Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
motorola in trouble?
- Thread starter pc_convert?
- Start date
- Sort by reaction score
The game cube does not use a G4.
I just got through researching this. I was wrong; the Game Cube does not run on a G4... It's a G3, hot-rodded. Here's a link with some of the spec info. In an effort to save people the time and trouble of going to another link, I'll qoute the page:
I will also qoute Satoru Iwata. He is, currently, Nintendo's creative director and is responsible for some of the bigger game titles Nintendo has done through the years. This quote was his answer to a question about the setup of the Game Cube:
An interesting side-note to all this is that the Gekko, evidently, out-performs the G3s that Apple has been getting from Motorola. So, another case against keeping them involved in operations.
Nintendo is already looking at using the Gekko in other game systems outside of the Cube (Game Boy 64?). It can split it's architecture to run two seperate and discrete 32 bit processes, much like the Sledgehammer. So, maybe a case against bringing AMD into this as well.
I just got through researching this. I was wrong; the Game Cube does not run on a G4... It's a G3, hot-rodded. Here's a link with some of the spec info. In an effort to save people the time and trouble of going to another link, I'll qoute the page:
The final production speed is actually 485MHz. They had to do that and slow down the Flipper audio/video card, as it ran hotter.
I will also qoute Satoru Iwata. He is, currently, Nintendo's creative director and is responsible for some of the bigger game titles Nintendo has done through the years. This quote was his answer to a question about the setup of the Game Cube:
The performance reports I've seen (4.7 GigaFLOPS!) indicate the Gekko is more hot-rodded than he's letting on.
An interesting side-note to all this is that the Gekko, evidently, out-performs the G3s that Apple has been getting from Motorola. So, another case against keeping them involved in operations.
Nintendo is already looking at using the Gekko in other game systems outside of the Cube (Game Boy 64?). It can split it's architecture to run two seperate and discrete 32 bit processes, much like the Sledgehammer. So, maybe a case against bringing AMD into this as well.
Point well made!
The PPC architecture has plenty of life in it. IBM pushes the envelope with the G3, customizing it, and basically turning it into a hotrod, while Motorola adds AltiVec to the G4 and struggles to get it past 1GHz...
What gives?
The PPC architecture has plenty of life in it. IBM pushes the envelope with the G3, customizing it, and basically turning it into a hotrod, while Motorola adds AltiVec to the G4 and struggles to get it past 1GHz...
What gives?
What gives?
I agree with pc_convert? (and thanks, by the way). Motorola is having trouble all the way around.
I think the problem stems from Motorola failing to take the industry seriously. They are fighting to maintain in all areas of their operations. They seem to have squandered their resources. But, why? Was it arrogance? I'm convinced it's a management issue, personally. Case in point: My father is a retired firefighter captain. As a firefighter, you regularly have to use your handy-talkie in the middle of a burning structure. It could be anything: a garage, someone's house, petroleum strorage facility; you name it. Either way, fire actively emits various disruptive, distortive radio and electro-magnetic signals. This is a bad environment to try to use -and rely upon- any form of radio device. Consequently, no brand of hand radios he has used worked worth a damn, except for one... Motorola's. And guess what product Motorola elected to stop production of, about 5 years ago. Ignorant bastards. Theirs was, also, the only hand radio capable of taking the physical punishment of his job. My dad wasn't the only one to say this, either. He subscribes to a trade publication called Firehouse Magazine. They had an article lamenting Moto's decision. It would seem that, all 'round, these people are, no longer, the sharpest tacks in the box.
They have been loudly trumeting some new 64 bit RISC processor of theirs. Almost as if to attract the attention that Apple has not afforded them, this time.
I see Motorola like this: in the mid '80s, they were a real contender and someone to be reckoned with. This was due, in no small part, to Apple's insistance on favoring them, and on not working with Big Blue. But as the 68k processors started getting recognition (from performance reviews of Macs), they came into greater demand. So, they started supplying them to Commodor and Atari. I understand you don't want to rely on one customer. But, you never want to rest on your laurels, either. And I think they did just that. Accordingly, they're struggling to catch up to -and keep up with- IBM, now that they're stuck having to work with them and their designs.
The fact that they involved themselves with IBM and Apple to [supposedly] co-design the 601 Power PC processor, in '91/'92, indicates that they didn't have an equal or better design of their own to offer, at the time. It wasn't until they worked with them that they became known for developing RISC processors. Their biggest in-house success came from the Cold Fire chip, which appeared around '95-'96. The only people, I've ever met, who know about it are synthesizer/sampler tech heads like myself. It's used in E-Mu Systems' Emulator 4 and, pretty much, every other unit in their lineup.
Motorola has been very quiet about what level microns they are capable of producing at. That is as important a statistic to front as any. It indicates the level of production sophistication you have achieved as a company. I doubt they could jump into .10 micron production without serious help or trouble. I can't imagine Steve Jobs giving them any more chances to redeem themselves for a while. If he does, I will be shocked and somewhat disappointed.
I see the line of logic with the second partner design. It protects Apple from being SOL if the primary vendor's factory breaks down or explodes or something. IBM demonstrated this when Motorola's new G4 production facility, totally, failed. Motorla was down for approximately 3 months. Before 2 weeks had passed, Steve Jobs had forced Motorola execs to pay a visit and got IBM up and running to catch up and take in the slack in the meantime. To our benefit, that's also when the 800MHz G4s were developed (thank you IBM). In 2 months time IBM had fully working mules of a G4 they could wind up to 900MHz or so. Motorola pitched an absolute fit. Apparently, they even threatened to sue(?)! Another indication that they are really pushing the threshold of their technical competency.
All this said, I think Moto is out. If not, they damn well should be.
I agree with pc_convert? (and thanks, by the way). Motorola is having trouble all the way around.
I think the problem stems from Motorola failing to take the industry seriously. They are fighting to maintain in all areas of their operations. They seem to have squandered their resources. But, why? Was it arrogance? I'm convinced it's a management issue, personally. Case in point: My father is a retired firefighter captain. As a firefighter, you regularly have to use your handy-talkie in the middle of a burning structure. It could be anything: a garage, someone's house, petroleum strorage facility; you name it. Either way, fire actively emits various disruptive, distortive radio and electro-magnetic signals. This is a bad environment to try to use -and rely upon- any form of radio device. Consequently, no brand of hand radios he has used worked worth a damn, except for one... Motorola's. And guess what product Motorola elected to stop production of, about 5 years ago. Ignorant bastards. Theirs was, also, the only hand radio capable of taking the physical punishment of his job. My dad wasn't the only one to say this, either. He subscribes to a trade publication called Firehouse Magazine. They had an article lamenting Moto's decision. It would seem that, all 'round, these people are, no longer, the sharpest tacks in the box.
They have been loudly trumeting some new 64 bit RISC processor of theirs. Almost as if to attract the attention that Apple has not afforded them, this time.
I see Motorola like this: in the mid '80s, they were a real contender and someone to be reckoned with. This was due, in no small part, to Apple's insistance on favoring them, and on not working with Big Blue. But as the 68k processors started getting recognition (from performance reviews of Macs), they came into greater demand. So, they started supplying them to Commodor and Atari. I understand you don't want to rely on one customer. But, you never want to rest on your laurels, either. And I think they did just that. Accordingly, they're struggling to catch up to -and keep up with- IBM, now that they're stuck having to work with them and their designs.
The fact that they involved themselves with IBM and Apple to [supposedly] co-design the 601 Power PC processor, in '91/'92, indicates that they didn't have an equal or better design of their own to offer, at the time. It wasn't until they worked with them that they became known for developing RISC processors. Their biggest in-house success came from the Cold Fire chip, which appeared around '95-'96. The only people, I've ever met, who know about it are synthesizer/sampler tech heads like myself. It's used in E-Mu Systems' Emulator 4 and, pretty much, every other unit in their lineup.
Motorola has been very quiet about what level microns they are capable of producing at. That is as important a statistic to front as any. It indicates the level of production sophistication you have achieved as a company. I doubt they could jump into .10 micron production without serious help or trouble. I can't imagine Steve Jobs giving them any more chances to redeem themselves for a while. If he does, I will be shocked and somewhat disappointed.
I see the line of logic with the second partner design. It protects Apple from being SOL if the primary vendor's factory breaks down or explodes or something. IBM demonstrated this when Motorola's new G4 production facility, totally, failed. Motorla was down for approximately 3 months. Before 2 weeks had passed, Steve Jobs had forced Motorola execs to pay a visit and got IBM up and running to catch up and take in the slack in the meantime. To our benefit, that's also when the 800MHz G4s were developed (thank you IBM). In 2 months time IBM had fully working mules of a G4 they could wind up to 900MHz or so. Motorola pitched an absolute fit. Apparently, they even threatened to sue(?)! Another indication that they are really pushing the threshold of their technical competency.
All this said, I think Moto is out. If not, they damn well should be.
Concern with the next generation of processers
I'm not that worried about it, right now. Apple has cleaned up the code base for OS-X that it tends to outrun -or at least match step with- Linux. Having achieved that, it also outruns Windows. OS-X is, currently good for up to 4 processors. Fine by me; I can live with a quad G4 Power Mac. If Total Impact can squeeze four G3's or G4s (your pick) onto a PCI board, and keep 512MB of RAM on it, then, so should Apple. What should, really, be done is optimize the G4. IBM proved there is a h*ll of a lot of kick left in the thing when you consider the performance increase they made by tweaking the G3 for Nintendo. I feel the G4 is being held back, compared to the G3. It's twice the bit depth, yet, (compred MHz to Mhz) has only around a 40% - 60% improvement in number of instructions per second it handles. And that's with that blasted Velocity Engine. I feel that IBM can have a G4e (sounds good) ready to go, practically, at the drop of a hat. Now, if we can get Apple to drop that hat!
The G5 probably won't show for another MacWorld convention. It's too big a deal not to keep for an Apple-centric shindig. I figure the G5 will, hopefully, be IBM's Power 4 processor. If I can ever dig up a URL on the thing -and I have been digging- I will post it here. I found a perfect one about a year ago. It listed performance specs and everything. I just can't remember where it is.
I'm not that worried about it, right now. Apple has cleaned up the code base for OS-X that it tends to outrun -or at least match step with- Linux. Having achieved that, it also outruns Windows. OS-X is, currently good for up to 4 processors. Fine by me; I can live with a quad G4 Power Mac. If Total Impact can squeeze four G3's or G4s (your pick) onto a PCI board, and keep 512MB of RAM on it, then, so should Apple. What should, really, be done is optimize the G4. IBM proved there is a h*ll of a lot of kick left in the thing when you consider the performance increase they made by tweaking the G3 for Nintendo. I feel the G4 is being held back, compared to the G3. It's twice the bit depth, yet, (compred MHz to Mhz) has only around a 40% - 60% improvement in number of instructions per second it handles. And that's with that blasted Velocity Engine. I feel that IBM can have a G4e (sounds good) ready to go, practically, at the drop of a hat. Now, if we can get Apple to drop that hat!
The G5 probably won't show for another MacWorld convention. It's too big a deal not to keep for an Apple-centric shindig. I figure the G5 will, hopefully, be IBM's Power 4 processor. If I can ever dig up a URL on the thing -and I have been digging- I will post it here. I found a perfect one about a year ago. It listed performance specs and everything. I just can't remember where it is.
I don't think the G5 will be IBM's POWER4. POWER4 is v.expensive. Yes, it kicks ass and is PPC binary compatible but it won't be the G5.
The POWER4 series are used in IBM's top servers, these things are amazing. Just check here for info POWER4 also go here and search for POWER4 to be amazed.
POWER4 is v.nice but way too expensive.
G4e sounds good...
I definately agree about quad G4's. Apple should have some quad G4 machines for the server side/ ray-tracing etc. especially if the G5 is late/delayed.
The POWER4 series are used in IBM's top servers, these things are amazing. Just check here for info POWER4 also go here and search for POWER4 to be amazed.
POWER4 is v.nice but way too expensive.
G4e sounds good...I definately agree about quad G4's. Apple should have some quad G4 machines for the server side/ ray-tracing etc. especially if the G5 is late/delayed.
price -vs- POWER 4 (with apologies for mammoth posts)
Quoting PC_Convert?:
This, unfortunately for us, is true <damn>. However, ther are a few reasons to keep it in the range of hopes and Christmas lists:
#1.) As you said, pc_convert? (can we call you Mr. _convert?..? ;-> j/k), the POWER 4 chip is being assigned to a number of IBM's top end servers. However, the G4 has been used in a number of other servers and Apple got a version of it; specifically, the A-6000 was listed, at one time, as running the G4 processor. Obviously, the difference between the logic board for it -vs- a Power Mac is like comparing a Stealth Bomber -vs- a Volkswagen GTI. But, that could hint at something. IBM/Motorola made more money, on the G4, once they started allowing sales of it by companies operating outside of the markets they, themselves, were aiming for. And in having done so, Apple got a much needed "big dog" processor while the Linux community gained a new best friend in brute force processing , Total Impact. If IBM, truly, wants to increase the speed of their return profits on the POWER 4, they had better consider a vendor from a market that doesn't directly compete with them. And, Apple is about as far away from competing with them as anyone is ever going to be.
#2.)Basic macro-economics! The more attainable the product, the greater the popularity; the greater the popularity, the greater the demand; the greater the demand, the greater the profits. Emphasis on the word "attainable", here. This is to imply, both, reasonable price and adequate supply. No matter what, IBM is going to have look at the cost of having developed that lovely little demon and figure the fastest way to make their money back on it. If they keep it in their premium machines, and nowhere else, it won't make them any money for at least a year and a half.
#3.)A possible POWER 2! One of the links pc_convert kindly provided us with (thank you, of course) noted the possibility of scaling the design of the POWER 4 processor down to a 4 core system as opposed to the bombastic 8 that it's sporting, now. Here's what it said:
Quoting PC_Convert?:
This, unfortunately for us, is true <damn>. However, ther are a few reasons to keep it in the range of hopes and Christmas lists:
#1.) As you said, pc_convert? (can we call you Mr. _convert?..? ;-> j/k), the POWER 4 chip is being assigned to a number of IBM's top end servers. However, the G4 has been used in a number of other servers and Apple got a version of it; specifically, the A-6000 was listed, at one time, as running the G4 processor. Obviously, the difference between the logic board for it -vs- a Power Mac is like comparing a Stealth Bomber -vs- a Volkswagen GTI. But, that could hint at something. IBM/Motorola made more money, on the G4, once they started allowing sales of it by companies operating outside of the markets they, themselves, were aiming for. And in having done so, Apple got a much needed "big dog" processor while the Linux community gained a new best friend in brute force processing , Total Impact. If IBM, truly, wants to increase the speed of their return profits on the POWER 4, they had better consider a vendor from a market that doesn't directly compete with them. And, Apple is about as far away from competing with them as anyone is ever going to be.
#2.)Basic macro-economics! The more attainable the product, the greater the popularity; the greater the popularity, the greater the demand; the greater the demand, the greater the profits. Emphasis on the word "attainable", here. This is to imply, both, reasonable price and adequate supply. No matter what, IBM is going to have look at the cost of having developed that lovely little demon and figure the fastest way to make their money back on it. If they keep it in their premium machines, and nowhere else, it won't make them any money for at least a year and a half.
#3.)A possible POWER 2! One of the links pc_convert kindly provided us with (thank you, of course) noted the possibility of scaling the design of the POWER 4 processor down to a 4 core system as opposed to the bombastic 8 that it's sporting, now. Here's what it said:
My reason for pointing this out is that it's been a consistent component of every reputable Macintosh rumor site's speculation that the next generation Power Mac is due to have an optional 2 or 4 core mother board and will hum along at about 2.? GHz. The POWER 4 already runs from 1.8 - 2.4 GHz... Hmmm. Considering the POWER 4 is a series of 4 chips wth 2 cores each, then, what about halving or quartering the core count [like the article said] into a simpler version of the same thing? One might be able to do it in the same factory, even. Each chip in the POWER 4 is supposed to be 64 bits. Granted, that would be dropping by 50%, in terms of bit depth, from the G4. But, you're gaining greater performance with the quad 64s in a single die, than, two separate 128s. This would also solve several current design/manufacturing hurdles for Apple. One socket size (for all mother boards) and one CPU type with variable core counts as opposed to two socket sizes and two very different CPUs. No scorching hot Velovity Engine to contend with either.
Lets hope Apple takes note....
They may have... This thread was unreachable for about 12 hours or so from Friday into Saturday. I tried using my bookmark for the thread and tried using the link from the last reply alert email I got on this; I even tried 3 different browsers. Nothing worked. I'm not into, deliberately, being paranoid; however, Apple has been known to troll this and a few other rumor sites. Plus, think of the ego boost if we're right.
They may have... This thread was unreachable for about 12 hours or so from Friday into Saturday. I tried using my bookmark for the thread and tried using the link from the last reply alert email I got on this; I even tried 3 different browsers. Nothing worked. I'm not into, deliberately, being paranoid; however, Apple has been known to troll this and a few other rumor sites. Plus, think of the ego boost if we're right.
With the Dual 1Ghz Powermac G4's it doesn't seem that motorola is doing that badly...
...at least they have enough money to manufacture enough 1 Ghz chips for dual processers.
...at least they have enough money to manufacture enough 1 Ghz chips for dual processers.
Yeah it's nice the G4 has finally hit 1GHZ, but it has nothing to do with Motorola's standing.
The original G4 appeared 31st August 1999 in the 400Mhz PowerMac, in September of that year it was scaled back to 350MHz. By the end of 1999 AMD was at 800MHz.
It's now Jan 2002 and after a redesign the G4+ has finally hit 1GHz while AMD and Intel are pushing 2GHz ish.
Thats 29 months (nearly to the day)for an 600MHz increase in clock speed!
29 Months! - !2 years 5 months! = 600MHz "not doing that badly!"
In 29 months x86 has had a 1.2GHz increase in clock speed - double that of Motorola.
I'm not bitching at Apple, they are doing a great job with the processors their getting, I just think they are being let down by Motorola in a BIG way. Now if the PPC had hit 2GHz you wouldn't here me complaining...(probably)
If you want to check the facts they are here.
I'm not bitching at Apple, they are doing a great job with the processors their getting, I just think they are being let down by Motorola in a BIG way. Now if the PPC had just hit 2GHz you wouldn't hear me complaining....(probably)
Motorola isn't doing that great, either...
The improved architecture of all the G4s that exceed 500MHz can be credited to IBM, not Motorola. Somewhere around a year ago, Motorola put a brand new factory into operation for their flavor of the G4. Three weeks into operation, the factory, still hadn't managed to meet minimum quality or production consistency standards. The assembly line, ultimately, broke down. At that time, the G4 was only good for 500MHz. Steve Jobs was understandably pissed. In almost no time, he had execs from Motorola and IBM visiting Cuppertino (I love how he can do that! Bill Gates can't even pull that off!) and worked out an crisis plan. IBM started producing the G4 processors and Mortorla was warded with sraightening out their sh*t. In the 3 or 4 months that IBM was producing Apple's G4s, they managed to take that design to a producable 800 MHz. They had test mules that could go to 900. Motorola went appaplectic! THEY prevented IBM from producing the higher clocked units stating all IBM was supposed to do was cover for them while they get their factory back on track. Motorola even threatened legal action.
Even once Motorola got their factory on line and working, they still had trouble kicking out adequate supply of the sped up chips.
Look at it this way. Motorola can't break 1.2 GHz. Meanwhile IBM makes Power PC chips up to a "G7" and a "POWER" class of chip that exceeds their own Power PCs (in MHz and processing), yet, still handles those instruction sets. Not to mention the supurlative job IBM did of hot-rodding the G3 for Nintendo to the tune of 4.7 GigaFLOPS @ 485MHz...
Now, you tell me who isn't doing bad...
The improved architecture of all the G4s that exceed 500MHz can be credited to IBM, not Motorola. Somewhere around a year ago, Motorola put a brand new factory into operation for their flavor of the G4. Three weeks into operation, the factory, still hadn't managed to meet minimum quality or production consistency standards. The assembly line, ultimately, broke down. At that time, the G4 was only good for 500MHz. Steve Jobs was understandably pissed. In almost no time, he had execs from Motorola and IBM visiting Cuppertino (I love how he can do that! Bill Gates can't even pull that off!) and worked out an crisis plan. IBM started producing the G4 processors and Mortorla was warded with sraightening out their sh*t. In the 3 or 4 months that IBM was producing Apple's G4s, they managed to take that design to a producable 800 MHz. They had test mules that could go to 900. Motorola went appaplectic! THEY prevented IBM from producing the higher clocked units stating all IBM was supposed to do was cover for them while they get their factory back on track. Motorola even threatened legal action.
Even once Motorola got their factory on line and working, they still had trouble kicking out adequate supply of the sped up chips.
Look at it this way. Motorola can't break 1.2 GHz. Meanwhile IBM makes Power PC chips up to a "G7" and a "POWER" class of chip that exceeds their own Power PCs (in MHz and processing), yet, still handles those instruction sets. Not to mention the supurlative job IBM did of hot-rodding the G3 for Nintendo to the tune of 4.7 GigaFLOPS @ 485MHz...
Now, you tell me who isn't doing bad...