It is not a question of ignoring your arguments, the problem is that your arguments are invalid and apparently not based on contemporary knowledge in the field.
My arguments are invalid are they? Not based on contemporary knowledge are they?
Dark side blinded you to reason, it has. Argue with you I will not.
Quote:
The point is that you have been arguing that the described technology is impossible because a new instruction set is required or new hardware is required or a new operating system is required. Clusters in general and the PPC cluster in particular, should demonstrate clearly that it is possible to implement the discussed technology WITHOUT the need for instruction set modifications or special processors.
Aside from the fact that you don't seem to understand the difference between the impossible and the merely unreasonably difficult, this assertion proves that you have a fundamental lack of understanding of what HPC is about.
Quote:
So At least we agree that the nodes are capable of running conventional code while being part of a cluster.
I said they could. That's not the point. You're arguing at cross purposes here because you really don't know what you're talking about. (About clustering, that is. You may know a lot about other things you talk about.)
Quote:
I do know a couple of things though. One is that Apple will have hard time stuffing a 970 into a laptop even with a die shrink. The second is that Apple is probally in the best position of any company, in the computing business, to be able to successfully launch such a machine. This is largely a result of previous efforts to support the G4 in dual processor configurations.
You don't know as much as you think you do. And as long as you continue to believe you know as much as you think you do, you never will.
If the current rumors on AI are correct about the PPC970+, we may well see 1.8 - 2.0GHz G5 Powerbooks this summer. What with SSOI and Power Tune, this could the "low energy" chip rumored to be in the works.
Hey, I'm willing to be educated. Show me something. So far you all I've seen from you is the insistance that STI are really just building a cluster and Apple is building a cluster in a laptop because a chip you haven't seen yet is going to be bad at power management.
Now, if in my ignorance, I have some part of that wrong please enlighten me.
Hey, I'm willing to be educated. Show me something. So far you all I've seen from you is the insistance that STI are really just building a cluster and Apple is building a cluster in a laptop because a chip you haven't seen yet is going to be bad at power management.
Now, if in my ignorance, I have some part of that wrong please enlighten me.
something you should know, is that comments like
You don't know as much as you think you do. And as long as you continue to believe you know as much as you think you do, you never will.
in response to:
I do know a couple of things though. One is that Apple will have hard time stuffing a 970 into a laptop even with a die shrink. The second is that Apple is probally in the best position of any company, in the computing business, to be able to successfully launch such a machine. This is largely a result of previous efforts to support the G4 in dual processor configurations
are neither constructive nor accurate.
a rumor is a rumor. you dont dispel another rumor using another rumor as proof.
a rumor is a rumor. you dont dispel another rumor using another rumor as proof.
That's funny coming from someone who accused IBM of lying about technology in their next processor.
It's common sense. Which do you think is more likely to be in the next Powerbook G5, a slimmed down, power saving PPC970 or a 4 CPU cluster with as many transistors as a PPC970?
Oh, wait, forget I asked. I think I know what your answer will be and it will have nothing to do with common sense.
I said they could. That's not the point. You're arguing at cross purposes here because you really don't know what you're talking about. (About clustering, that is. You may know a lot about other things you talk about.)
My point with respect to clustering, is that a cluster can be built out of common hardware. These clusters can be built at the board level like VT does or similarly on a MCM. SOC just provides another scaling factor. New hardware instructions and massive rewrites to the operating system are not required.
Quote:
If the current rumors on AI are correct about the PPC970+, we may well see 1.8 - 2.0GHz G5 Powerbooks this summer. What with SSOI and Power Tune, this could the "low energy" chip rumored to be in the works.
I would absolutely love to hear in feburary that the 970+, along with its support chips, are running with low enough power disipation to go into a laptop. That would be just fantastic. On the other hand I'm not too sure that IBM is going to be able to cut power usage by 50%. Even if they hit 50% it would still be a rather power hungery machine relative to what else is possible.
I really hope nr9 is right. I'm an EE and I would love to have a powerbook with a processor that's actually a revolutionary development. Whether it's real or not, I think this rumor is damn cool.
Yeah the thought is really cool, it certianly would be good if Apple where to introduce really low power technology into a laptop. For a laptop in would be revolutionary, but in the past attempts that somewhat mimic what is being described where done on the PC.
I wonder if anybody remembers the Transputer? What has been described is something like a Transputer but with fewer porcessors of the same family as the control processor. With a bit of modernization it would be really cool. Especially if they can map some of the current operating system functionality onto the implementation.
Dave
Quote:
Originally posted by Splinemodel
I really hope nr9 is right. I'm an EE and I would love to have a powerbook with a processor that's actually a revolutionary development. Whether it's real or not, I think this rumor is damn cool.
I believe Tomb is right about the G5 PowerBook. IBM is doing an energy saving G5 and will be talking about it in February if my memory is correct. It's possible that this chip is not just for Apple. IBM may see a big benefit to significantly lower power for its blade servers too. If there is not excessive overhead, power conservation is a good thing for any application.
Yet the cell appears to have a great deal of promise. If Sony is pursuing the cell for PlayStation 3, it is surely way beyond a 'blue sky' idea. They have it working, at least in prototype stage. Since the PlayStation is a low cost consumer product, the cell must have potential for low cost as well as a high performance at low power. So it should not be surprising to hear that Apple is working on the same thing. But realistically, at what stage is Apple in producing a product?
My bet is that Apple will not beat Sony to market with cell technology. If this rumor is true about a four core processor using message passing, it is very likely a prototype chip for prototype computers. There would be a tremendous amount of software experimentation and testing to do on a new concept like this. Apple would have to get hundreds of these boxes in the hands of their own software engineers as well as a few key vendors.
It would be foolish to come out with such a revolutionary product quickly. When does Sony plan to introduce the PS/3? This would be the very earliest date that Apple could have a product. Likely it would be longer. If after a lot of experimentation and testing Apple finds cell technology does not offer some significant advantage, then it may never see the light of day.
That's funny coming from someone who accused IBM of lying about technology in their next processor.
It's common sense. Which do you think is more likely to be in the next Powerbook G5, a slimmed down, power saving PPC970 or a 4 CPU cluster with as many transistors as a PPC970?
Oh, wait, forget I asked. I think I know what your answer will be and it will have nothing to do with common sense.
i enjoy the luxury of making such comments because I have sources. you don't. you are just speculating
I just want to add a final thought to my last posting.
I believe there is a very good chance that the cell type processor will make it into Apple products eventually. However, the last statement in my posting could also be true. After testing prototype computers built on the cell approach, and comparing results with present architectures, it's possible that the cell has no significant advantage in any market. It's just an engineering fact of life. Ideas that look great on paper, that people get all excited about, these ideas don't always pan out the way we hope. Win some, lose some.
Just because a cluster of high powered computers can do wondrous things, it is not assured that a cluster of low power processors will do great things in an individual computer. When it's broken down to the smallest element in the mix, there may be no substitute for raw power in individual processor cores.
The PowerBook are great. They are really fast what-so-ever. No G5 soon. G4s are perfect. 8)
It seems to me like the aluminum PowerBook is a design Apple is going to use temporarily -- a stop gap to the Next Big Thing?.
I have heard from a few people and read online that manufacturing the Titanium PowerBook was wasteful and costly.
Apple knew from its ongoing work on the G5 that it was going to have a unique thermal zone design, with an aluminum enclosure. I believe that Apple decided to bring things into parity with the PowerBooks by altering a design already in use with the iBooks, as a stop-gap measure.
I mean, look at the aluminum PowerBook's keyboard -- it has the same font style as is on the iBook! Not that this is the sole indicator that the design of the iBook was lifted for the PowerBook, but I think that even the fact that the iBook team built the 12-inch PowerBook is telling.
The timing of the announcement of new PowerBooks last winter at MWSF seemed rather odd, given the release of the Titanium, only two months previously.
The whole thing seems kind of transitionary, if you ask me. Maybe I'm just in a strange funk
It seems to me like the aluminum PowerBook is a design Apple is going to use temporarily -- a stop gap to the Next Big Thing?.
I think it was a concession, after years of trying, that titanium was just an all-around bad idea as a laptop shell. If we've learned anything over the last few years, it's that Apple has crack materials and electrical engineers, and they flailed around more with the TiBook than with any other model I've ever seen come out of Cupertino - including the Cube.
Aluminum is significantly cheaper and much, much easier to work with, and it allowed them to keep the original design - which, despite all its foibles, is probably the most successful since the WallStreet inspired comparisons to Jayne Mansfield.
Quote:
Apple knew from its ongoing work on the G5 that it was going to have a unique thermal zone design, with an aluminum enclosure. I believe that Apple decided to bring things into parity with the PowerBooks by altering a design already in use with the iBooks, as a stop-gap measure.
I mean, look at the aluminum PowerBook's keyboard -- it has the same font style as is on the iBook! Not that this is the sole indicator that the design of the iBook was lifted for the PowerBook, but I think that even the fact that the iBook team built the 12-inch PowerBook is telling.
There aren't really hardware teams within Apple. Apple Executive has clarified that. That said, it's not at all surprising that the 12" PowerBook borrows liberally from the 12" iBook. Why shouldn't it? It's a lot easier than trying to scale a bigger but still cramped motherboard down.
I wonder whether, in fact, the aluminum PowerBooks informed the G5 case, or whether Ives realized that it was better for both, or something like that. At any rate, now that the G5 is aluminum as well, I can see the general style sticking around for a while.
I don't think it's possible to partition a laptop in the way that the G5 (or a rackmount server) is partitioned. It's just too tight in there, and the roar of an Xserve is not acceptable to the average laptop user. You'll continue to see a lot more pipes wicking heat around to radiators, and a lot more strategies adapted to the fact that everything is stacked on top of everything else and there's hardly any room to circulate air.
The WallStreet reminds me: Maybe Apple should go back to stainless steel, only this time without the WallStreet's plastic covering. That would be badass.
It seems to me like the aluminum PowerBook is a design Apple is going to use temporarily -- a stop gap to the Next Big Thing?.
So the question is what is the next big thing? Dual G4s, a low power 970, a new processor follow on to the 750 series or something really new as this thread has suggested.
At this point I have a hard time believing that Apple and IBM are going to get a high enough power reduction in the 970 series to enable placement in the PowerBook. Another go around with the G4 is possible either with or without SMP, but I don't think Apple really want that. So that leave a new processor which by the way seems to have its own rumors or a multiprocessor as described here in this thread.
Quote:
>>>>>>>>>>snipped
The whole thing seems kind of transitionary, if you ask me. Maybe I'm just in a strange funk
Strangenest is the PowerBooks life this year. What it will morph into is an open question. It would be nice if there was actually solid evidence that the 970 can be reimplemented in a low power device at a marketable frequency. Since there is no evidence of this vry low power 970 your funk is justified.
It would be a sad day for Apple if the tried to deliever to us 1.2 or 1.4 GHz 970's as laptop machines. A more substantial performance increase is need in the laptop arena. Beyond that Apple needs to win back its low power leadership.
Aluminum cases are the "hottest" cases in the hobbyist community. Everyone wants one. Form their lightness to their heat dissipation properties to their nice sheen. Apple embraced the hobbyist community when they release the PowerMac G5 and now has the leading Case design with the awesome heat zones.
The Titanium Notebooks were the most lusted after notebooks and the Aluminum improve on their design, only now there is no paint chipping problem. They are beautiful notebooks. But, Apple is getting some heat as the Pentium-M is starting to really put some pressure on, though the Powerbooks are still superior, imo, it is getting to close to be comfortable. I don't think Apple wants to go through a repeat of the desktop marketshare losses. And, why should they? They have excellent notebooks across the price range.
They only now need to differentiate the Powerbooks by moving them forward into the 970 world and the sooner the better. The iBooks then can get faster with the 7447 and its lower heat and lead on the low-end, while the Powerbooks catapult forward into 970 speeds.
Comments
Originally posted by wizard69
It is not a question of ignoring your arguments, the problem is that your arguments are invalid and apparently not based on contemporary knowledge in the field.
My arguments are invalid are they? Not based on contemporary knowledge are they?
Dark side blinded you to reason, it has. Argue with you I will not.
The point is that you have been arguing that the described technology is impossible because a new instruction set is required or new hardware is required or a new operating system is required. Clusters in general and the PPC cluster in particular, should demonstrate clearly that it is possible to implement the discussed technology WITHOUT the need for instruction set modifications or special processors.
Aside from the fact that you don't seem to understand the difference between the impossible and the merely unreasonably difficult, this assertion proves that you have a fundamental lack of understanding of what HPC is about.
So At least we agree that the nodes are capable of running conventional code while being part of a cluster.
I said they could. That's not the point. You're arguing at cross purposes here because you really don't know what you're talking about. (About clustering, that is. You may know a lot about other things you talk about.)
I do know a couple of things though. One is that Apple will have hard time stuffing a 970 into a laptop even with a die shrink. The second is that Apple is probally in the best position of any company, in the computing business, to be able to successfully launch such a machine. This is largely a result of previous efforts to support the G4 in dual processor configurations.
You don't know as much as you think you do. And as long as you continue to believe you know as much as you think you do, you never will.
If the current rumors on AI are correct about the PPC970+, we may well see 1.8 - 2.0GHz G5 Powerbooks this summer. What with SSOI and Power Tune, this could the "low energy" chip rumored to be in the works.
Originally posted by Tomb of the Unknown
You don't know as much as you think you do. And as long as you continue to believe you know as much as you think you do, you never will.
you should talk in the mirror
Originally posted by Nr9
you should talk in the mirror
Hey, I'm willing to be educated. Show me something. So far you all I've seen from you is the insistance that STI are really just building a cluster and Apple is building a cluster in a laptop because a chip you haven't seen yet is going to be bad at power management.
Now, if in my ignorance, I have some part of that wrong please enlighten me.
Originally posted by Tomb of the Unknown
Hey, I'm willing to be educated. Show me something. So far you all I've seen from you is the insistance that STI are really just building a cluster and Apple is building a cluster in a laptop because a chip you haven't seen yet is going to be bad at power management.
Now, if in my ignorance, I have some part of that wrong please enlighten me.
something you should know, is that comments like
You don't know as much as you think you do. And as long as you continue to believe you know as much as you think you do, you never will.
in response to:
I do know a couple of things though. One is that Apple will have hard time stuffing a 970 into a laptop even with a die shrink. The second is that Apple is probally in the best position of any company, in the computing business, to be able to successfully launch such a machine. This is largely a result of previous efforts to support the G4 in dual processor configurations
are neither constructive nor accurate.
a rumor is a rumor. you dont dispel another rumor using another rumor as proof.
Originally posted by Tomb of the Unknown
Dark side blinded you to reason, it has. Argue with you I will not.
Urge to make Voltaire quote rising...
Barto
Originally posted by Nr9
a rumor is a rumor. you dont dispel another rumor using another rumor as proof.
That's funny coming from someone who accused IBM of lying about technology in their next processor.
It's common sense. Which do you think is more likely to be in the next Powerbook G5, a slimmed down, power saving PPC970 or a 4 CPU cluster with as many transistors as a PPC970?
Oh, wait, forget I asked. I think I know what your answer will be and it will have nothing to do with common sense.
Originally posted by Tomb of the Unknown
I said they could. That's not the point. You're arguing at cross purposes here because you really don't know what you're talking about. (About clustering, that is. You may know a lot about other things you talk about.)
My point with respect to clustering, is that a cluster can be built out of common hardware. These clusters can be built at the board level like VT does or similarly on a MCM. SOC just provides another scaling factor. New hardware instructions and massive rewrites to the operating system are not required.
Quote:
If the current rumors on AI are correct about the PPC970+, we may well see 1.8 - 2.0GHz G5 Powerbooks this summer. What with SSOI and Power Tune, this could the "low energy" chip rumored to be in the works.
I would absolutely love to hear in feburary that the 970+, along with its support chips, are running with low enough power disipation to go into a laptop. That would be just fantastic. On the other hand I'm not too sure that IBM is going to be able to cut power usage by 50%. Even if they hit 50% it would still be a rather power hungery machine relative to what else is possible.
Quote:
I wonder if anybody remembers the Transputer? What has been described is something like a Transputer but with fewer porcessors of the same family as the control processor. With a bit of modernization it would be really cool. Especially if they can map some of the current operating system functionality onto the implementation.
Dave
Originally posted by Splinemodel
I really hope nr9 is right. I'm an EE and I would love to have a powerbook with a processor that's actually a revolutionary development. Whether it's real or not, I think this rumor is damn cool.
Yet the cell appears to have a great deal of promise. If Sony is pursuing the cell for PlayStation 3, it is surely way beyond a 'blue sky' idea. They have it working, at least in prototype stage. Since the PlayStation is a low cost consumer product, the cell must have potential for low cost as well as a high performance at low power. So it should not be surprising to hear that Apple is working on the same thing. But realistically, at what stage is Apple in producing a product?
My bet is that Apple will not beat Sony to market with cell technology. If this rumor is true about a four core processor using message passing, it is very likely a prototype chip for prototype computers. There would be a tremendous amount of software experimentation and testing to do on a new concept like this. Apple would have to get hundreds of these boxes in the hands of their own software engineers as well as a few key vendors.
It would be foolish to come out with such a revolutionary product quickly. When does Sony plan to introduce the PS/3? This would be the very earliest date that Apple could have a product. Likely it would be longer. If after a lot of experimentation and testing Apple finds cell technology does not offer some significant advantage, then it may never see the light of day.
Originally posted by Tomb of the Unknown
That's funny coming from someone who accused IBM of lying about technology in their next processor.
It's common sense. Which do you think is more likely to be in the next Powerbook G5, a slimmed down, power saving PPC970 or a 4 CPU cluster with as many transistors as a PPC970?
Oh, wait, forget I asked. I think I know what your answer will be and it will have nothing to do with common sense.
i enjoy the luxury of making such comments because I have sources. you don't. you are just speculating
I believe there is a very good chance that the cell type processor will make it into Apple products eventually. However, the last statement in my posting could also be true. After testing prototype computers built on the cell approach, and comparing results with present architectures, it's possible that the cell has no significant advantage in any market. It's just an engineering fact of life. Ideas that look great on paper, that people get all excited about, these ideas don't always pan out the way we hope. Win some, lose some.
Just because a cluster of high powered computers can do wondrous things, it is not assured that a cluster of low power processors will do great things in an individual computer. When it's broken down to the smallest element in the mix, there may be no substitute for raw power in individual processor cores.
Originally posted by Nr9
i enjoy the luxury of making such comments because I have sources. you don't. you are just speculating
what a bunch of BS, this guy is a classic ignoramous troll
Originally posted by Agent Macintosh
The PowerBook are great. They are really fast what-so-ever. No G5 soon. G4s are perfect.
It seems to me like the aluminum PowerBook is a design Apple is going to use temporarily -- a stop gap to the Next Big Thing?.
I have heard from a few people and read online that manufacturing the Titanium PowerBook was wasteful and costly.
Apple knew from its ongoing work on the G5 that it was going to have a unique thermal zone design, with an aluminum enclosure. I believe that Apple decided to bring things into parity with the PowerBooks by altering a design already in use with the iBooks, as a stop-gap measure.
I mean, look at the aluminum PowerBook's keyboard -- it has the same font style as is on the iBook! Not that this is the sole indicator that the design of the iBook was lifted for the PowerBook, but I think that even the fact that the iBook team built the 12-inch PowerBook is telling.
The timing of the announcement of new PowerBooks last winter at MWSF seemed rather odd, given the release of the Titanium, only two months previously.
The whole thing seems kind of transitionary, if you ask me. Maybe I'm just in a strange funk
Originally posted by DHagan4755
The whole thing seems kind of transitionary, if you ask me. Maybe I'm just in a strange funk
pretty sure that's the case, especially after that 20 inch powerbook thread
Originally posted by DHagan4755
It seems to me like the aluminum PowerBook is a design Apple is going to use temporarily -- a stop gap to the Next Big Thing?.
I think it was a concession, after years of trying, that titanium was just an all-around bad idea as a laptop shell. If we've learned anything over the last few years, it's that Apple has crack materials and electrical engineers, and they flailed around more with the TiBook than with any other model I've ever seen come out of Cupertino - including the Cube.
Aluminum is significantly cheaper and much, much easier to work with, and it allowed them to keep the original design - which, despite all its foibles, is probably the most successful since the WallStreet inspired comparisons to Jayne Mansfield.
Apple knew from its ongoing work on the G5 that it was going to have a unique thermal zone design, with an aluminum enclosure. I believe that Apple decided to bring things into parity with the PowerBooks by altering a design already in use with the iBooks, as a stop-gap measure.
I mean, look at the aluminum PowerBook's keyboard -- it has the same font style as is on the iBook! Not that this is the sole indicator that the design of the iBook was lifted for the PowerBook, but I think that even the fact that the iBook team built the 12-inch PowerBook is telling.
There aren't really hardware teams within Apple. Apple Executive has clarified that. That said, it's not at all surprising that the 12" PowerBook borrows liberally from the 12" iBook. Why shouldn't it? It's a lot easier than trying to scale a bigger but still cramped motherboard down.
I wonder whether, in fact, the aluminum PowerBooks informed the G5 case, or whether Ives realized that it was better for both, or something like that. At any rate, now that the G5 is aluminum as well, I can see the general style sticking around for a while.
I don't think it's possible to partition a laptop in the way that the G5 (or a rackmount server) is partitioned. It's just too tight in there, and the roar of an Xserve is not acceptable to the average laptop user. You'll continue to see a lot more pipes wicking heat around to radiators, and a lot more strategies adapted to the fact that everything is stacked on top of everything else and there's hardly any room to circulate air.
The WallStreet reminds me: Maybe Apple should go back to stainless steel, only this time without the WallStreet's plastic covering. That would be badass.
Originally posted by DHagan4755
It seems to me like the aluminum PowerBook is a design Apple is going to use temporarily -- a stop gap to the Next Big Thing?.
So the question is what is the next big thing? Dual G4s, a low power 970, a new processor follow on to the 750 series or something really new as this thread has suggested.
At this point I have a hard time believing that Apple and IBM are going to get a high enough power reduction in the 970 series to enable placement in the PowerBook. Another go around with the G4 is possible either with or without SMP, but I don't think Apple really want that. So that leave a new processor which by the way seems to have its own rumors or a multiprocessor as described here in this thread.
Quote:
>>>>>>>>>>snipped
The whole thing seems kind of transitionary, if you ask me. Maybe I'm just in a strange funk
Strangenest is the PowerBooks life this year. What it will morph into is an open question. It would be nice if there was actually solid evidence that the 970 can be reimplemented in a low power device at a marketable frequency. Since there is no evidence of this vry low power 970 your funk is justified.
It would be a sad day for Apple if the tried to deliever to us 1.2 or 1.4 GHz 970's as laptop machines. A more substantial performance increase is need in the laptop arena. Beyond that Apple needs to win back its low power leadership.
Dave
Quote:
The Titanium Notebooks were the most lusted after notebooks and the Aluminum improve on their design, only now there is no paint chipping problem. They are beautiful notebooks. But, Apple is getting some heat as the Pentium-M is starting to really put some pressure on, though the Powerbooks are still superior, imo, it is getting to close to be comfortable. I don't think Apple wants to go through a repeat of the desktop marketshare losses. And, why should they? They have excellent notebooks across the price range.
They only now need to differentiate the Powerbooks by moving them forward into the 970 world and the sooner the better. The iBooks then can get faster with the 7447 and its lower heat and lead on the low-end, while the Powerbooks catapult forward into 970 speeds.