<strong>"64-bit-ness in itself does not mean "twice as fast",
THAT'S COMMON KNOWLEDGE.
</strong><hr></blockquote>
Then what exactly was the point of saying "Short of going 64 bit there is no way for Apple to level the playing field."?
[quote]<strong>"even though AMD would like to have you believe otherwise. As has already been pointed out numerous times on these boards, the cases where it will provide significant performance advantages over 32 bit systems are relatively few and not commonly encountered by most people."
It can, in fact mean the opposite -> half as fast. bandwidth requirements double. The G4 is... bandwidth constrained.
</strong><hr></blockquote>
Only pointers necessarily become larger, so bandwidth requirements don't really double. Also, the G4 is mainly bandwidth-constrained for tasks involving AltiVec or large amounts of FP anyway, so I don't think this would be have too much of an impact on performance.
[quote]<strong>32 bit addressing -> 4GB/thread max.</strong><hr></blockquote>
Actually, unless I'm mistaken, it should be 4GB per process rather than per thread.
And why do people keep saying "the cpu is cheap thus computer is cheap"
Can I interest you in a nice shinny 1ghz G4 cpu for your quicksilver 733?
A case itself can cost more then the CPU for idiots sake.
Surfice to say, I bet these are people who pay $50 an hour to compusa techs to install their ram.
The margin for a G4 is not really all that big by my guesses. the new mirror case itself would probably retail in the $100-200 arena from what you can see on the PC side.
Now that that's out of the way, 64 bit is nothing to blink and eye at for at least another 2-5yrs for consumers. Someone has to pioneer one onto the consumer market of course, but with computer depreciation as it is, give me a cheaper 32bit processing anyday.
The dual 887 have painted Apple in the corner. With OS X being the OS of choise also among the users 2x887 is even inclucluding some overhead about 1.5 GHz. So to replace it with a SP that CPU have to be at least a G4@1.5 GHz to not be slower then the predecessor!
For the topend the corresponding figure is 2.5 GHz. Considering it took about 3 years to get from 0.5 to 1 GHz a 2.5 around the coner sound like a pipe dream.
Assume that the next low end will be a 1x 1.5GHz and the top end a 2x 1.5 GHz how do you get a midrange model in there? A single 2 GHz? Not likely. A 2x1.0 -1.25 GHz that will be beaten in all classic and non SMP applications by the lowend version...
People complain about Motorola and the 500 MHz wall. IMHO depressing thing is not that that wall left the PM at half the clock speed of the Pentium/AMDs but that instead of catching up we have been sliding further back for years. The currently fastest shiping G4 is at 1 GHz and the P4 is at 2.8 GHz within spitting distance of being 3 times the clock speed of the G4.
If Apple do not solve the CPU problem next year at the latest we will go NEXT. That is Apple will become a marginal software company and no hardware at all.
<strong>The dual 887 have painted Apple in the corner. With OS X being the OS of choise also among the users 2x887 is even inclucluding some overhead about 1.5 GHz. So to replace it with a SP that CPU have to be at least a G4@1.5 GHz to not be slower then the predecessor!</strong><hr></blockquote>
First of all, the chip is at 867 MHz, not 887. Second, why do you think the dual 867 is being replaced by a single G4 or even a G4 at all? Do you honestly think Apple is that stupid?
[quote]<strong>Assume that the next low end will be a 1x 1.5GHz and the top end a 2x 1.5 GHz how do you get a midrange model in there? A single 2 GHz? Not likely. A 2x1.0 -1.25 GHz that will be beaten in all classic and non SMP applications by the lowend version...</strong><hr></blockquote>
I have a better idea. Let's assume you don't know what speed or type of chip or how many of them the next Power Macs will be using. Oh wait, that's reality. Sorry to burst your bubble there.
[quote]<strong>IMHO depressing thing is not that that wall left the PM at half the clock speed of the Pentium/AMDs but that instead of catching up we have been sliding further back for years. The currently fastest shiping G4 is at 1 GHz and the P4 is at 2.8 GHz within spitting distance of being 3 times the clock speed of the G4.</strong><hr></blockquote>
And if Apple used x86 chips in their Macs you might have had a point - but they don't, and you don't. I didn't think there was anybody left on this board that really believes PowerPC chips and x86 chips are directly comparable, MHz-wise... guess I was wrong. If you take nothing else away from this post, at least take this: MHz doesn't matter when you're comparing two entirely different chip architectures, because they're entirely different. Make sense now?
[quote]<strong>If Apple do not solve the CPU problem next year at the latest we will go NEXT. That is Apple will become a marginal software company and no hardware at all.</strong><hr></blockquote>
Okay, you can go sit in the corner with Michael Dell and the two of you can predict Apple's doom every 18 months. But unless you're severely learning impaired, you might have noticed at this point that Apple has not in fact gone out of business any time in the last 23 years, no matter how many times Dell said they would. But if you still want to bet on the old "Apple's going out of business", then be my guest.
<strong>Might find this an interesting read at O'Gradys.
Next-Generation Power4 Processor No-Go for Power Macs
Fri, 06 Sep 2002, 09:03
[Ed: The following was posted publicly by an Apple SciTech mailing list member; PowerPage.org has confirmed the source but the source has requested anonymity on our site. -PK]
I attended a briefing today on IBM's high performance computing technology, which is hinged on their Power4 CPU (this CPU has awesome performance in the various real-world benchmarks I have seen). After the briefing, I asked the presenter (a chief engineering manager from IBM) about the Power4 derivative for desktops and low end servers to be announced in October. You may recall that there has been speculation that this CPU would find its way into PowerMacs in the future. Well, it sounds like this CPU is not in Apple's future -- the "over 160" vector instructions are not AltiVec (even though AltiVec has 162 instructions), and there are technical issues that would prevent AltiVec from ever marrying with Power4 or its successors. Furthermore, the guy came right out and said that they have pitched the desktop Power4 to Apple, but Apple was not interested.
So, although Power4-based PowerMacs seemed like a promising (and likely) possibility, it looks like it won't happen. I guess we will have to wait and see what Apple has in store for the future . . . .
</strong><hr></blockquote>
idunno if i believe that. moto has already said they only wanna do embedded chips. so moto is out, apple would rather sell its soul to the devil than go x86 (but would if they had no other choice). if they turn down this chip from IBM what do they have left with?
also. i dont believe IBM would build a new fab facility that doubles their current fabbing ability just to make a CPU that is going to be developed for desktop computers and would ONLY be able to be used by apple, and have apple say no.
i think IBM would want a signed contract from steve before investing so much money into this new chip.
The dual 887 have painted Apple in the corner. With OS X being the OS of choise also among the users 2x887 is even inclucluding some overhead about 1.5 GHz. So to replace it with a SP that CPU have to be at least a G4@1.5 GHz to not be slower then the predecessor!
First of all, the chip is at 867 MHz, not 887. Second, why do you think the dual 867 is being replaced by a single G4 or even a G4 at all? Do you honestly think Apple is that stupid?
The very first line of this thread reads"According to Spymac the next itineration of Powermacs will revert back to a mixture of single processor and dual processor machines." My coments are in the context of previous statments in this thread
ssume that the next low end will be a 1x 1.5GHz and the top end a 2x 1.5 GHz how do you get a midrange model in there? A single 2 GHz? Not likely. A 2x1.0 -1.25 GHz that will be beaten in all classic and non SMP applications by the lowend version...
I have a better idea. Let's assume you don't know what speed or type of chip or how many of them the next Power Macs will be using. Oh wait, that's reality. Sorry to burst your bubble there.
If I claimed to know what CPU they would use it would be a bubble, it was a general argument about the problem of going from a all dual line to a mixed line up. So it seems that you burst something else...
IMHO depressing thing is not that that wall left the PM at half the clock speed of the Pentium/AMDs but that instead of catching up we have been sliding further back for years. The currently fastest shiping G4 is at 1 GHz and the P4 is at 2.8 GHz within spitting distance of being 3 times the clock speed of the G4.
And if Apple used x86 chips in their Macs you might have had a point - but they don't, and you don't. I didn't think there was anybody left on this board that really believes PowerPC chips and x86 chips are directly comparable, MHz-wise... guess I was wrong. If you take nothing else away from this post, at least take this: MHz doesn't matter when you're comparing two entirely different chip architectures, because they're entirely different. Make sense now?
Take a look at <a href="http://www.realworldtech.com/page.cfm?AID=RWT051400000000"; target="_blank">http://www.realworldtech.com/page.cfm?AID=RWT051400000000</a>; were the clock speed, interger and FPU of PPC and x86 during the period 1994-2000. For 1995-99 the clock speeds were quite close as was Integer and FPU performance with the PPC having 100-120% of the FPU performance of the x86 and the Integer in the 80-100% range. I do not claim that a 600 MHz pentium is 20% faster than a 500 G4, I claim that a G4 and a AMD or a P4 is not so vastly different in performance per clock cycle. Say that the most efficient is less than 50 % more efficent than the least efficent. So that the currently fastest G4 is then at most as fast as a 1.5 GHz P4. If you think a G4/400 is as fast as a P4 2.8 because it gives the same RC5 code cracking then it a religious belife and as such outside the realms or reason.
Lastly do not forget marketing, there nothing have to make sense it is all about perception and 1 GHz is percived as slower than 2.8 GHz
If Apple do not solve the CPU problem next year at the latest we will go NEXT. That is Apple will become a marginal software company and no hardware at all.
Okay, you can go sit in the corner with Michael Dell and the two of you can predict Apple's doom every 18 months. But unless you're severely learning impaired, you might have noticed at this point that Apple has not in fact gone out of business any time in the last 23 years, no matter how many times Dell said they would. But if you still want to bet on the old "Apple's going out of business", then be my guest.
Apple certified service tech
- Mac user since 1985
- All around Mac dork
[i]Back in 1992 when I bought my slow LCII instead of a much faster 386, I got a OS and GUI that was vastly superior and much more plug an play that stale cake DOS 5 with the Windows 3.1 as the icing! Like it or not but XP is much less behind X than 5.0/3.1 was 7.0.
Dell is a dealer Apple is a creator and inovator and have contributed vastly to the evolution of the personal computer, that is why I have such high expectatations on them. If I wuld buy a PC form Dell I my highest expectation is that it would be the same as everybody else at a slightly lower price...
Regarding my learning impairment and the general tone of your reply: If you ever become a mature human you will show respect to people that have other views than your own.
Comments
<strong>"64-bit-ness in itself does not mean "twice as fast",
THAT'S COMMON KNOWLEDGE.
</strong><hr></blockquote>
Then what exactly was the point of saying "Short of going 64 bit there is no way for Apple to level the playing field."?
[quote]<strong>"even though AMD would like to have you believe otherwise. As has already been pointed out numerous times on these boards, the cases where it will provide significant performance advantages over 32 bit systems are relatively few and not commonly encountered by most people."
Crap. Utter crap. Crap, crap, CRAP! Fluffy crap.</strong><hr></blockquote>
Do you actually read my posts before typing such insightful replies as your little crap orgy?
First the 32/64 bit issue is "COMMON KNOWLEDGE", and one paragraph later it's suddenly "Crap. Utter crap. Crap, crap, CRAP! Fluffy crap."? Huh?
[quote]<strong>
A 3.4 gig Xp Hammer will pound the beezeesus out of any G4 Motorola can drag out of its sorry ass in the next 18 months.</strong><hr></blockquote>
Reread my post and you'll find that I never said anything to the contrary.
[quote]<strong>'BYE'
</strong><hr></blockquote>
Uh, please, monkeying my sig, how much more childish can it get?
Bye,
RazzFazz
<strong>
It can, in fact mean the opposite -> half as fast. bandwidth requirements double. The G4 is... bandwidth constrained.
</strong><hr></blockquote>
Only pointers necessarily become larger, so bandwidth requirements don't really double. Also, the G4 is mainly bandwidth-constrained for tasks involving AltiVec or large amounts of FP anyway, so I don't think this would be have too much of an impact on performance.
[quote]<strong>32 bit addressing -> 4GB/thread max.</strong><hr></blockquote>
Actually, unless I'm mistaken, it should be 4GB per process rather than per thread.
Bye,
RazzFazz
<strong>The new powermacs are definitely catching up to wintel . . . circa 1999</strong><hr></blockquote>
HA HA HA HA HA!!!! <img src="graemlins/oyvey.gif" border="0" alt="[No]" />
Supplier information will always remain internal.
And why do people keep saying "the cpu is cheap thus computer is cheap"
Can I interest you in a nice shinny 1ghz G4 cpu for your quicksilver 733?
A case itself can cost more then the CPU for idiots sake.
Surfice to say, I bet these are people who pay $50 an hour to compusa techs to install their ram.
The margin for a G4 is not really all that big by my guesses. the new mirror case itself would probably retail in the $100-200 arena from what you can see on the PC side.
Now that that's out of the way, 64 bit is nothing to blink and eye at for at least another 2-5yrs for consumers. Someone has to pioneer one onto the consumer market of course, but with computer depreciation as it is, give me a cheaper 32bit processing anyday.
~Kuku
For the topend the corresponding figure is 2.5 GHz. Considering it took about 3 years to get from 0.5 to 1 GHz a 2.5 around the coner sound like a pipe dream.
Assume that the next low end will be a 1x 1.5GHz and the top end a 2x 1.5 GHz how do you get a midrange model in there? A single 2 GHz? Not likely. A 2x1.0 -1.25 GHz that will be beaten in all classic and non SMP applications by the lowend version...
People complain about Motorola and the 500 MHz wall. IMHO depressing thing is not that that wall left the PM at half the clock speed of the Pentium/AMDs but that instead of catching up we have been sliding further back for years. The currently fastest shiping G4 is at 1 GHz and the P4 is at 2.8 GHz within spitting distance of being 3 times the clock speed of the G4.
If Apple do not solve the CPU problem next year at the latest we will go NEXT. That is Apple will become a marginal software company and no hardware at all.
G-News
<strong>You guys just gotta relax and wait instead fo wail.
G-News</strong><hr></blockquote>
Wheres the fun in that?
YOU NEED TO RELAX CAUSE I"M RELAXED EVERYONE IS RELAXED!
<strong>
PS have we lost a large number of contributors to this forum? I have never known it to be so quiet.
[ 09-06-2002: Message edited by: Addison ]</strong><hr></blockquote>
Dunno about others, but I was unable to reach AI for over two weeks, it got lost in a circular route in the states.
<strong>
Dunno about others, but I was unable to reach AI for over two weeks, it got lost in a circular route in the states.</strong><hr></blockquote>
Same here.
<strong>The dual 887 have painted Apple in the corner. With OS X being the OS of choise also among the users 2x887 is even inclucluding some overhead about 1.5 GHz. So to replace it with a SP that CPU have to be at least a G4@1.5 GHz to not be slower then the predecessor!</strong><hr></blockquote>
First of all, the chip is at 867 MHz, not 887. Second, why do you think the dual 867 is being replaced by a single G4 or even a G4 at all? Do you honestly think Apple is that stupid?
[quote]<strong>Assume that the next low end will be a 1x 1.5GHz and the top end a 2x 1.5 GHz how do you get a midrange model in there? A single 2 GHz? Not likely. A 2x1.0 -1.25 GHz that will be beaten in all classic and non SMP applications by the lowend version...</strong><hr></blockquote>
I have a better idea. Let's assume you don't know what speed or type of chip or how many of them the next Power Macs will be using. Oh wait, that's reality. Sorry to burst your bubble there.
[quote]<strong>IMHO depressing thing is not that that wall left the PM at half the clock speed of the Pentium/AMDs but that instead of catching up we have been sliding further back for years. The currently fastest shiping G4 is at 1 GHz and the P4 is at 2.8 GHz within spitting distance of being 3 times the clock speed of the G4.</strong><hr></blockquote>
And if Apple used x86 chips in their Macs you might have had a point - but they don't, and you don't. I didn't think there was anybody left on this board that really believes PowerPC chips and x86 chips are directly comparable, MHz-wise... guess I was wrong. If you take nothing else away from this post, at least take this: MHz doesn't matter when you're comparing two entirely different chip architectures, because they're entirely different. Make sense now?
[quote]<strong>If Apple do not solve the CPU problem next year at the latest we will go NEXT. That is Apple will become a marginal software company and no hardware at all.</strong><hr></blockquote>
Okay, you can go sit in the corner with Michael Dell and the two of you can predict Apple's doom every 18 months. But unless you're severely learning impaired, you might have noticed at this point that Apple has not in fact gone out of business any time in the last 23 years, no matter how many times Dell said they would. But if you still want to bet on the old "Apple's going out of business", then be my guest.
<strong>Might find this an interesting read at O'Gradys.
Next-Generation Power4 Processor No-Go for Power Macs
Fri, 06 Sep 2002, 09:03
[Ed: The following was posted publicly by an Apple SciTech mailing list member; PowerPage.org has confirmed the source but the source has requested anonymity on our site. -PK]
I attended a briefing today on IBM's high performance computing technology, which is hinged on their Power4 CPU (this CPU has awesome performance in the various real-world benchmarks I have seen). After the briefing, I asked the presenter (a chief engineering manager from IBM) about the Power4 derivative for desktops and low end servers to be announced in October. You may recall that there has been speculation that this CPU would find its way into PowerMacs in the future. Well, it sounds like this CPU is not in Apple's future -- the "over 160" vector instructions are not AltiVec (even though AltiVec has 162 instructions), and there are technical issues that would prevent AltiVec from ever marrying with Power4 or its successors. Furthermore, the guy came right out and said that they have pitched the desktop Power4 to Apple, but Apple was not interested.
So, although Power4-based PowerMacs seemed like a promising (and likely) possibility, it looks like it won't happen. I guess we will have to wait and see what Apple has in store for the future . . . .
idunno if i believe that. moto has already said they only wanna do embedded chips. so moto is out, apple would rather sell its soul to the devil than go x86 (but would if they had no other choice). if they turn down this chip from IBM what do they have left with?
also. i dont believe IBM would build a new fab facility that doubles their current fabbing ability just to make a CPU that is going to be developed for desktop computers and would ONLY be able to be used by apple, and have apple say no.
i think IBM would want a signed contract from steve before investing so much money into this new chip.
Originally posted by DrBoar:
The dual 887 have painted Apple in the corner. With OS X being the OS of choise also among the users 2x887 is even inclucluding some overhead about 1.5 GHz. So to replace it with a SP that CPU have to be at least a G4@1.5 GHz to not be slower then the predecessor!
First of all, the chip is at 867 MHz, not 887. Second, why do you think the dual 867 is being replaced by a single G4 or even a G4 at all? Do you honestly think Apple is that stupid?
The very first line of this thread reads"According to Spymac the next itineration of Powermacs will revert back to a mixture of single processor and dual processor machines." My coments are in the context of previous statments in this thread
ssume that the next low end will be a 1x 1.5GHz and the top end a 2x 1.5 GHz how do you get a midrange model in there? A single 2 GHz? Not likely. A 2x1.0 -1.25 GHz that will be beaten in all classic and non SMP applications by the lowend version...
I have a better idea. Let's assume you don't know what speed or type of chip or how many of them the next Power Macs will be using. Oh wait, that's reality. Sorry to burst your bubble there.
If I claimed to know what CPU they would use it would be a bubble, it was a general argument about the problem of going from a all dual line to a mixed line up. So it seems that you burst something else...
IMHO depressing thing is not that that wall left the PM at half the clock speed of the Pentium/AMDs but that instead of catching up we have been sliding further back for years. The currently fastest shiping G4 is at 1 GHz and the P4 is at 2.8 GHz within spitting distance of being 3 times the clock speed of the G4.
And if Apple used x86 chips in their Macs you might have had a point - but they don't, and you don't. I didn't think there was anybody left on this board that really believes PowerPC chips and x86 chips are directly comparable, MHz-wise... guess I was wrong. If you take nothing else away from this post, at least take this: MHz doesn't matter when you're comparing two entirely different chip architectures, because they're entirely different. Make sense now?
Take a look at <a href="http://www.realworldtech.com/page.cfm?AID=RWT051400000000"; target="_blank">http://www.realworldtech.com/page.cfm?AID=RWT051400000000</a>; were the clock speed, interger and FPU of PPC and x86 during the period 1994-2000. For 1995-99 the clock speeds were quite close as was Integer and FPU performance with the PPC having 100-120% of the FPU performance of the x86 and the Integer in the 80-100% range. I do not claim that a 600 MHz pentium is 20% faster than a 500 G4, I claim that a G4 and a AMD or a P4 is not so vastly different in performance per clock cycle. Say that the most efficient is less than 50 % more efficent than the least efficent. So that the currently fastest G4 is then at most as fast as a 1.5 GHz P4. If you think a G4/400 is as fast as a P4 2.8 because it gives the same RC5 code cracking then it a religious belife and as such outside the realms or reason.
Lastly do not forget marketing, there nothing have to make sense it is all about perception and 1 GHz is percived as slower than 2.8 GHz
If Apple do not solve the CPU problem next year at the latest we will go NEXT. That is Apple will become a marginal software company and no hardware at all.
Okay, you can go sit in the corner with Michael Dell and the two of you can predict Apple's doom every 18 months. But unless you're severely learning impaired, you might have noticed at this point that Apple has not in fact gone out of business any time in the last 23 years, no matter how many times Dell said they would. But if you still want to bet on the old "Apple's going out of business", then be my guest.
Apple certified service tech
- Mac user since 1985
- All around Mac dork
[i]Back in 1992 when I bought my slow LCII instead of a much faster 386, I got a OS and GUI that was vastly superior and much more plug an play that stale cake DOS 5 with the Windows 3.1 as the icing! Like it or not but XP is much less behind X than 5.0/3.1 was 7.0.
Dell is a dealer Apple is a creator and inovator and have contributed vastly to the evolution of the personal computer, that is why I have such high expectatations on them. If I wuld buy a PC form Dell I my highest expectation is that it would be the same as everybody else at a slightly lower price...
Regarding my learning impairment and the general tone of your reply: If you ever become a mature human you will show respect to people that have other views than your own.
[ 09-17-2002: Message edited by: DrBoar ]</p>