Oh, by the way, I did a search of Appleinsider for the MPC7457-rm and found the link to Motorola's pdf with the roadmap for the proposed MPC 7457-rm, it appears to have been removed.
Off to Arse to search there to see if it was saved anywhere else other than Motorola's website.
Motorola is a YEAR late moving the G4 to 130nm. But now I'm expected to believe in any capacity that Motorla can fab at 90nm. Riiiiiiiiiight. MacBidoobie must think we're fools.
Motorola is dead, Apple has been burned by this company too much. I say good riddance. They've been screwing Mac users for almost 2 freakin' decades with delays and underperforming processors.
By that roadmap, the 7457RM wasn't indicated until 2004 anyway, and RIO not until 2005. As always Motorola is behind the curve.
It would be nearly a miracle if they managed to bring out a low power 7457 (first generation) that might be suitable for notebooks. But at the rate they move, it will still be too little too late.
By the time they manage to do anything at all with the G4 line and this process, IBM will have already brought out the 980.
I'm sure Apple will get some performance out of them. In court. I'm sure they can put the half-1 billion dollars in damages to good use...
Lemon Bon Bon
I feel they should pay me to . My boss tells me each day that i don't work hard enough, in fact it's my G4 lagging behind.
The emotional damage this has caused me easy surpasses 500 mil. dollars
I had to drink to much coffee since I had to wait till the machine finnished something. During that time the price of the coffee went up. I'll make that a round figure...100.000 dollars.
Due to heavy coffee drinking, I had to pee more. Expenses (for the lady who cleans the toilet) plus 1 kidny stone removal procedure. Mmmm...let me say 50.000 dollars.
Since coffee also means a sigaret for me, I smoked the living crap out of me. They costs alot, so 1 package extra each day for, let's say, 3 years comes to 4606 dollars.
While waiting in my office i started masturbating each day. That makes 5 liters of sperm. If I would have donated them to the sperm bank, that would have made me 60.000 dollars.
Lemon, I hope Apple get's more than those 500 mil.
Apple will only use the new improved G4s if they are above 2,5 Ghz. Then and only then will Jobs feel that it is justified not having the chance of beating the craps out of Mot from on stage.
Motorola is dead, Apple has been burned by this company too much. I say good riddance. They've been screwing Mac users for almost 2 freakin' decades with delays and underperforming processors.
Heh. Somebody has a selective memory. Until about 4 years ago, Motorola's chips were better, cheaper, and faster than Intel's offerings. Mac users used to crow about how their hardware was faster the best the PC world could offer.
It's only in the last few years that Mot has started to suck. Now, suddenly, they sucked for decades? Whatever.
Here is a much better translation thanks to Mac Rumors
"According to an internal source (thank-you Superced) Motorola has started producing 0.09 µm chips in their Crolles 1 plant.
That's all we know, but if Motorola manages to produce G4s this small, they might be able to re-enter the frequency race.
It would be difficult to catch up with the PPC 970 in brute force, but a 0.09 µm G4 running at 2 GHz would be an excellent replacement for the G3 in Apple's consumer line, particularily in the iBook.
It remains to be seen whether Apple would be willing to overlook their differences [with Motorola], and accept working with them towards a common future. If we follow Motorola's Roadmap (which they rarely follow themselves), the 0.09 µm G4s would be the 7457-RMs (originally slated for production in 0.1 µm) which incorporate Rapid-IO to maximize memory bandwidth, the current G4's Achilles heel."
By that roadmap, the 7457RM wasn't indicated until 2004 anyway, and RIO not until 2005. As always Motorola is behind the curve.
Um. "RM" stands for 'RapidIO + Memory Controller'. The wavy symbol on the chip is the 'sign' of RapidIO... -> the entire point of the 7457RM was RapidIO + the memory controller.
Now, perhaps _everything_ doesn't shift to RapidIO until 2005, but the 'RM' parts that have that logo on them were intended to have RapidIO.
It's only in the last few years that Mot has started to suck. Now, suddenly, they sucked for decades? Whatever.
Well, I wouldn't have said "sucked", but they have had other major snafus. The 68060 lingered and lingered and eventually it got serious enough for Apple to bail... to something that was (essentially) an IBM design. The formation of AIM was to 'fix' a problem that was fundamentally Mots: If the 68060 had just 'followed' Moore's law from a price/performance perspective, Apple would never have been THAT interested in switching to a POWER derivative.
Maybe someone read my post earlier speculating Mot's next move, because this dovetails with it nicely. To wit, I speculated that:
* Mot realized that their 130nm fab was unsuitable for fabbing anything more intricate than flash memory.
* Mot realized that by the time they released what amounted to a 130nm version of the 7455 (the 7457), no-one would care; when they released the 7457-RM sometime later, no-one would care. Particularly since IBM is now aiming their G3 line at the G4.
* Mot has access to the 90nm fab jointly ventiured with Philips Semi and another major European firm.
So, given all that:
* Mot decides to can the 7457 as planned, and substitute in the 7457-RM (which then disappears from the roadmap as a separate product).
* Mot skips over 130nm and goes straight to 90nm, which they're on the public record as saying they wanted to do.
Result: Mot tries to release a 90nm, RapidIO-enabled G4 in 4Q2003. I can believe that they're trying to do this, since it's just about the only way they'll be able to remain competitive.
Whether they've cleaned up their managerial incompetence enough to pull this off is an open question. But I don't see that they have any other (attractive) options. The above-linked roadmap looks pretty lame. It looks like MacBidouille (or a correspondent) agrees with me... which, given my track record at predicting hardware developments, makes it almost certain that he's wrong.
Availability is listed on the following pdf as 2 quarter of 2003 in a chart near the bottom(in green).Interesting they list it as 1.3 gig while they list the 7455 at 1 gig,so if apple uses it the actual speed should be much higher than that.
Whether they've cleaned up their managerial incompetence enough to pull this off is an open question. But I don't see that they have any other (attractive) options.
We may not know the whole grim picture. <speculation>Motorola may have lost too much money on R&D which hasn't justified itself. While still pretty good for embedded purposes, G4 and its development may well not be going anywhere now. It is even possible that if Moto gives up on G4 and concentrates on 85xx, they will be more successful than ever. If this is so (and both Apple and Moto knew it for a couple of years), then Moto is prepared for Apple to leave. Moreover, in this context, IBM may even become Moto's savior.</speculation> Oops... What am I all about?
uh, I think I must clarify my opinion. The only reason I tracked down that roadmap was to show that the elusive MPC 7457-rm wasn't a rumor(well technically not, but since it was shown only as proposed, I guess you could say that even Motorola considered it a rumor ).
I doubt that it will appear before the end of 2004, if ever. Its' architecture looks to target desktops and Motorola seems to have reduced any emphasis on designing a Host Processor capable of competeting in the desktop market, but well you never know.
As far as 0.09µm processes, well, err, umm, what is Motorola's track record. They're going to jump from 0.18µm to 0.09µm, hhmmmm. With the help of STMicroelectronics and Phillips, it may be possible, but that seems to be a fairly dramatic jump.
On top of that, the MPC7457 should max out with a 200MHz front side bus(not impressive) and to lesson the impact of this slow bus speed, L3 cache would most likely be required. Apparently L3 cache ain't cheap, so although the MPC7457 would be significantly less expensive than the MPC7455 and the IBM 970, the L3 cache might eat up any savings, no?
Unless, and this a BIG if, Motorola has secretly developed the MPC 7457-rm and will begin producing it in quantity withing the next 6 months, I believe it will be hard pressed to supply Apple cpu's, the MPC7457, for any desktops other the the low end and because of the L3 cache, the cost(price/performance) won't be a pretty sight.
Comments
Originally posted by rickag
The entire message including the jpg.
Just tried to send an email to myself and it looks like my email server is down.
Can you enable PM-ing so I can send you a different mail-adress?
Originally posted by NETROMac
Can you enable PM-ing so I can send you a different mail-adress?
It wouldn't help, I'm at work and the email server is down and even if I could get into the server room I wouldn't know what to do.
I'll go ahead and enable PM-ing.
Off to Arse to search there to see if it was saved anywhere else other than Motorola's website.
The elusive roadmap
Motorola is a YEAR late moving the G4 to 130nm. But now I'm expected to believe in any capacity that Motorla can fab at 90nm. Riiiiiiiiiight. MacBidoobie must think we're fools.
Motorola is dead, Apple has been burned by this company too much. I say good riddance. They've been screwing Mac users for almost 2 freakin' decades with delays and underperforming processors.
Originally posted by rickag
Thank heaven for search engines on web boards.
The elusive roadmap
By that roadmap, the 7457RM wasn't indicated until 2004 anyway, and RIO not until 2005. As always Motorola is behind the curve.
It would be nearly a miracle if they managed to bring out a low power 7457 (first generation) that might be suitable for notebooks. But at the rate they move, it will still be too little too late.
By the time they manage to do anything at all with the G4 line and this process, IBM will have already brought out the 980.
Originally posted by rickag
Thank heaven for search engines on web boards.
The elusive roadmap
Exactly, and the RM is no longer on the most recent roadmaps and presentations.
Originally posted by Lemon Bon Bon
I'm sure Apple will get some performance out of them. In court. I'm sure they can put the half-1 billion dollars in damages to good use...
Lemon Bon Bon
I feel they should pay me to . My boss tells me each day that i don't work hard enough, in fact it's my G4 lagging behind.
The emotional damage this has caused me easy surpasses 500 mil. dollars
I had to drink to much coffee since I had to wait till the machine finnished something. During that time the price of the coffee went up. I'll make that a round figure...100.000 dollars.
Due to heavy coffee drinking, I had to pee more. Expenses (for the lady who cleans the toilet) plus 1 kidny stone removal procedure. Mmmm...let me say 50.000 dollars.
Since coffee also means a sigaret for me, I smoked the living crap out of me. They costs alot, so 1 package extra each day for, let's say, 3 years comes to 4606 dollars.
While waiting in my office i started masturbating each day. That makes 5 liters of sperm. If I would have donated them to the sperm bank, that would have made me 60.000 dollars.
Lemon, I hope Apple get's more than those 500 mil.
Originally posted by rickag
Thank heaven for search engines on web boards.
The elusive roadmap
What is with the teeth?
Originally posted by hmurchison
Motorola is dead, Apple has been burned by this company too much. I say good riddance. They've been screwing Mac users for almost 2 freakin' decades with delays and underperforming processors.
Heh. Somebody has a selective memory. Until about 4 years ago, Motorola's chips were better, cheaper, and faster than Intel's offerings. Mac users used to crow about how their hardware was faster the best the PC world could offer.
It's only in the last few years that Mot has started to suck. Now, suddenly, they sucked for decades? Whatever.
"According to an internal source (thank-you Superced) Motorola has started producing 0.09 µm chips in their Crolles 1 plant.
That's all we know, but if Motorola manages to produce G4s this small, they might be able to re-enter the frequency race.
It would be difficult to catch up with the PPC 970 in brute force, but a 0.09 µm G4 running at 2 GHz would be an excellent replacement for the G3 in Apple's consumer line, particularily in the iBook.
It remains to be seen whether Apple would be willing to overlook their differences [with Motorola], and accept working with them towards a common future. If we follow Motorola's Roadmap (which they rarely follow themselves), the 0.09 µm G4s would be the 7457-RMs (originally slated for production in 0.1 µm) which incorporate Rapid-IO to maximize memory bandwidth, the current G4's Achilles heel."
Originally posted by Shaktai
By that roadmap, the 7457RM wasn't indicated until 2004 anyway, and RIO not until 2005. As always Motorola is behind the curve.
Um. "RM" stands for 'RapidIO + Memory Controller'. The wavy symbol on the chip is the 'sign' of RapidIO... -> the entire point of the 7457RM was RapidIO + the memory controller.
Now, perhaps _everything_ doesn't shift to RapidIO until 2005, but the 'RM' parts that have that logo on them were intended to have RapidIO.
Originally posted by robo
It's only in the last few years that Mot has started to suck. Now, suddenly, they sucked for decades? Whatever.
Well, I wouldn't have said "sucked", but they have had other major snafus. The 68060 lingered and lingered and eventually it got serious enough for Apple to bail... to something that was (essentially) an IBM design. The formation of AIM was to 'fix' a problem that was fundamentally Mots: If the 68060 had just 'followed' Moore's law from a price/performance perspective, Apple would never have been THAT interested in switching to a POWER derivative.
* Mot realized that their 130nm fab was unsuitable for fabbing anything more intricate than flash memory.
* Mot realized that by the time they released what amounted to a 130nm version of the 7455 (the 7457), no-one would care; when they released the 7457-RM sometime later, no-one would care. Particularly since IBM is now aiming their G3 line at the G4.
* Mot has access to the 90nm fab jointly ventiured with Philips Semi and another major European firm.
So, given all that:
* Mot decides to can the 7457 as planned, and substitute in the 7457-RM (which then disappears from the roadmap as a separate product).
* Mot skips over 130nm and goes straight to 90nm, which they're on the public record as saying they wanted to do.
Result: Mot tries to release a 90nm, RapidIO-enabled G4 in 4Q2003. I can believe that they're trying to do this, since it's just about the only way they'll be able to remain competitive.
Whether they've cleaned up their managerial incompetence enough to pull this off is an open question. But I don't see that they have any other (attractive) options. The above-linked roadmap looks pretty lame. It looks like MacBidouille (or a correspondent) agrees with me... which, given my track record at predicting hardware developments, makes it almost certain that he's wrong.
Motorola? What is it? Is it a trash can manufacturer?
Lemon Bon Bon
http://e-www.motorola.com/webapp/sps...TR046708718653
http://e-www.motorola.com/brdata/PDF.../MPC7457EC.pdf
Availability is listed on the following pdf as 2 quarter of 2003 in a chart near the bottom(in green).Interesting they list it as 1.3 gig while they list the 7455 at 1 gig,so if apple uses it the actual speed should be much higher than that.
http://e-www.motorola.com/brdata/PDF...CSALESFACT.pdf
Originally posted by Amorph
Whether they've cleaned up their managerial incompetence enough to pull this off is an open question. But I don't see that they have any other (attractive) options.
We may not know the whole grim picture. <speculation>Motorola may have lost too much money on R&D which hasn't justified itself. While still pretty good for embedded purposes, G4 and its development may well not be going anywhere now. It is even possible that if Moto gives up on G4 and concentrates on 85xx, they will be more successful than ever. If this is so (and both Apple and Moto knew it for a couple of years), then Moto is prepared for Apple to leave. Moreover, in this context, IBM may even become Moto's savior.</speculation> Oops... What am I all about?
I doubt that it will appear before the end of 2004, if ever. Its' architecture looks to target desktops and Motorola seems to have reduced any emphasis on designing a Host Processor capable of competeting in the desktop market, but well you never know.
As far as 0.09µm processes, well, err, umm, what is Motorola's track record. They're going to jump from 0.18µm to 0.09µm, hhmmmm. With the help of STMicroelectronics and Phillips, it may be possible, but that seems to be a fairly dramatic jump.
On top of that, the MPC7457 should max out with a 200MHz front side bus(not impressive) and to lesson the impact of this slow bus speed, L3 cache would most likely be required. Apparently L3 cache ain't cheap, so although the MPC7457 would be significantly less expensive than the MPC7455 and the IBM 970, the L3 cache might eat up any savings, no?
Unless, and this a BIG if, Motorola has secretly developed the MPC 7457-rm and will begin producing it in quantity withing the next 6 months, I believe it will be hard pressed to supply Apple cpu's, the MPC7457, for any desktops other the the low end and because of the L3 cache, the cost(price/performance) won't be a pretty sight.