I still believe Apple will NEED a 32bit low power G4 class CPU for the "consumer" range.
Now if Moto can provide it,.. cough then fine, but the entire product line cannot move because the G4 is stuck where it is. The 970 will free the "Pro" line allowing faster and more palatable upgrades to Apple's "Consumer" range.
A 1.42G (7455) G4 iMac would sell well right now, with 7457 speed bumps and maybe even for the iBook, unless IBM has its low power G4 class CPU available.
The reality is that Apple NEEDS to get its products well beyond the 1GHz if it intends on staying business, and FAST!
I agree, Apple will not dump Motorola out of emotion. They will keep them on the back burner while they improve their consumer line. The newer G4s will be for the consumer lines, iBooks, iMacs, maybe a low end screenless Mac. There is obviously use for their processors.
When we put aside our hatred of Motorola, we will see this is a Good Thing? in the long run.
I suspect it will/would be a great loss to Moto if they lost Apple as a PPC customer. Would it put them under? No, but it would hurt profits until significant changes were made in their SPS division.
And they would do as much as they could to prevent that. This revised roadmap is proof to that. Well not exactly proof but it seems like the stuff Apple would ask to include in a new revision of the G4.
Is this "roadmap revision" thing a recent development, or have you just discovered it?
My take is that it just doesn't matter. Mot can't get it up anymore in the high performance field, and I really don't think they'll be able too any time in the future. I'd like to believe differently, but they have let us down time and again.
It's great if they keep developing decent PPCs. Maybe Apple can use them again if they become competitive again...but for now, it's hasta la vista, baby.
I had a feeling the issue would be heavy compiler optimizations similar to the problem the IA-64 platform has presently. I would imagine that IBM thinks the ideal balance of execution units in a desktop processor is 2 fixed point, 2 floating point, and a vector unit. Now let me ask this: instead of a dual core processor with each core having the same execution units as I described above, how about a quad core processor with a single fixed and floating point unit and a vector unit with half the cache for each core? With a properly threaded application (and operating system) would this be advantageous or just an over engineered flop?
One VERY significant difference between the 970 and IA-64 is that the 970 will look at the instructions it is given and attempt to reschedule them so that it can keep itself busy. The IA 64 does not do this. The IA 64 forces such scheduling off on the compiler, making for a much more complex and difficult compiler. If I remember correctly, the IA-64 has four way parallelism in its integer units, so it can in theory do quite a bit of work per clock cycle, so long as the compiler was able to keep all the pipes full. No, the magic number isn't 2 integer units and 2 floating point units, the magic number actually seems to be 4 of both because it is very very difficult to figure out a way to keep more than four units busy on most code. You could make a quad core die with only one execution unit, but keeping two busy isn't too difficult, so you might as well do two. The possible performance benefit of such a system would be that if you could get the transistor count to be very low, that you run the CPU faster, but such a CPU would definitely be the poster child for the MHz myth. It is generally understood that parallelism is a good thing in CPU design.
What you are describing is similar to Intel's "hyperthreading" which is where the CPU exposes itself as if it was a dual core CPU so that multiple threads can simultaneously run on a single core CPU. "Hyperthreading" was first done on the Alpha chips and I seem to recall that it is on tap of the PPC 980 (I could be wrong so don't hold me to this). Hyperthreading is a good way to get a speed boost out of your chips because it can run more threads more of the time. The drawback is that the CPU has a tendency to get warm (so don't look for such a CPU in a laptop any time soon!).
Quote:
Originally posted by Outsider
So would it make more sence to design a processor with minimal L1(like the P4) and L2 cache (or none at all) and concentrate on a super fast system bus, say a fairly wide (64bit) fast (1GHz DDR) bus thatconnects to a similarly fast memory interface? I always thought of cache as a temporary solution to a problem computer technology always had in the 80's and 90's: slow memory interfaces. I envision a futuristic processor with the memory right on the die. Several cores tapping into a central store of main memory, maybe 4GB or so, with an attachment to external solid state magnetic memory, negating the use of any mechanical disk storage. But this is already off topic.
Well, you need some cache. Of course, the ammount of cache that you are going to need depends on what the CPU does. For bandwidth intensive operations, cache is less relevant than your bus speed (because you need to hit main memory for large chunks of data). If you aren't doing bandwidth intensive stuff, but access the same data over and over again, then cache is extremely useful. The latter scenario is common for servers, which is why IBM ships their Power 4 servers with up to 64 MB of L3 cache. More cache is rarely ever a bad thing, but more cache will definitely cost more money and so it doesn't always make sense for the machine that an average user uses.
CPU's will always be faster than the system bus. It is just easier to run a little piece of silicon fast than it is to run something the size of your motherboard fast. Maybe one day we'll get away from synchronous CPU's and busses (this could help quite a bit). Back in the day, CPU's ran at the same speed as thier system bus, but that will probably never happen again (until we go to light based CPU's and motherboards-unless they use different frequencies of light or something like that).
You could make a CPU with memory on the die, and this would essentially be a huge L3 cache. The drawbacks to this are:
Larger die size mean more costly chips
Something of a pain to upgrade your memory
This post has been an entertaining diversion from the bugs that I have to fix...
"Hyperthreading" was first done on the Alpha chips and I seem to recall that it is on tap of the PPC 980 (I could be wrong so don't hold me to this).
Its normally called SMT (simultaneous multi-threading), Hyperthreading is Intel's marketing term. Since that's how the general public first heard of it, no doubt that's how it'll be know in the future... which could be a problem if Intel has trademarked it and now only they can have Hyperthreading. Ugh.
Anyhow, IBM has said that the POWER5 will have SMT. They have said nothing at all about any 9xx series chips except the 970, but there is a general assumption (correct or not) that the 980 will be to the POWER5 as the 970 is to the POWER4.
I think that if Moto can keep the G4's competetive and that if they can the power consumption down, that Apple will probably continue to use them in future laptops. Having said that, I think that the presentation slides are encouraging, but that I will believe that Moto can get something done after they actually deliver. Talk is cheap with these people so I am going to wait for their actions.
This Motorola roadmap with a dual core PowerPC chip sounds like a response to Broadcom's (SiByte) BCM1250 network processor:
Two 64-bit MIPS CPUs
600 MHz-1 GHz
On-chip Multi-Processor Bus
512K On-chip L2 Cache
DDR Memory Controller with 50 Gbps Peak Bandwidth
Integrated Network (3 Ethernet)
System I/O (Hypertransport, PCI bus, serial)
It's sort of race to see if Motorola can ship its 8540 and 8560 chips before Broadcom ships single core variants of this chip architecture. If Broadcom can ship the BCM1250 in the near future, Moto will have a problem. So, I don't think it is a response to IBM. If IBM ships a PPC750 with AltiVec and on-chip memory controller and I/O, then Moto will have a problem because of IBM's process advantage. The Power4/5 derivatives are simply in a different market.
I think that if Moto can keep the G4's competetive and that if they can the power consumption down, that Apple will probably continue to use them in future laptops. Having said that, I think that the presentation slides are encouraging, but that I will believe that Moto can get something done after they actually deliver. Talk is cheap with these people so I am going to wait for their actions.
If rumours about the 970 priceing turn out to be true, then there the 970 will cost equil to or less than the current crop of G4's. This could be a nail in the G4's coffin if it turns out to be true. Personally, I like what I am hearing about the future plans for the PPC platform from IBM a lot better than from Moto. They are moving too slow with the development and implementation of new tech into thier PPC chips. Probably good for the embeded market, but not for the high profile chips that bring the platform press.
If rumours about the 970 priceing turn out to be true, then there the 970 will cost equil to or less than the current crop of G4's. This could be a nail in the G4's coffin if it turns out to be true. Personally, I like what I am hearing about the future plans for the PPC platform from IBM a lot better than from Moto. They are moving too slow with the development and implementation of new tech into thier PPC chips. Probably good for the embeded market, but not for the high profile chips that bring the platform press.
I agree with you that the G4 would be hard pressed to keep up with a 970. No doubts there. Of course, there is more to a chip than its cost and processing power. There is also the power usage problem which is very significant for laptop users. I think that there is still a market for a fast G4 (+ faster bus) in a laptop.
Also, Apple needs to differentiate its machines. Maybe the powerbooks go 970 and the iBooks go G4. I think that this is also a reason to keep the G4 around. (the G4 desperately needs a die shrink!)
To be honest, I am sick and tired of seeing G4's in Apple's configurations. I want them to be gone and to wake up from the bad dream that has been Apple's products for the last year. At the same time, I am a realist and know that the G4 still has a place, though maybe not for long.
To be honest, I am sick and tired of seeing G4's in Apple's configurations. I want them to be gone and to wake up from the bad dream that has been Apple's products for the last year.
A fitting epitaph for the G4 if we leave it there.
I think when the .09 970 hits, Moto' is dead and buried...from top to bottom in Apple's line.#
1.8 970 on .09? Equal to a 3.6 gig G4? In a laptop? Can you say, 'OH-MY-WORD!?' Yeah. The G4 'has' a future. But not at Apple computer. Not by early 2004. That's just where I'm placing my bet. You guys may differ yer mileage.
It would take a dual core G4 NOW to even begin to compete with the 970. Then you've got the 970's bandwidth. AND it's moving to .09 very quickly. Apple will have all the differentiation they need. And I have to wonder...if Apple is getting 25-35% reduct' over Moto's G4 line as it stands...perhaps if Apple buys MORE 970s...they'll get a greater deal...and that pours MORE money into 9XX development.
IBM will probably have a G3 cpu that will answer anything Moto' has to offer the iBook come early 2004. It will be interesting to see where IBM takes the G3 onces Apple 'dumps' Moto'.
Plus, IF Apple sues Moto' for the 'aborted' G5...then that 1 billion in damages can also be pumped into developing the 980 and 990...the kinda dosh IBM aint gonna turn down if it wants to recoup some of that money spent on their New York Fabber...
Still, the Microprocessor Report Editor's assertion that the 'G5' was 'still on track' troubles me... Apple can be left field...at times.
If Apple had any plans to use this new dual core G4, do you think Steve would let Mot talk about it now?
Sure, why not. Apple let Moto talk about the Appolo G4's something like a year before they made it into a Mac. Moto does sell chips to people other than Apple, and at the very least, Moto needs to discuss their chip fab plans in their SEC filings. I profoundly doubt that Apple will be using these chips in their future machines because I am biased against Moto in that I think that they can't deliver on their talk. When Moto ships these chips in late 2005, they are going to be so far behind the curve, they are going to get killed by any competition (embedded or not). Moto is going down.
I profoundly doubt that Apple will be using these chips in their future machines because I am biased against Moto in that I think that they can't deliver on their talk. When Moto ships these chips in late 2005, they are going to be so far behind the curve, they are going to get killed by any competition (embedded or not). Moto is going down.
It's sort of race to see if Motorola can ship its 8540 and 8560 chips before Broadcom ships single core variants of this chip architecture. If Broadcom can ship the BCM1250 in the near future, Moto will have a problem.
I think the BCM1250 is already used in Cisco routers, and the single-core BCM1125 was announced a while ago so it's probably shipping as well.
I agree with you that the G4 would be hard pressed to keep up with a 970. No doubts there. Of course, there is more to a chip than its cost and processing power. There is also the power usage problem which is very significant for laptop users. I think that there is still a market for a fast G4 (+ faster bus) in a laptop.
Also, Apple needs to differentiate its machines. Maybe the powerbooks go 970 and the iBooks go G4. I think that this is also a reason to keep the G4 around. (the G4 desperately needs a die shrink!)
To be honest, I am sick and tired of seeing G4's in Apple's configurations. I want them to be gone and to wake up from the bad dream that has been Apple's products for the last year. At the same time, I am a realist and know that the G4 still has a place, though maybe not for long.
If the 970 costs less than (latest rumour 25% + less than current G4's), then it would be idiotic for Apple to put the more expensive chip in their low cost consumer computers wouldnt it? It would be better for Apple to differentiate between Pro and Consumer with single/dual configs than to use A more expensive chip in the iMac and iBook which are their lowest margin computers.
A fisherman casts a lure into a lake and slowly reels it in with the hope that some fish will sense the lure and mistakenly interpret it as prey. The fisherman using the lure only wishes to trick the fish, there is never any real possibility of the fish obtaining any sort of food.
I don't think Moto is exactly a fisherman in this case, but this new PPC Roadmap is definitely a lure. The Roadmap is Moto's answer to IBM's 970, and the obvious threat to Moto. With G4s powering all Macs save the iBook, Moto stands to lose a lot of business if Apple does something rash, like putting 970s in every Mac line. Maybe this is exactly the sort of competition that was needed to light a fire under Moto's ass, or more precisely, to convince the idiot management at Moto that they should invest resources in desktop PPC development.
Whether Moto provides the meat is another question. To me, this roadmap reads like something thrown together the night before, sort of a "lets combine all the G4 back-burner projects into one chip". Even Moto delivers on every planned promise, the resulting G4 will still fall short of IBM's 970.
Notice how on the roadmap this new 7447/57 .13 chip has a max speed of 1.3 Ghz.
Apple already has a 1.42Ghz 7455. And on the roadmap the 7455 has a max of 1Ghz.
And at the top of the roadmap it says "Frequences listed are for Commercial spec." I've always heard this myth that "Apple overclocks" the recent G4s, and I'm sure that's not true, but perhaps they just don't make the 7455 above 1Ghz for the embedded market - only Apple.
Given the frosty MOT - AAPL relationship right now, do you really think MOT would care whether Jobs wanted them to keep quiet or not?
Money talks.
I'm pretty sure that their very public roadmap is an indication that they've lost Apple in the high end. We might see 1.3GHz notebooks or iMacs or what-have-you, or G4 iBooks, but none of those will warrant secrecy.
It's funny; I first heard mumblings about dual-core G4s back before "Max" was known to be the design that Motorola had settled on - all the way back in '97. It took 8 years, but it looks like it'll finally happen. Waaaay too late for us to care. By then we'll be looking forward to the 980+, or possibly the 990.
Notice how on the roadmap this new 7447/57 .13 chip has a max speed of 1.3 Ghz.
Apple already has a 1.42Ghz 7455. And on the roadmap the 7455 has a max of 1Ghz.
And at the top of the roadmap it says "Frequences listed are for Commercial spec." I've always heard this myth that "Apple overclocks" the recent G4s, and I'm sure that's not true, but perhaps they just don't make the 7455 above 1Ghz for the embedded market - only Apple.
There are several versions of the 7455:
P - 1.85V 65°C
S - 1.85V 75°C
L - 1.6V 105°C
N - 1.3V 105°C
xlr8yourmac listed the part number of a Giga Designs 1.4GHz upgrade to be XPC7455BRX1400PF. It's a 7455B at 1.4 GHz packed in a CBGA, version P, revision F (=3.3).
Comments
Originally posted by hasapi
I still believe Apple will NEED a 32bit low power G4 class CPU for the "consumer" range.
Now if Moto can provide it,.. cough then fine, but the entire product line cannot move because the G4 is stuck where it is. The 970 will free the "Pro" line allowing faster and more palatable upgrades to Apple's "Consumer" range.
A 1.42G (7455) G4 iMac would sell well right now, with 7457 speed bumps and maybe even for the iBook, unless IBM has its low power G4 class CPU available.
The reality is that Apple NEEDS to get its products well beyond the 1GHz if it intends on staying business, and FAST!
I agree, Apple will not dump Motorola out of emotion. They will keep them on the back burner while they improve their consumer line. The newer G4s will be for the consumer lines, iBooks, iMacs, maybe a low end screenless Mac. There is obviously use for their processors.
When we put aside our hatred of Motorola, we will see this is a Good Thing? in the long run.
Originally posted by Junkyard Dawg
I suspect it will/would be a great loss to Moto if they lost Apple as a PPC customer. Would it put them under? No, but it would hurt profits until significant changes were made in their SPS division.
And they would do as much as they could to prevent that. This revised roadmap is proof to that. Well not exactly proof but it seems like the stuff Apple would ask to include in a new revision of the G4.
Is this "roadmap revision" thing a recent development, or have you just discovered it?
My take is that it just doesn't matter. Mot can't get it up anymore in the high performance field, and I really don't think they'll be able too any time in the future. I'd like to believe differently, but they have let us down time and again.
It's great if they keep developing decent PPCs. Maybe Apple can use them again if they become competitive again...but for now, it's hasta la vista, baby.
Originally posted by Outsider
First off, excellent post.
I had a feeling the issue would be heavy compiler optimizations similar to the problem the IA-64 platform has presently. I would imagine that IBM thinks the ideal balance of execution units in a desktop processor is 2 fixed point, 2 floating point, and a vector unit. Now let me ask this: instead of a dual core processor with each core having the same execution units as I described above, how about a quad core processor with a single fixed and floating point unit and a vector unit with half the cache for each core? With a properly threaded application (and operating system) would this be advantageous or just an over engineered flop?
One VERY significant difference between the 970 and IA-64 is that the 970 will look at the instructions it is given and attempt to reschedule them so that it can keep itself busy. The IA 64 does not do this. The IA 64 forces such scheduling off on the compiler, making for a much more complex and difficult compiler. If I remember correctly, the IA-64 has four way parallelism in its integer units, so it can in theory do quite a bit of work per clock cycle, so long as the compiler was able to keep all the pipes full. No, the magic number isn't 2 integer units and 2 floating point units, the magic number actually seems to be 4 of both because it is very very difficult to figure out a way to keep more than four units busy on most code. You could make a quad core die with only one execution unit, but keeping two busy isn't too difficult, so you might as well do two. The possible performance benefit of such a system would be that if you could get the transistor count to be very low, that you run the CPU faster, but such a CPU would definitely be the poster child for the MHz myth. It is generally understood that parallelism is a good thing in CPU design.
What you are describing is similar to Intel's "hyperthreading" which is where the CPU exposes itself as if it was a dual core CPU so that multiple threads can simultaneously run on a single core CPU. "Hyperthreading" was first done on the Alpha chips and I seem to recall that it is on tap of the PPC 980 (I could be wrong so don't hold me to this). Hyperthreading is a good way to get a speed boost out of your chips because it can run more threads more of the time. The drawback is that the CPU has a tendency to get warm (so don't look for such a CPU in a laptop any time soon!).
Originally posted by Outsider
So would it make more sence to design a processor with minimal L1(like the P4) and L2 cache (or none at all) and concentrate on a super fast system bus, say a fairly wide (64bit) fast (1GHz DDR) bus thatconnects to a similarly fast memory interface? I always thought of cache as a temporary solution to a problem computer technology always had in the 80's and 90's: slow memory interfaces. I envision a futuristic processor with the memory right on the die. Several cores tapping into a central store of main memory, maybe 4GB or so, with an attachment to external solid state magnetic memory, negating the use of any mechanical disk storage. But this is already off topic.
Well, you need some cache. Of course, the ammount of cache that you are going to need depends on what the CPU does. For bandwidth intensive operations, cache is less relevant than your bus speed (because you need to hit main memory for large chunks of data). If you aren't doing bandwidth intensive stuff, but access the same data over and over again, then cache is extremely useful. The latter scenario is common for servers, which is why IBM ships their Power 4 servers with up to 64 MB of L3 cache. More cache is rarely ever a bad thing, but more cache will definitely cost more money and so it doesn't always make sense for the machine that an average user uses.
CPU's will always be faster than the system bus. It is just easier to run a little piece of silicon fast than it is to run something the size of your motherboard fast. Maybe one day we'll get away from synchronous CPU's and busses (this could help quite a bit). Back in the day, CPU's ran at the same speed as thier system bus, but that will probably never happen again (until we go to light based CPU's and motherboards-unless they use different frequencies of light or something like that).
You could make a CPU with memory on the die, and this would essentially be a huge L3 cache. The drawbacks to this are:
Larger die size mean more costly chips
Something of a pain to upgrade your memory
This post has been an entertaining diversion from the bugs that I have to fix...
Originally posted by Yevgeny
"Hyperthreading" was first done on the Alpha chips and I seem to recall that it is on tap of the PPC 980 (I could be wrong so don't hold me to this).
Its normally called SMT (simultaneous multi-threading), Hyperthreading is Intel's marketing term. Since that's how the general public first heard of it, no doubt that's how it'll be know in the future... which could be a problem if Intel has trademarked it and now only they can have Hyperthreading. Ugh.
Anyhow, IBM has said that the POWER5 will have SMT. They have said nothing at all about any 9xx series chips except the 970, but there is a general assumption (correct or not) that the 980 will be to the POWER5 as the 970 is to the POWER4.
Two 64-bit MIPS CPUs
600 MHz-1 GHz
On-chip Multi-Processor Bus
512K On-chip L2 Cache
DDR Memory Controller with 50 Gbps Peak Bandwidth
Integrated Network (3 Ethernet)
System I/O (Hypertransport, PCI bus, serial)
It's sort of race to see if Motorola can ship its 8540 and 8560 chips before Broadcom ships single core variants of this chip architecture. If Broadcom can ship the BCM1250 in the near future, Moto will have a problem. So, I don't think it is a response to IBM. If IBM ships a PPC750 with AltiVec and on-chip memory controller and I/O, then Moto will have a problem because of IBM's process advantage. The Power4/5 derivatives are simply in a different market.
Originally posted by Yevgeny
I think that if Moto can keep the G4's competetive and that if they can the power consumption down, that Apple will probably continue to use them in future laptops. Having said that, I think that the presentation slides are encouraging, but that I will believe that Moto can get something done after they actually deliver. Talk is cheap with these people so I am going to wait for their actions.
If rumours about the 970 priceing turn out to be true, then there the 970 will cost equil to or less than the current crop of G4's. This could be a nail in the G4's coffin if it turns out to be true. Personally, I like what I am hearing about the future plans for the PPC platform from IBM a lot better than from Moto. They are moving too slow with the development and implementation of new tech into thier PPC chips. Probably good for the embeded market, but not for the high profile chips that bring the platform press.
Originally posted by JCG
If rumours about the 970 priceing turn out to be true, then there the 970 will cost equil to or less than the current crop of G4's. This could be a nail in the G4's coffin if it turns out to be true. Personally, I like what I am hearing about the future plans for the PPC platform from IBM a lot better than from Moto. They are moving too slow with the development and implementation of new tech into thier PPC chips. Probably good for the embeded market, but not for the high profile chips that bring the platform press.
I agree with you that the G4 would be hard pressed to keep up with a 970. No doubts there. Of course, there is more to a chip than its cost and processing power. There is also the power usage problem which is very significant for laptop users. I think that there is still a market for a fast G4 (+ faster bus) in a laptop.
Also, Apple needs to differentiate its machines. Maybe the powerbooks go 970 and the iBooks go G4. I think that this is also a reason to keep the G4 around. (the G4 desperately needs a die shrink!)
To be honest, I am sick and tired of seeing G4's in Apple's configurations. I want them to be gone and to wake up from the bad dream that has been Apple's products for the last year. At the same time, I am a realist and know that the G4 still has a place, though maybe not for long.
To be honest, I am sick and tired of seeing G4's in Apple's configurations. I want them to be gone and to wake up from the bad dream that has been Apple's products for the last year.
A fitting epitaph for the G4 if we leave it there.
I think when the .09 970 hits, Moto' is dead and buried...from top to bottom in Apple's line.#
1.8 970 on .09? Equal to a 3.6 gig G4? In a laptop? Can you say, 'OH-MY-WORD!?' Yeah. The G4 'has' a future. But not at Apple computer. Not by early 2004. That's just where I'm placing my bet. You guys may differ yer mileage.
It would take a dual core G4 NOW to even begin to compete with the 970. Then you've got the 970's bandwidth. AND it's moving to .09 very quickly. Apple will have all the differentiation they need. And I have to wonder...if Apple is getting 25-35% reduct' over Moto's G4 line as it stands...perhaps if Apple buys MORE 970s...they'll get a greater deal...and that pours MORE money into 9XX development.
IBM will probably have a G3 cpu that will answer anything Moto' has to offer the iBook come early 2004. It will be interesting to see where IBM takes the G3 onces Apple 'dumps' Moto'.
Plus, IF Apple sues Moto' for the 'aborted' G5...then that 1 billion in damages can also be pumped into developing the 980 and 990...the kinda dosh IBM aint gonna turn down if it wants to recoup some of that money spent on their New York Fabber...
Still, the Microprocessor Report Editor's assertion that the 'G5' was 'still on track' troubles me... Apple can be left field...at times.
Lemon Bon Bon
If Apple had any plans to use this new dual core G4, do you think Steve would let Mot talk about it now?
Originally posted by Ensign Pulver
People, we're all overlooking one basic fact:
If Apple had any plans to use this new dual core G4, do you think Steve would let Mot talk about it now?
Sure, why not. Apple let Moto talk about the Appolo G4's something like a year before they made it into a Mac. Moto does sell chips to people other than Apple, and at the very least, Moto needs to discuss their chip fab plans in their SEC filings. I profoundly doubt that Apple will be using these chips in their future machines because I am biased against Moto in that I think that they can't deliver on their talk. When Moto ships these chips in late 2005, they are going to be so far behind the curve, they are going to get killed by any competition (embedded or not). Moto is going down.
I profoundly doubt that Apple will be using these chips in their future machines because I am biased against Moto in that I think that they can't deliver on their talk. When Moto ships these chips in late 2005, they are going to be so far behind the curve, they are going to get killed by any competition (embedded or not). Moto is going down.
Lemon Bon Bon
Originally posted by THT
It's sort of race to see if Motorola can ship its 8540 and 8560 chips before Broadcom ships single core variants of this chip architecture. If Broadcom can ship the BCM1250 in the near future, Moto will have a problem.
I think the BCM1250 is already used in Cisco routers, and the single-core BCM1125 was announced a while ago so it's probably shipping as well.
Originally posted by Yevgeny
I agree with you that the G4 would be hard pressed to keep up with a 970. No doubts there. Of course, there is more to a chip than its cost and processing power. There is also the power usage problem which is very significant for laptop users. I think that there is still a market for a fast G4 (+ faster bus) in a laptop.
Also, Apple needs to differentiate its machines. Maybe the powerbooks go 970 and the iBooks go G4. I think that this is also a reason to keep the G4 around. (the G4 desperately needs a die shrink!)
To be honest, I am sick and tired of seeing G4's in Apple's configurations. I want them to be gone and to wake up from the bad dream that has been Apple's products for the last year. At the same time, I am a realist and know that the G4 still has a place, though maybe not for long.
If the 970 costs less than (latest rumour 25% + less than current G4's), then it would be idiotic for Apple to put the more expensive chip in their low cost consumer computers wouldnt it? It would be better for Apple to differentiate between Pro and Consumer with single/dual configs than to use A more expensive chip in the iMac and iBook which are their lowest margin computers.
I don't think Moto is exactly a fisherman in this case, but this new PPC Roadmap is definitely a lure. The Roadmap is Moto's answer to IBM's 970, and the obvious threat to Moto. With G4s powering all Macs save the iBook, Moto stands to lose a lot of business if Apple does something rash, like putting 970s in every Mac line. Maybe this is exactly the sort of competition that was needed to light a fire under Moto's ass, or more precisely, to convince the idiot management at Moto that they should invest resources in desktop PPC development.
Whether Moto provides the meat is another question. To me, this roadmap reads like something thrown together the night before, sort of a "lets combine all the G4 back-burner projects into one chip". Even Moto delivers on every planned promise, the resulting G4 will still fall short of IBM's 970.
Apple already has a 1.42Ghz 7455. And on the roadmap the 7455 has a max of 1Ghz.
And at the top of the roadmap it says "Frequences listed are for Commercial spec." I've always heard this myth that "Apple overclocks" the recent G4s, and I'm sure that's not true, but perhaps they just don't make the 7455 above 1Ghz for the embedded market - only Apple.
Originally posted by jouster
Given the frosty MOT - AAPL relationship right now, do you really think MOT would care whether Jobs wanted them to keep quiet or not?
Money talks.
I'm pretty sure that their very public roadmap is an indication that they've lost Apple in the high end. We might see 1.3GHz notebooks or iMacs or what-have-you, or G4 iBooks, but none of those will warrant secrecy.
It's funny; I first heard mumblings about dual-core G4s back before "Max" was known to be the design that Motorola had settled on - all the way back in '97. It took 8 years, but it looks like it'll finally happen. Waaaay too late for us to care. By then we'll be looking forward to the 980+, or possibly the 990.
Originally posted by BRussell
Notice how on the roadmap this new 7447/57 .13 chip has a max speed of 1.3 Ghz.
Apple already has a 1.42Ghz 7455. And on the roadmap the 7455 has a max of 1Ghz.
And at the top of the roadmap it says "Frequences listed are for Commercial spec." I've always heard this myth that "Apple overclocks" the recent G4s, and I'm sure that's not true, but perhaps they just don't make the 7455 above 1Ghz for the embedded market - only Apple.
There are several versions of the 7455:
P - 1.85V 65°C
S - 1.85V 75°C
L - 1.6V 105°C
N - 1.3V 105°C
xlr8yourmac listed the part number of a Giga Designs 1.4GHz upgrade to be XPC7455BRX1400PF. It's a 7455B at 1.4 GHz packed in a CBGA, version P, revision F (=3.3).
So Moto doesn't make them for Apple alone
Moto did only announce two versions of the 7457:
L - 1.3V 105°C
N - 1.1V 105°C