look here, im not trying to argue that single threaded performance isn't important or that scaling to higher frequencies does not make sense.
Of course people want fast single threaded performance, however everyone has just hit the wall. People have tried for over a year to fix issues with 90nm processes. They have given up.
What i am saying is based on INSIDE INFORMATION that the companies involved have GIVEN UP on faster single thread performance.
im not a pessimist, of course I would like fast single thread performance too. But facts are facts, the power wall is coming down on everyone.
Mark my words, there will not be a 5GHz processor for at least the next five years and there probably will never be.
look here, im not trying to argue that single threaded performance isn't important or that scaling to higher frequencies does not make sense.
Of course people want fast single threaded performance, however everyone has just hit the wall. People have tried for over a year to fix issues with 90nm processes. They have given up.
What i am saying is based on INSIDE INFORMATION that the companies involved have GIVEN UP on faster single thread performance.
im not a pessimist, of course I would like fast single thread performance too. But facts are facts, the power wall is coming down on everyone.
Mark my words, there will not be a 5GHz processor for at least the next five years and there probably will never be.
What inside info? You haven't given us any yet. CPU cores will get faster. Do you work for Intel, AMD, IBM, and Moto/freescale? No you don't. Are you the President Bush? No. You have nothing and offically calling you a troll.
I remember back in the day, CPU development stalled at 200mhz for quite some time. Then it was the 450-500mhz wall. Then I stopped caring for a while... and now it's stalled at 2.5ghz. In the past, they managed to overcome their boundaries by refining their designs and development methods. Why would it be any different this time? I'm confident that the 2.5 mark will be broken, it'll just take some time to do, just like in the past.
Am I the only one that thinks that this supposed speed "wall" is good for the IT business ? I mean, now that the whole IT business has come to a halt regarding speed improvement, it puts the pressure back on software developers to accually optimise their applications, and not go by the old "Nah, no need for optimising, there'll be more powerful computers out soon enough anyways".
I mean look at the console games. One hardware specification that stays the same for 3+ years. But still the game developers pushes more and more features into their games, simply by optimising their software to run optimaly on existing hardware...
Maybe its just me, but I think this is a good thing
You don't have to be a prophet to see that the future is multi-core designs. Current technologies, roughly speaking, don't allow any more clock increase because they have approached their physical limits. This does not exclude the possibility that new materials will appear as soon as tomorrow which will help reduce power dissipation and allow lower voltages for gates operation. If and when these new techs come, brutal clock speed will rise again, up to the physical limits of these techs, and so on, and so forth.
Right now the only ways to substantially increase the computing power are multi-chip and multi-core chip designs. This is very good for the future because multi-core hardware will at last force software developers to write better multi-threaded apps, and this experience will never become obsolete, even when newer technologies push clock speed further.
Until that, it's time to optimize your mentality, be you a hardware or software developer, or an end-user with an inefficient workflow.
there was never a industry-wide consensus like this time that scaling has stopped
Well.. Freescale doesn't agree with you at all. They are currently at 1420 MHz with 7447A bur according to very current SNDF report (page 53) they expect to scale to 3+ GHz at 65 nm. That's more than double the current frequenzy. Even if they are trying to catch up with the rest of the industry they don't share your pessimism at all.
Actually, IBM are planning to introduce Power 6 based chips at up to 6GHz by 2006/7 according to the best and most accurate source of rumors on this front - the same guy who revealed the existence of 970Mp. The project is known as eCLipz
Mark my words, there will not be a 5GHz processor for at least the next five years and there probably will never be.
Never a 5 gig processor?
I agree with much of what you're saying, that the industry is moving from clock rate mongering over to more parallel designs. However, claiming 'never' makes me discount your posts as merely, accidently insightful.
they go up to 130 celsius with aggressive water cooling
they just melt down. your processor life is probably measured in days.
it was the same problem with the canned motorola G5s
Just curious, those 3GHz chips that couldn't run at all, were they made using an aggressive low-k dialectric, Strained Silicon and/or Strained Silicon directly on Insulator?
You don't have to be a prophet to see that the future is multi-core designs. Current technologies, roughly speaking, don't allow any more clock increase because they have approached their physical limits. This does not exclude the possibility that new materials will appear as soon as tomorrow which will help reduce power dissipation and allow lower voltages for gates operation. If and when these new techs come, brutal clock speed will rise again, up to the physical limits of these techs, and so on, and so forth.
Right now the only ways to substantially increase the computing power are multi-chip and multi-core chip designs. This is very good for the future because multi-core hardware will at last force software developers to write better multi-threaded apps, and this experience will never become obsolete, even when newer technologies push clock speed further.
Until that, it's time to optimize your mentality, be you a hardware or software developer, or an end-user with an inefficient workflow.
This is a very funny thread to read. Nr9 comes in with a serious piece of information that we really ought to important ramifications of, and he gets slapped down because people don't want to hear it. The evidence in support of this information is there, but almost nobody will acknowledge it. Instead he is called a troll. This reminds me of various other transitions in history where people refused to believe something was going to happen until they were steamrollered by it. Can't you even accept the possibility that he has a legitimate piece of information?
First of all, this is not a decree that no line of chips will every gain even a single MHz of speed from this day forth for all time. Different lineups are at different places on their performance/power curves, they have different requirements, they are on different processes, and they have different characteristics. What Nr9 is saying is that the companies who have reached the 90nm node have discovered, to their surprise (!!), that the power cost of frequency increases is no longer viable as a primary strategy to increasing performance. This trend has been obvious to anyone paying attention over the past 5 years, so I don't understand why it should be a surprise to anyone. Its also not a hard-and-fast number like the speed of sound, or the speed of light... but clearly the balance of tradeoffs in processor design has shifted far enough to effect the basic strategy of the designers. Intel announced this 6+ months ago when they decided that their future was not their clock rate champ, the NETBurst architecture.
Second, the various counter-examples mentioned are from companies that "expect" to be able to do better or from old roadmaps that don't take into account the learnings of the last 6-9 months. These are projections made without the foreknowledge of the problems found at 90nm. Freescale saying that they can reach 3 GHz, for example, is either them not being aware of the precise troubles they're about to hit... or it is them deciding that they can make the same design choices that IBM & Intel made to achieve 3 GHz.
Third, the engineering planning for this technology has very long lead times. If all the bleeding edge technology in existance doesn't provide any hope for correcting this problem, then any truly new ideas are probably about 5 years from reaching the maturity level where they can be used to produce millions and billions of new chips. I always take "never" with a grain of salt, but in the absence of any real ideas about how to break a given barrier then we ought to assume that for any practical purpose that barrier is as inevitable as the speed of light. Let the researchers hunt for a solution, that is the nature of research.
In any case, give Nr9 a fair shake... he hasn't done anything to deserve being slammed like this.
Programmer, I agree with much of what you and Nr9 are saying. Chip manufacturers are now focusing more on parallel designs than increasing clock frequencies.
However, it is also perfectly understandable that a terse, me-too prophet is met with skepticism on these boards. Perhaps his message would have gone over better if not for troll-like one-line posts like the following:
Quote:
plans change
Quote:
i make them
Quote:
EH, NO
Quote:
the liquid colling is definately required for the 2.5
Quote:
trust me, it is 2.5-4.0 GHz
Certainly not the best way to portray yourself as an insightful and knowledgable person instead of a troll.
Perhaps it was the other one line post that made you fall for his lovable persona:
Can't you even accept the possibility that he has a legitimate piece of information?
I see no reason why we should trust him over plenty of contradictory information. His track record isn't great. It's convenient to blame mispredictions of the past on unverifyable change of plans. His claims might be true, but we have absolutely no way of knowing.
We all know that the MHz race is nearing its end but Nr9 is firmly stating that it's over and that everyone in the industry agrees with that.
He obviously knows his stuff but his way of conveying them leavs no room for second oppinions or alternate views. He speaks like he IS the industry and that he knows everything. That attitude isn't to flattering. His last prediction turned out wrong, so why really should we trust his firm assurances now?
He comes to us from nowhere, demands of us to accept his gospel. He drops us stuff that we can debate in perpetuum and without any means of verifying his claims. He's very selective in what discussions he enters, and then stays there to fuel the debate with more unverifuable claims. He shows every sign of beeing a troll.
That said.. he might be dead right. I have no way of knowing.
People around here aren't even willing to contemplate a theoretical exercise. The past is the future. It all reminds me of Intel when Steve Jobs tried to sell them on the idea of a Personal Computer.
Maybe he's an insider. Maybe he's a troll. There's more than a grain of truth in what he says. The way forward is not the well worn path we've been on for years.
Just to add to what dfiler said. Nr9 also posted this highly sceptical information.
Quote:
Nr9
Member
From: Old Post posted 11-16-2003 06:08 PM
The PowerBook G5 has 2 MCMs with 2 processor each
4 processor PowerBook G5 with modified 440 core with altivec
Last edited by Nr9 on 11-16-2003 at 06:16 PM
While in and of itself, using 4 440 cores maybe possible, it would require major modifications to OS X and all applications to make it feasible to run software at any reasonable speeds.
In this case, not only does Nr9 make the chips, he also has a working knowledge of software vendors development plans.
While I don't discount the current wall being hit, there is no way to predict the ultimate outcome. As yet the G5 does not use an aggressive low-k dialectric and does not use SSDOI. Breakthroughs do happen, don't they? Making unqualified statements about an unbreakable wall being 2-4 GHz reduces credibility.
People around here aren't even willing to contemplate a theoretical exercise....
Sure they are.
But such exercises aren't fostered by supposed insiders posting one liners like:
Quote:
trust me, it is 2.5-4.0 GHz
Posts like these are indistinguishable from trolling.
However, when knowledgeable people post properly reasoned and supported assertions, they are usually greeted by a much different reaction.
I greatly prefer Programmer'S approach to Nr9's. You are held in high esteem by me and many others Programmer, and we appreciate your contributions to this forum. Hopefully others follow your lead.
Comments
Of course people want fast single threaded performance, however everyone has just hit the wall. People have tried for over a year to fix issues with 90nm processes. They have given up.
What i am saying is based on INSIDE INFORMATION that the companies involved have GIVEN UP on faster single thread performance.
im not a pessimist, of course I would like fast single thread performance too. But facts are facts, the power wall is coming down on everyone.
Mark my words, there will not be a 5GHz processor for at least the next five years and there probably will never be.
Originally posted by Nr9
look here, im not trying to argue that single threaded performance isn't important or that scaling to higher frequencies does not make sense.
Of course people want fast single threaded performance, however everyone has just hit the wall. People have tried for over a year to fix issues with 90nm processes. They have given up.
What i am saying is based on INSIDE INFORMATION that the companies involved have GIVEN UP on faster single thread performance.
im not a pessimist, of course I would like fast single thread performance too. But facts are facts, the power wall is coming down on everyone.
Mark my words, there will not be a 5GHz processor for at least the next five years and there probably will never be.
What inside info? You haven't given us any yet. CPU cores will get faster. Do you work for Intel, AMD, IBM, and Moto/freescale? No you don't. Are you the President Bush? No. You have nothing and offically calling you a troll.
even though you don't even have to work to find out
just join ACM SIGARCH
im sure you will learn a lot
Guys don't feed the trolls...
I mean look at the console games. One hardware specification that stays the same for 3+ years. But still the game developers pushes more and more features into their games, simply by optimising their software to run optimaly on existing hardware...
Maybe its just me, but I think this is a good thing
.:BoeManE:.
Right now the only ways to substantially increase the computing power are multi-chip and multi-core chip designs. This is very good for the future because multi-core hardware will at last force software developers to write better multi-threaded apps, and this experience will never become obsolete, even when newer technologies push clock speed further.
Until that, it's time to optimize your mentality, be you a hardware or software developer, or an end-user with an inefficient workflow.
Originally posted by Nr9
there was never a industry-wide consensus like this time that scaling has stopped
Well.. Freescale doesn't agree with you at all. They are currently at 1420 MHz with 7447A bur according to very current SNDF report (page 53) they expect to scale to 3+ GHz at 65 nm. That's more than double the current frequenzy. Even if they are trying to catch up with the rest of the industry they don't share your pessimism at all.
Originally posted by Nr9
Mark my words, there will not be a 5GHz processor for at least the next five years and there probably will never be.
Never a 5 gig processor?
I agree with much of what you're saying, that the industry is moving from clock rate mongering over to more parallel designs. However, claiming 'never' makes me discount your posts as merely, accidently insightful.
Originally posted by Nr9
the 3GHz chips can't run at all
they go up to 130 celsius with aggressive water cooling
they just melt down. your processor life is probably measured in days.
it was the same problem with the canned motorola G5s
Just curious, those 3GHz chips that couldn't run at all, were they made using an aggressive low-k dialectric, Strained Silicon and/or Strained Silicon directly on Insulator?
Originally posted by costique
You don't have to be a prophet to see that the future is multi-core designs. Current technologies, roughly speaking, don't allow any more clock increase because they have approached their physical limits. This does not exclude the possibility that new materials will appear as soon as tomorrow which will help reduce power dissipation and allow lower voltages for gates operation. If and when these new techs come, brutal clock speed will rise again, up to the physical limits of these techs, and so on, and so forth.
Right now the only ways to substantially increase the computing power are multi-chip and multi-core chip designs. This is very good for the future because multi-core hardware will at last force software developers to write better multi-threaded apps, and this experience will never become obsolete, even when newer technologies push clock speed further.
Until that, it's time to optimize your mentality, be you a hardware or software developer, or an end-user with an inefficient workflow.
That's what I was trying to go for with my posts
I just suck at saying it.
First of all, this is not a decree that no line of chips will every gain even a single MHz of speed from this day forth for all time. Different lineups are at different places on their performance/power curves, they have different requirements, they are on different processes, and they have different characteristics. What Nr9 is saying is that the companies who have reached the 90nm node have discovered, to their surprise (!!), that the power cost of frequency increases is no longer viable as a primary strategy to increasing performance. This trend has been obvious to anyone paying attention over the past 5 years, so I don't understand why it should be a surprise to anyone. Its also not a hard-and-fast number like the speed of sound, or the speed of light... but clearly the balance of tradeoffs in processor design has shifted far enough to effect the basic strategy of the designers. Intel announced this 6+ months ago when they decided that their future was not their clock rate champ, the NETBurst architecture.
Second, the various counter-examples mentioned are from companies that "expect" to be able to do better or from old roadmaps that don't take into account the learnings of the last 6-9 months. These are projections made without the foreknowledge of the problems found at 90nm. Freescale saying that they can reach 3 GHz, for example, is either them not being aware of the precise troubles they're about to hit... or it is them deciding that they can make the same design choices that IBM & Intel made to achieve 3 GHz.
Third, the engineering planning for this technology has very long lead times. If all the bleeding edge technology in existance doesn't provide any hope for correcting this problem, then any truly new ideas are probably about 5 years from reaching the maturity level where they can be used to produce millions and billions of new chips. I always take "never" with a grain of salt, but in the absence of any real ideas about how to break a given barrier then we ought to assume that for any practical purpose that barrier is as inevitable as the speed of light. Let the researchers hunt for a solution, that is the nature of research.
In any case, give Nr9 a fair shake... he hasn't done anything to deserve being slammed like this.
However, it is also perfectly understandable that a terse, me-too prophet is met with skepticism on these boards. Perhaps his message would have gone over better if not for troll-like one-line posts like the following:
plans change
i make them
EH, NO
the liquid colling is definately required for the 2.5
trust me, it is 2.5-4.0 GHz
Certainly not the best way to portray yourself as an insightful and knowledgable person instead of a troll.
Perhaps it was the other one line post that made you fall for his lovable persona:
programmer is correct
Originally posted by Programmer
Can't you even accept the possibility that he has a legitimate piece of information?
I see no reason why we should trust him over plenty of contradictory information. His track record isn't great. It's convenient to blame mispredictions of the past on unverifyable change of plans. His claims might be true, but we have absolutely no way of knowing.
We all know that the MHz race is nearing its end but Nr9 is firmly stating that it's over and that everyone in the industry agrees with that.
He obviously knows his stuff but his way of conveying them leavs no room for second oppinions or alternate views. He speaks like he IS the industry and that he knows everything. That attitude isn't to flattering. His last prediction turned out wrong, so why really should we trust his firm assurances now?
He comes to us from nowhere, demands of us to accept his gospel. He drops us stuff that we can debate in perpetuum and without any means of verifying his claims. He's very selective in what discussions he enters, and then stays there to fuel the debate with more unverifuable claims. He shows every sign of beeing a troll.
That said.. he might be dead right. I have no way of knowing.
Maybe he's an insider. Maybe he's a troll. There's more than a grain of truth in what he says. The way forward is not the well worn path we've been on for years.
Nr9
Member
From: Old Post posted 11-16-2003 06:08 PM
The PowerBook G5 has 2 MCMs with 2 processor each
4 processor PowerBook G5 with modified 440 core with altivec
Last edited by Nr9 on 11-16-2003 at 06:16 PM
While in and of itself, using 4 440 cores maybe possible, it would require major modifications to OS X and all applications to make it feasible to run software at any reasonable speeds.
In this case, not only does Nr9 make the chips, he also has a working knowledge of software vendors development plans.
While I don't discount the current wall being hit, there is no way to predict the ultimate outcome. As yet the G5 does not use an aggressive low-k dialectric and does not use SSDOI. Breakthroughs do happen, don't they? Making unqualified statements about an unbreakable wall being 2-4 GHz reduces credibility.
Originally posted by murk
People around here aren't even willing to contemplate a theoretical exercise....
Sure they are.
But such exercises aren't fostered by supposed insiders posting one liners like:
trust me, it is 2.5-4.0 GHz
Posts like these are indistinguishable from trolling.
However, when knowledgeable people post properly reasoned and supported assertions, they are usually greeted by a much different reaction.
I greatly prefer Programmer'S approach to Nr9's. You are held in high esteem by me and many others Programmer, and we appreciate your contributions to this forum. Hopefully others follow your lead.