If you're only interested in iTunes and email (no Web surfing?), you wouldn't be buying a PowerMac to begin with, unless you just like to throw your money around, so why did you even read this thread? Sounds like an iMac G5 is right up your alley.billyboy said:So, having read through this thread, have i got this right? the Chud on the SMT means dual dual processor freeserves for 3.5GHz PowermacBooks addressing 64GB of memory next April?
It cracks me up how the techno heads surface and flex their enormous mental muscles quoting FSBs on the backend diode at x KHz per tetra cycle newton.
In a nutshell, will the next set of Powermacs be faster than what I already have, and still be simple enough to switch on with a push of a button? If so, what date and how much? And where, and why does anyone just using iTunes and email need to care? Thanks.
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.
IBM PPC 970MP A Reality?
- Thread starter quagmire
- Start date
- Sort by reaction score
Just a few thoughts...
ummmmmm, I dont think comparing Pentium 4 EE to a 970 class processor is fair cuz arent they for two different markets (consumer vs server)? I mean from what I understand all the IBM processors were designed to work best in multiple processor systems... As far as I know there really isnt too many dual Pentum 4 systems (in fact i think its zero) but there are Xeons duals and strangely a single Xeon may be a bit faster than a 970 but when they are in multiple processor arrangements, for example, a dual 970 clocked by almost a GHZ down is equal or beats a Xeon dual station in over 50% of benchmarks I have seen so far as well as from personal experience and sometimes the performance difference is rather extremely significant... On the other hand, we can argue all we want but for the past couple of years, IBM AMD and Intel has released processors that when it comes down to it all are about equal to performance at the high end... I would like to see Apple adopt some kind of Cell derivative technology to really outclass the competition for a while which I think would have been good for the market because computer technology has become stagnant lately, all I see is increases of couple of MHZ in Graphics cards, memory increases, but no one is bold enough to really attack the market the big way... Apple is on a roll right now, if they play it right they will gain a significant marketshare at consumer level, people are tired of Windows and its problems, now if Apple would offer a computer that IS the fastest in the world along with OS X and a couple of games and programs such as Autodesks Autocad and such they would gain a lot but I am sure they know it lol I am just wondering what IBM promised Steve Jobs when they knew that the 3GHZ wasnt goint to be a reality last year, I mean IBM has to be somehow obligated to Apple because there must be some kind of binding agreements and/contracts that certain timeframes must be met and such... Somehow I have a feeling that 970MP wont be such a distinguishable processor BUT I do believe that the Power5 derivative (which to my knowledge has a lot of similarities as Cell including the multiple OS support as well as the same amount of integer units) will be the processor that will begin to distance Apple in front of the computer world...
ummmmmm, I dont think comparing Pentium 4 EE to a 970 class processor is fair cuz arent they for two different markets (consumer vs server)? I mean from what I understand all the IBM processors were designed to work best in multiple processor systems... As far as I know there really isnt too many dual Pentum 4 systems (in fact i think its zero) but there are Xeons duals and strangely a single Xeon may be a bit faster than a 970 but when they are in multiple processor arrangements, for example, a dual 970 clocked by almost a GHZ down is equal or beats a Xeon dual station in over 50% of benchmarks I have seen so far as well as from personal experience and sometimes the performance difference is rather extremely significant... On the other hand, we can argue all we want but for the past couple of years, IBM AMD and Intel has released processors that when it comes down to it all are about equal to performance at the high end... I would like to see Apple adopt some kind of Cell derivative technology to really outclass the competition for a while which I think would have been good for the market because computer technology has become stagnant lately, all I see is increases of couple of MHZ in Graphics cards, memory increases, but no one is bold enough to really attack the market the big way... Apple is on a roll right now, if they play it right they will gain a significant marketshare at consumer level, people are tired of Windows and its problems, now if Apple would offer a computer that IS the fastest in the world along with OS X and a couple of games and programs such as Autodesks Autocad and such they would gain a lot but I am sure they know it lol I am just wondering what IBM promised Steve Jobs when they knew that the 3GHZ wasnt goint to be a reality last year, I mean IBM has to be somehow obligated to Apple because there must be some kind of binding agreements and/contracts that certain timeframes must be met and such... Somehow I have a feeling that 970MP wont be such a distinguishable processor BUT I do believe that the Power5 derivative (which to my knowledge has a lot of similarities as Cell including the multiple OS support as well as the same amount of integer units) will be the processor that will begin to distance Apple in front of the computer world...
You're right, today there aren't any dual-core Pentiums or even dual-processor Pentium systems. However, that's going to change in a very big way over the next year. The dual-core Pentium EE is going to ship in the second quarter and it's even possible that we'll see dual-core Pentium systems before we get the dual-core 970MP Power Mac. Intel says that by the end of 2006 their product mix for both desktop and notebook processors will be something like 70% dual core. So, it is not unfair to compare the dual-core Pentium EE to what Apple may be shipping in the same time frame. And, note, we know with almost certainty that the dual-core Pentiums are going to ship before summer, but as yet it is only rumor as to when the 970MP will appear.
As you seem to be willing to indicate, the current dual Power Mac G5s are at rough parity with current dual Xeons. However, both the dual G5s and the dual Xeons are pretty expensive systems (and the Xeon is more of a server chip, not really a desktop system). But, dual-core processors are going to change the price/performance equation pretty quickly. We're not going to get twice the performance for the same money, but we're likely to see significantly better performance for about the same money or similar performance for even less money (if that makes any sense). Simply put, more bang for the buck.
When you recognize that the entire PC industry has only increased desktop processor performance by about 25% over the last two years you can understand why this coming shift could be important. In the next three months we could see high-end desktop PCs and Macs jump by over 25% in performance, in some cases it could even be greater than that.
Finally, I'm pretty certain that the dual-core Pentium EE will be targeted and will first appear as a high-end desktop system for enthusiasts and workstation type applications. And that's exactly what the Power Macs are designed for.
As you seem to be willing to indicate, the current dual Power Mac G5s are at rough parity with current dual Xeons. However, both the dual G5s and the dual Xeons are pretty expensive systems (and the Xeon is more of a server chip, not really a desktop system). But, dual-core processors are going to change the price/performance equation pretty quickly. We're not going to get twice the performance for the same money, but we're likely to see significantly better performance for about the same money or similar performance for even less money (if that makes any sense). Simply put, more bang for the buck.
When you recognize that the entire PC industry has only increased desktop processor performance by about 25% over the last two years you can understand why this coming shift could be important. In the next three months we could see high-end desktop PCs and Macs jump by over 25% in performance, in some cases it could even be greater than that.
Finally, I'm pretty certain that the dual-core Pentium EE will be targeted and will first appear as a high-end desktop system for enthusiasts and workstation type applications. And that's exactly what the Power Macs are designed for.
blitzkrieg79 said:
fpnc said:
Right on the money.
I would also add that dual core Itaniums are scheduled to release 3rd quarter 2005 with Tanglewood (up to 8 cores) coming in the second half of 2006.
Obviously only god will be able to afford it, but multicore processors are the future of computing.
billyboy said:So, having read through this thread, have i got this right? the Chud on the SMT means dual dual processor freeserves for 3.5GHz PowermacBooks addressing 64GB of memory next April?
It cracks me up how the techno heads surface and flex their enormous mental muscles quoting FSBs on the backend diode at x KHz per tetra cycle newton.
In a nutshell, will the next set of Powermacs be faster than what I already have, and still be simple enough to switch on with a push of a button? If so, what date and how much? And where, and why does anyone just using iTunes and email need to care? Thanks.
While I agree that this particular thread does appear to have been assimilated by the Borg, these sort of discussions CAN often be helpful in sorting possible from impossible, which can help with everyones buying decisions...
...as long as they eventually reach some sort of conclusion as opposed to endless techno jousting and measurebating.
This happened the last time a REV B PowerMac was rumored.
It cracks me up.
I just want a sub-two grand dual 2.0 G5.
It cracks me up.
I just want a sub-two grand dual 2.0 G5.
wdlove said:
No, not likely. Performance will be substantially better but not by a factor of two.
Sun is getting 70-80% more out of their dual core UltraSPARC IV, compared to the older single core, at the same frequency. It does, of course, depend on the load you are running. In Sun's case, the measurements are all against enterprise server applications, i.e. Oracle DB, Oracle Financials, SAP, PeopleSoft, etc. All stuff that has been written to take full advantage of large SMP machines with anywhere up to 72 CPUs in a single box. Your mileage on the desktop may varyfpnc said:
billyboy said:So, having read through this thread, have i got this right? the Chud on the SMT means dual dual processor freeserves for 3.5GHz PowermacBooks addressing 64GB of memory next April?
It cracks me up how the techno heads surface and flex their enormous mental muscles quoting FSBs on the backend diode at x KHz per tetra cycle newton.
In a nutshell, will the next set of Powermacs be faster than what I already have, and still be simple enough to switch on with a push of a button? If so, what date and how much? And where, and why does anyone just using iTunes and email need to care? Thanks.
You shouldn't belittle technologists. Without technology, and the people prepared to make the effort to understand it, we would all be sitting in a cave somewhere cold and hungry.
daveL said:Sun is getting 70-80% more out of their dual core UltraSPARC IV, compared to the older single core, at the same frequency. It does, of course, depend on the load you are running. In Sun's case, the measurements are all against enterprise server applications, i.e. Oracle DB, Oracle Financials, SAP, PeopleSoft, etc. All stuff that has been written to take full advantage of large SMP machines with anywhere up to 72 CPUs in a single box. Your mileage on the desktop may vary
I can believe that's possible under certain conditions and as I previously posted Intel is saying the following about their new dual-core Pentium EE:
I suspect, however, that even those 50-65% values are pretty special case and they most likely used highly-optimized, multi-threaded application to get those kinds of numbers. One area where you most likely won't see big improvements is in game play. There you are sometimes bound by the graphics card and driver performance or by the extent that the software developers have multi-threaded the game (which in many cases is not very much or well). But, as dual-core chips become the norm more and more software developers will begin to optimize their games for multi-threaded environments. So, over time game performance might actually get noticeably better simply as a result of threading-related software changes.
The Pentium 4 EE is basically just a P4 with a BIG cache.
The 970 use separate front side busses for each processor, whereas the P4 must share the front side bus between processors.
So it makes sense that Intel would dual core the EE version because it beats on the FSB less.
The P4 EE is not a cheap processor. So dual dual P4 EE will really beat the ^%& out of the FSB.
The 970MP dual core has more cache (good). Both cores will share the FSB, the nature of a dual core. Architecture advantage to the 970.
AMD is even better by putting the memory controller on chip. Low latency, I think there might be cache snoop, coherency issues.
The 970 use separate front side busses for each processor, whereas the P4 must share the front side bus between processors.
So it makes sense that Intel would dual core the EE version because it beats on the FSB less.
The P4 EE is not a cheap processor. So dual dual P4 EE will really beat the ^%& out of the FSB.
The 970MP dual core has more cache (good). Both cores will share the FSB, the nature of a dual core. Architecture advantage to the 970.
AMD is even better by putting the memory controller on chip. Low latency, I think there might be cache snoop, coherency issues.
wizard said:
I know that. But it keeps the temp down. With temperature, there is heat. So in the lappy dual core G5 and single core G5, use similar diodes and set it to a lower temperature resulting in less heat. I have been wrong though.
The byproduct of processing is heat. If you don't get rid of the heat, the temperature goes up. So to keep the temperature down, you need to keep the generate heat down, so this means you need to keep the processing down. Bummer, but ...
If you under clock a 970MP, this will reduce the heat generated. Since the clock is running slower, the core voltage can be reduced and this brings the heat even lower. Perhaps they could even run the FSB at the same rate as the CPU.
If you under clock a 970MP, this will reduce the heat generated. Since the clock is running slower, the core voltage can be reduced and this brings the heat even lower. Perhaps they could even run the FSB at the same rate as the CPU.
Ibm has now blocked the page its now says
http://www-306.ibm.com/chips/techlib/techlib.nsf/techdocs/7ADD0446E94B93ED87256FC30083B0A9
Our Apologies
You have attempted to enter a controlled access area that you are not currently entitled to view. IBM Microelectronics uses a entitlement process to protect potentially sensitive data.
Your entitlement is not an all or nothing proposition. Each user has a specific set of entitlements associated with their user profile. IBM employees, Business Partners and Customers may all have differing levels of entitlement. For example, all IBM employees would not have a blanket entitlement to all data. The IBM employee may only be entitled to information relating to their specific job responsibilities. This same principle applies to all users of this site.
Data accessible from this site may relate to IBM, IBM Business Partners, and Customers, and we take the responsibility for protecting that data very seriously. If you feel that you qualify for additional entitlements, please contact your IBM Representative or other IBM Contact person to secure additional entitlements. Thank you for your understanding of our privacy concerns, and again, we apologize for any inconvenience this may have caused you.
HTTP Web Server: Lotus Notes Exception - Document has been deleted.
The posting on IBM's server should never have happened, they did not want us to know about the 970MP.
http://www-306.ibm.com/chips/techlib/techlib.nsf/techdocs/7ADD0446E94B93ED87256FC30083B0A9
Our Apologies
You have attempted to enter a controlled access area that you are not currently entitled to view. IBM Microelectronics uses a entitlement process to protect potentially sensitive data.
Your entitlement is not an all or nothing proposition. Each user has a specific set of entitlements associated with their user profile. IBM employees, Business Partners and Customers may all have differing levels of entitlement. For example, all IBM employees would not have a blanket entitlement to all data. The IBM employee may only be entitled to information relating to their specific job responsibilities. This same principle applies to all users of this site.
Data accessible from this site may relate to IBM, IBM Business Partners, and Customers, and we take the responsibility for protecting that data very seriously. If you feel that you qualify for additional entitlements, please contact your IBM Representative or other IBM Contact person to secure additional entitlements. Thank you for your understanding of our privacy concerns, and again, we apologize for any inconvenience this may have caused you.
HTTP Web Server: Lotus Notes Exception - Document has been deleted.
The posting on IBM's server should never have happened, they did not want us to know about the 970MP.
mvc said:While I agree that this particular thread does appear to have been assimilated by the Borg, these sort of discussions CAN often be helpful in sorting possible from impossible, which can help with everyones buying decisions...
...as long as they eventually reach some sort of conclusion as opposed to endless techno jousting and measurebating.
Sorry if my British cynicism offended the DaveL s of this thread. I actually have and need a G5 Powermac as I have it pretty much smoking with the work I do on it, and a lo tech iMac G5 just wouldnt cut it. So with a few of these feeding frenzy threads under my belt, gradually (or is it eventually?) the brain boxes have made sense of Ghz and FSBs and what not to me, so I did know pretty much what I was buying into when I stomped up for the model of PowerMac I chose.
Thanks technology gurus, keep up the informative entertainment, but please, some people need to cut down on their distortion reality field medication.
wizard said:
Yes, Dave, but they do allow you to cut the processor speed which will cool the core. If each core could have three processors speeds as they do now (halted, half, full speed) and there are two cores then you have 2 x 3 = 6 possible processor speeds and thus 6 possible power consumption levels and thus 6 possible heat generation levels.* You also have a fan and possibly liquid cooling to reduce heat levels but which you don't want running all the time for noise and power consumption reasons.
If the chip is over heating, as measured on the thermal diode, then you turn on the fan and start lowering the power consumption by cutting the speed of the cores one at a time. If the heat is too great you end up turning off the machine, all automatically, until it cools sufficiently.
-Walter
*Hmm... six heat generation levels... Now we're cooking! Think, PowerBook's with hot plates for keeping your coffee warm, heating up a cup of soup, frying an egg! Wow! The next great thing! I wan't my G5 PowerBook Kitchenette!
fpnc said:
You are right in what u stated but I think I should have been more clear, when I compared the Pentium 4 EE and 970 I mean the current versions not the future ones because I dont really know what they really will be... And as far as who will bring the first dual core to the consumer market if I had to bet I would bet on everyone BUT Intel considering they lost the 1GHZ race and the 64BIT race, reality is that AMD and IBM have been more innovative companies for the past 2-3 year or so, all Intel did was slap more cache on to the chip and thats about it (and that HT stuff doesnt really work as well to gain any noticable difference in average performance, and Pentium M is basically a modified Pentium 3) where AMD and IBM introduced totally new cores... I know that for the past year or so the processor desktop performance was increased by some 25% and that this dual core thing has a potential to jump over 25% in the next few months BUT 25% is hardly considered a computer revolution (more of evolution) if u ask me, I am no expert but somehow I have a feeling that actually the so called Power5 derivative (as I mentioned in my earlier posts has a lot of similarities with the CELL such as 4 floating point and 4 integer units plus the main processor kind of give u nine cores, multiple OS support, and more parallizated structure) might be the revolutionary processor that Apple will distance it self in the front, I know other companies are not sleeping but if IBM and Apple was able to quickly develop under pressure a Power 4 derivative (970 series) to battle the PC market(and that processor keeps up with all the current high ends from other companies), the Power 5 and Power 5 derivative is supposedly developed simultaneously from scratch with Apple in mind so I can only imagine what the engineers from Apple and IBM can do without any pressure at all, and as I mentioned earlier, IBM has to be somewhat obligated to Apple for that 3GHZ fiasco and Steve Jobs has been awfully quiet about it where in Motorola years all he was doing was cursing and complaining about it, I know IBM had to promise something rather big so Steve wouldnt be so outspoken and give IBM a black eye...
