Doubt it, thats where low-power PPCG4 and MIPS rule. IBM itself and Apple is probably going to be the two biggest customers of this Power4.
News.com is writing the chip, most likely, will be single-core. Thats not too god.
Will a Power4 be much faster ( Mhz for Mhz) compared to dual G4? I guess not for applications highly altivec-optimized. After all, the dual G4s got 2 set of Altivec-hardware.
On the other hand, the PPC was always held back by lack of memory bandwidth, and the new design clearly isnt.
<strong>I don't believe all the poppycock about agreements and IBM feeling slighted by Apple. Apple went with Moto 'cause IBM had nothing. </strong><hr></blockquote>
Actually when Motorola stalled at 500MHz IBM stepped in and said "Look we can fab it at higher speeds in decent yields for you". Motorola then blocked Apple from using them.
My guess is since then IBM has never truly attempted to overtake the G4 with the G3 in terms of clockspeeds as they never had a good reason to.
So I'm guilty of building speculation on a solid base of supposition, leavened with a healthy dose of rumor and equal parts guesswork and bald-faced fantasy. <img src="graemlins/embarrassed.gif" border="0" alt="[Embarrassed]" />
I actually hadn't remembered reading it in this thread as I was jumping around a lot and that bit stuck. Perhaps timortis could shed somemore light?
</strong><hr></blockquote>
Well, here is how it all went.
Last year, as people here with good memory may recall, Motorola and IBM both updated their PPC roadmaps, one after another. IBM's was particularly intriguing, because for the first time they were talking about a next-generation PowerPC chip with SIMD, dual-cores etc.
This made many people at the time, including me, comment that IBM had to be leveraging all the R&D they had put into Power4, which would only make sense. It was commented ad infinitum that ultimately Power4 technology would trickle down into desktop implementations and chips that Apple could use, since IBM needed to also update their own Unix workstation & low-end server line, which still used 300 Mhz 32 bit PowerPC chips.
A couple of months later, however, IBM made some announcements such as <a href="http://news.com.com/2100-1001-257421.html"; target="_blank">the one mentioned here by Cnet,</a> and made it known that these processors would still be low-power/low-cost embedded chips, which disappointed me and quite a few other people. That's what I was talking about.
Good ol News.com, better than any rumor site out there.</strong><hr></blockquote>
And this was released today along with The Registry article and the conference anouncement. Background on the new chip Apple will be using building up for their announcement on Tuesday.
I have to say this is probably the most believable story for the next Apple used chip in the last year and a half.
The PowerPC Architecture was derived from the Power Architecture. Removing what was considered unecessary instructions from Power to create PowerPC. By removing these instructions they could remove units from a chip to process code. (Obviously there is more to it than just this but you can read in detail on IBM's website if you feel the need).
So if a PowerPC chip is simply a lightweight version of a Power chip it shouldn't be a huge undertaking to make a Power 4 into a PowerPC 4. At the time PowerPC was created I think IBM had a Power 2, so if the Power 4 is the heavyweight chip of today a PowerPC derivative should be fairly impressive.
This of course could pan out to be a processor for something different, but there is so much pointing to it being the real deal.
Please remember that I'm making alot of generalizations here. I don't want to take the time to go in depth on Power vs PowerPC, instruction sets, units, etc.... If someone really wants I might invest an hour or 2 into a in depth explanation.
I'm inclined to think Motorola is done. There's no future. In the words of someone who knows Motorola in response to a question of "Do they have a future with Apple", "Doubt it". Steve can't be that dumb, can he now? Apple has the ball in its court to make killer hardware to go along with an operating system that already has many Windows users foaming at the mouth. I do have to take a step back when I read things like this though. The question may be one of time. Thinking back, this could also explain something I've been told. The G5 exists, but it is not the G5. It won't be called the G5. Maybe it will or it won't, but this could be the basis for an argument saying a new PowerMac based on the POWER4 cpu. The evidence certainly adds up. I realize that IBM has announced products well before they actually make their way into a system, but things can be different. Also, I believe IBM is "disclosing" the technical details on this new processor, which means yes it is possible we see them announced in Apple systems next week. If not, I think we get the last reincarnation of the G4. LetÕs face it, IBM can give us some sweet tech.
Very interesting, and now AI is buzzing all over again.
<strong>I'm inclined to think Motorola is done. There's no future. In the words of someone who knows Motorola in response to a question of "Do they have a future with Apple", "Doubt it". Steve can't be that dumb, can he now? Apple has the ball in its court to make killer hardware to go along with an operating system that already has many Windows users foaming at the mouth. I do have to take a step back when I read things like this though. The question may be one of time. Thinking back, this could also explain something I've been told. The G5 exists, but it is not the G5. It won't be called the G5. Maybe it will or it won't, but this could be the basis for an argument saying a new PowerMac based on the POWER4 cpu. The evidence certainly adds up. I realize that IBM has announced products well before they actually make their way into a system, but things can be different. Also, I believe IBM is "disclosing" the technical details on this new processor, which means yes it is possible we see them announced in Apple systems next week. If not, I think we get the last reincarnation of the G4. LetÕs face it, IBM can give us some sweet tech.
Very interesting, and now AI is buzzing all over again. </strong><hr></blockquote>
I don't think Apple's wish for secrecy will prevent IBM from discussing their latest technical achievements at the MDF in October. There is very little precedent for this actually -- many PowerPCs have been discussed at that conference well before Apple could introduce them. The chip probably won't make it to market until at least 6 months after that conference.
Whether or not there was any agreement is moot, as Apple wouldn't let the low end machines overtake the Pro ones. Also, IBM would probably be discouraged from upstaging Motorola with a faster G3, as this would damage the AIM alliance
What bus do the new G3s use? 60x? MPX? Something else?
Current G4 transistor count: 33M
[quote]Pull those x86 whoop-ass kickin' boots on...<hr></blockquote>
They've been in the cupboard since '97
Gamecube and PS3 using PowerPC CPUs, hmmm. There's even a PowerPC logo on the Gamecube box. Could we see the emergence of PowerPC as a brand?
I don't think Apple's wish for secrecy will prevent IBM from discussing their latest technical achievements at the MDF in October. There is very little precedent for this actually -- many PowerPCs have been discussed at that conference well before Apple could introduce them. The chip probably won't make it to market until at least 6 months after that conference.</strong><hr></blockquote>
I agree but how much will they discuss production dates?
An excellent post by Big Mac. I think he said it all there.
IF this 'Power4' lite is the chip we're getting...it has massive ramifications for Apple's 'Switch' campaign.
If Apple are going to be comparing their computers with x86 computers...customers will see a big difference if the potential for this chip is realised. A Power Lite. 64 bit. Altivec? Gonna give even a 4 gig Pentium 4 a real headache... Add to that supposed Apple / Nvidia cooperation on some multimedia chipset or graphic workstation powerhouse card...then...next year is shaping up to be Apple's year!
Hmmm. Just thought. If 10.3 is going to be the 64 bit OS...and it's launched late next year...I hope this doesn't mean we get the Power chip early 2004
IBM's got the bag of tricks. It's going after Sun in the Unix Server space...
Apple can benefit from this.
Apple tagging it's 'power'Mac to the tails of IBM's POWER chips is a shrewd move. ie we and they can see the future. IBM invest big in CPU tech'...and if the Power 5 and 6 come...there's a chance that one day we'll get trickle down from that too.
It gives the PPC desktop market a revelation of realisation. We can forget about Intels and AMDs...because THE next generation of PPCs are going to rock AND they have a fantastic road map. Apple secrecy or not...we just have to look at what IBM are doing with their Power line to find out what to look for in future Macs. And having read some of those links...their Power chips are getting better and better...which takes them well into the second half of this decade.
:eek:
It's not cut and dried yet. But if Apple go with IBM it gives them alot of security. IBM are a big player. Unlike Motorola, they aren't losing billions. I sure hope Apple and IBM can deliver by Summer next year at the latest. By the time AMD's troubled Sledgehammer gets to market...Apple will have their answer waiting...
If this chip is being made for Apple and seeing as how its not an embedded processor, wouldn't Apple get to announce models using it before IBM spews out technical details?<hr></blockquote>
I had a look at this last night because I was wondering the same thing. I think that basically Motorola had published the tech involved in the G4 for some time prior to announcement (mid 98), however the G4 itself (MPC7400) was not announced until about the same time as Apple unveiled the PowerMac G4 (sept '99). I reckon that Apple (and IBM/Motorola wrt Apple) are also more secretive now than they were even then.
Comments
Doubt it, thats where low-power PPCG4 and MIPS rule. IBM itself and Apple is probably going to be the two biggest customers of this Power4.
News.com is writing the chip, most likely, will be single-core. Thats not too god.
Will a Power4 be much faster ( Mhz for Mhz) compared to dual G4? I guess not for applications highly altivec-optimized. After all, the dual G4s got 2 set of Altivec-hardware.
On the other hand, the PPC was always held back by lack of memory bandwidth, and the new design clearly isnt.
<strong>I don't believe all the poppycock about agreements and IBM feeling slighted by Apple. Apple went with Moto 'cause IBM had nothing. </strong><hr></blockquote>
Actually when Motorola stalled at 500MHz IBM stepped in and said "Look we can fab it at higher speeds in decent yields for you". Motorola then blocked Apple from using them.
My guess is since then IBM has never truly attempted to overtake the G4 with the G3 in terms of clockspeeds as they never had a good reason to.
<strong>
So I'm guilty of building speculation on a solid base of supposition, leavened with a healthy dose of rumor and equal parts guesswork and bald-faced fantasy.
I actually hadn't remembered reading it in this thread as I was jumping around a lot and that bit stuck. Perhaps timortis could shed somemore light?
</strong><hr></blockquote>
Well, here is how it all went.
Last year, as people here with good memory may recall, Motorola and IBM both updated their PPC roadmaps, one after another. IBM's was particularly intriguing, because for the first time they were talking about a next-generation PowerPC chip with SIMD, dual-cores etc.
This made many people at the time, including me, comment that IBM had to be leveraging all the R&D they had put into Power4, which would only make sense. It was commented ad infinitum that ultimately Power4 technology would trickle down into desktop implementations and chips that Apple could use, since IBM needed to also update their own Unix workstation & low-end server line, which still used 300 Mhz 32 bit PowerPC chips.
A couple of months later, however, IBM made some announcements such as <a href="http://news.com.com/2100-1001-257421.html"; target="_blank">the one mentioned here by Cnet,</a> and made it known that these processors would still be low-power/low-cost embedded chips, which disappointed me and quite a few other people. That's what I was talking about.
[ 08-08-2002: Message edited by: timortis ]</p>
<strong>Check <a href="http://news.com.com/2100-1001-949030.html?tag=fd_top"; target="_blank">this</a> out! Note this quote:
Good ol News.com, better than any rumor site out there.</strong><hr></blockquote>
And this was released today along with The Registry article and the conference anouncement. Background on the new chip Apple will be using building up for their announcement on Tuesday.
The PowerPC Architecture was derived from the Power Architecture. Removing what was considered unecessary instructions from Power to create PowerPC. By removing these instructions they could remove units from a chip to process code. (Obviously there is more to it than just this but you can read in detail on IBM's website if you feel the need).
So if a PowerPC chip is simply a lightweight version of a Power chip it shouldn't be a huge undertaking to make a Power 4 into a PowerPC 4. At the time PowerPC was created I think IBM had a Power 2, so if the Power 4 is the heavyweight chip of today a PowerPC derivative should be fairly impressive.
This of course could pan out to be a processor for something different, but there is so much pointing to it being the real deal.
Please remember that I'm making alot of generalizations here. I don't want to take the time to go in depth on Power vs PowerPC, instruction sets, units, etc.... If someone really wants I might invest an hour or 2 into a in depth explanation.
Apple Dropping Moto like a bad case of the clap.
Apple looks to IBM for thier new "64-bit PowerPC Processor with 'more than 160' vector instructions"
The IBM chips have their debut at the microprocessor conference this OCT.
They announce the new Hardware in Jan or Feb 2003, along with the chipset they've been cooking up with nVidia.
Bill Gates takes a bite out of his Panama hat and Grimaces at the macworld SF '03 celebration.
Quake 3 framerates reach 12,000/s on my new IBM/PPC/G5. and I still get fragged twice as much as my brother playing on his B&W 350.
Even money - any takers?
Very interesting, and now AI is buzzing all over again.
<strong>
Well, here is how it all went.
*really great explanation snipped for space*
</strong><hr></blockquote>
Thanks, timortis!
<strong>I'm inclined to think Motorola is done. There's no future. In the words of someone who knows Motorola in response to a question of "Do they have a future with Apple", "Doubt it". Steve can't be that dumb, can he now? Apple has the ball in its court to make killer hardware to go along with an operating system that already has many Windows users foaming at the mouth. I do have to take a step back when I read things like this though. The question may be one of time. Thinking back, this could also explain something I've been told. The G5 exists, but it is not the G5. It won't be called the G5. Maybe it will or it won't, but this could be the basis for an argument saying a new PowerMac based on the POWER4 cpu. The evidence certainly adds up. I realize that IBM has announced products well before they actually make their way into a system, but things can be different. Also, I believe IBM is "disclosing" the technical details on this new processor, which means yes it is possible we see them announced in Apple systems next week. If not, I think we get the last reincarnation of the G4. LetÕs face it, IBM can give us some sweet tech.
Very interesting, and now AI is buzzing all over again.
I don't think Apple's wish for secrecy will prevent IBM from discussing their latest technical achievements at the MDF in October. There is very little precedent for this actually -- many PowerPCs have been discussed at that conference well before Apple could introduce them. The chip probably won't make it to market until at least 6 months after that conference.
See RandyB comment at bottom
<a href="http://arstechnica.infopop.net/OpenTopic/page?q=Y&a=tpc&s=50009562&f=8300945231&m=856098792 5&p=3" target="_blank">http://arstechnica.infopop.net/OpenTopic/page?q=Y&a=tpc&s=50009562&f=8300945231&m=856098792 5&p=3</a>
<strong>
When was it ?</strong><hr></blockquote>
About the end of 1997.
End of Line
What bus do the new G3s use? 60x? MPX? Something else?
Current G4 transistor count: 33M
[quote]Pull those x86 whoop-ass kickin' boots on...<hr></blockquote>
They've been in the cupboard since '97
Gamecube and PS3 using PowerPC CPUs, hmmm. There's even a PowerPC logo on the Gamecube box. Could we see the emergence of PowerPC as a brand?
<strong>
I don't think Apple's wish for secrecy will prevent IBM from discussing their latest technical achievements at the MDF in October. There is very little precedent for this actually -- many PowerPCs have been discussed at that conference well before Apple could introduce them. The chip probably won't make it to market until at least 6 months after that conference.</strong><hr></blockquote>
I agree but how much will they discuss production dates?
IF this 'Power4' lite is the chip we're getting...it has massive ramifications for Apple's 'Switch' campaign.
If Apple are going to be comparing their computers with x86 computers...customers will see a big difference if the potential for this chip is realised. A Power Lite. 64 bit. Altivec? Gonna give even a 4 gig Pentium 4 a real headache... Add to that supposed Apple / Nvidia cooperation on some multimedia chipset or graphic workstation powerhouse card...then...next year is shaping up to be Apple's year!
Hmmm. Just thought. If 10.3 is going to be the 64 bit OS...and it's launched late next year...I hope this doesn't mean we get the Power chip early 2004
IBM's got the bag of tricks. It's going after Sun in the Unix Server space...
Apple can benefit from this.
Apple tagging it's 'power'Mac to the tails of IBM's POWER chips is a shrewd move. ie we and they can see the future. IBM invest big in CPU tech'...and if the Power 5 and 6 come...there's a chance that one day we'll get trickle down from that too.
It gives the PPC desktop market a revelation of realisation. We can forget about Intels and AMDs...because THE next generation of PPCs are going to rock AND they have a fantastic road map. Apple secrecy or not...we just have to look at what IBM are doing with their Power line to find out what to look for in future Macs. And having read some of those links...their Power chips are getting better and better...which takes them well into the second half of this decade.
:eek:
It's not cut and dried yet. But if Apple go with IBM it gives them alot of security. IBM are a big player. Unlike Motorola, they aren't losing billions. I sure hope Apple and IBM can deliver by Summer next year at the latest. By the time AMD's troubled Sledgehammer gets to market...Apple will have their answer waiting...
Lemon Bon Bon
Instead of duh-doh-duh-duh...
...Apple could just let IBM use the Apple start up chord...!
PS3, featuring the latest PPC processor (DAHHHH!)
That would be a cool advert...
Lemon Bon Bon
<strong>
About the end of 1997.
End of Line</strong><hr></blockquote>
Thank you for the answer.
1997+5 years contract=2002 !!!
So 2003 will be AI year (not AppleInsider sorry, but AppleIBM
Aw
If this chip is being made for Apple and seeing as how its not an embedded processor, wouldn't Apple get to announce models using it before IBM spews out technical details?<hr></blockquote>
I had a look at this last night because I was wondering the same thing. I think that basically Motorola had published the tech involved in the G4 for some time prior to announcement (mid 98), however the G4 itself (MPC7400) was not announced until about the same time as Apple unveiled the PowerMac G4 (sept '99). I reckon that Apple (and IBM/Motorola wrt Apple) are also more secretive now than they were even then.
Does anyone remember more fully the timeline?
[ 08-09-2002: Message edited by: Krassy ]</p>
Looking at the roadmap you can see G5 is 85xx rapidIO, extensible architecture...
There are 2 85xx processors :
8540 (one year old)
8560 (PowerQUICC III not for us)
First intersting point the MIPS are the same with the 7455 : 2315@1GHz (8540) vs 2310@1GHz (7455)
The 8540 is based on Book E, has RapidIO DDR memory controller @ 333MHz, it's .13 CMOS no SOI, SIMD,... All the technologies are ther
Will new mac with the 8540 (or a derivative of it) and new MB be faster than the 7455 G4 PM ?
So, does this 8540 processor sound like the "G5" we will in the next weeks ?
<a href="http://e-www.motorola.com/webapp/sps/site/overview.jsp?nodeId=03M943030450467M983989030230"; target="_blank">roadmap</a>
<a href="http://e-www.motorola.com/webapp/sps/site/prod_summary.jsp?code=MPC8560&nodeId=03M9430304504 67M98657" target="_blank">8560 PwQUICC III</a>
<a href="http://e-www.motorola.com/webapp/sps/site/prod_summary.jsp?code=MPC8540&nodeId=03M9430304504 67M98657" target="_blank">8540</a>
Aw