing the same size dies on their 200mm wafers as IBM is on their 300mm wafers?
Me thinks I smell spin.
Originally posted by Tomb of the Unknown
Heh. Is that what " 20% ?more good die? on conventional 200mm wafers" means? Or does it mean their 200mm yields are 20% higher than IBM's yields on 300mm wafers? Which could still leave IBM with higher quantities.
And what size dies are they talking about? Are TSMC and UMC producing the same size dies on their 200mm wafers as IBM is on their 300mm wafers?
Me thinks I smell spin.
it is a widely acknowledge fact that fishkil has problems. go do a search on it. IBM executives even admit it. why are you so defensive of IBM?
Originally posted by Nr9
it is a widely acknowledge fact that fishkil has problems.
No, it's a widely acknowledged fact that the East Fishkill had problems starting up. That was last quarter.
go do a search on it. IBM executives even admit it. why are you so defensive of IBM?
I'm not defensive. I just don't care for trollish behavior. And making claims like the above, dissing the quality of American educational institutions and engineers is trolling.
And another thing. High school students in your country may well take higher level courses than our high school students do. But you know what? Every American is entitled to a high school education. Every single one. Unlike some countries that only allow kids who pass tough entrance exams to attend or only allow the privileged to be educated. Sure, some kids take high level courses, but the majority may never get beyond trade schools, or the equivalent of a Jr. High level.
So take your superior attitude and shove it.
Originally posted by Nr9
Look at the above linked article. UMC and TSMC are having 20% better yields than IBM.
Eh. I'm guessing you aren't actually an engineer or else you'd realise such is engineering. Often you're presented with multiple routes to take and sometimes what looks good sometimes screws up.
Given IBM pioneered several of the most important, and difficult, technologies and processes currently used in semiconductor manufacturing and managed to bring them successfully online years ahead of competition to claim they don't know what they're doing is naïve.
Unlike UMC and TSMC IBM only charges by the working chip rather than by the wafer so the poor yields hurt their bottom line more, except in terms of production capacity and nobody has made any great secret that the Fishkill plant has had difficulties ramping up.
I know plenty of well educated scientists that couldn't solder two wires together or build a shack to live in, let alone change a tire on their car. They work real well sitting in an office working on one specific problem their entire lifetime. In many cases they don't even know where the problem they are working on fits in the scheme of things.
...and trolling is an interesting form of science and a waste of an education...