"Apple said today that a three-gigahertz version of the G5 chip would be available within 12 months." see nytimes.com
That'll be good for cheapskates like me who can't shell out the bucks now but looking forward to it in the future. Apple revamping the line quickly means the low end improves drastically.
Bye Motorola, don't let the door smack ya in the ass.
Jobs just said this to contrast the G5 introduction with the G4 introduction. It instills confidence in the platform among developers and switchers, and as a bonus it's a nice bitch-slap to Motorola.
Considering that the rest of the industry doubles performance every 18 months, a 50% increase in 12 months isn't exactly good news. Why the "wow"?
Well, I put money on Steve Jobs giving us a conservative estimate with the "3GHz in 12 months" statement. Also, in the x86 world it is more like 2x in 18-24 months. Since the 970 matches and exceeds the P4 and Xeons in the test results given I think we can take from Mr Job's statement that the product line is not going to fall way behind again.
Considering that the rest of the industry doubles performance every 18 months, a 50% increase in 12 months isn't exactly good news. Why the "wow"?
That statement refers to process technologies not performance and it's closer to 2 years anyway.
On the topic of scaling as it happens the PPC970 will scale faster than the Athlon 64 despite the fact the Athlon 64 is going to launch at virtually the same frequency.
i believe its double the transistors every 18 months no?
It's not really set in stone, but that's about right.
And as far as jumping from 2 to 3 GHz (+50%) in 12 months is concerned...
Since AMD can't keep up, Intel has gotten lazy. They only plan on reaching 3.4 GHz by the end of the year...whereas they released 2.53 GHz P4s in May 2002. A 34% jump in ~18 months.
AMD has been even worse off. They 've increased clockspeeds by 32% since January 2002, going from 1.67 GHz to 2.2 GHz. They briefly had a 2.25 GHz chip but they couldn't make an appreciable number of them.
Just for the record, it's about transistor count, not MHz. Also, there is a theoretical slowing in the rate of progress as designng and fabbing even tinier bits gets harder and harder and more expensive......of course, every time someone predicts a plateau, some smart people figure out how to beat it.....
Someone please clarify, but I am not sure that overall performance will scale linearly when both the processor and the bus take a 50% increase.
Well, no, performance won't scale lineraly. Hopefully Apple will tweak and tune the system to get as much performance as they can out of the computer. So the total performance will be (much) greater but not at the same level the clock scaling would indicate.
To get much better performance at that clock speed they would need to adjust the architecture of the chip: give it larger caches (which I suspect they will), greater execution resources (won't) or add something like SMT which uses idle resources much more efficiently (coming in the 980).
Comments
Originally posted by burningwheel
"Apple said today that a three-gigahertz version of the G5 chip would be available within 12 months." see nytimes.com
That'll be good for cheapskates like me who can't shell out the bucks now but looking forward to it in the future. Apple revamping the line quickly means the low end improves drastically.
Originally posted by Booga
Considering that the rest of the industry doubles performance every 18 months, a 50% increase in 12 months isn't exactly good news. Why the "wow"?
What do you mean by "rest of industry?"
Consider Intel:
January 2002: 2.2 Ghz
18 months later
June 2003: 3.2 Ghz
Where's the doubling?
(I'm going straight by Mhz, which of course is not accurate, but that's how you're comparing, so...)
Originally posted by Booga
Considering that the rest of the industry doubles performance every 18 months, a 50% increase in 12 months isn't exactly good news. Why the "wow"?
i believe its double the transistors every 18 months no?
Originally posted by applenut
i believe its double the transistors every 18 months no?
Trolls eat lots of transistors.
Jobs just said this to contrast the G5 introduction with the G4 introduction. It instills confidence in the platform among developers and switchers, and as a bonus it's a nice bitch-slap to Motorola.
"We have the new, all powerful G4. it's shipping at 550Mhz!!!"
ok, make that 500Mhz. for hte next ****ING YEAR. (or was it longer?)
in any case, the G4 initially got stuck a LONG time at its initial speeds. this would be to remove that fear with this new chip, IMO.
Originally posted by Booga
Considering that the rest of the industry doubles performance every 18 months, a 50% increase in 12 months isn't exactly good news. Why the "wow"?
Well, I put money on Steve Jobs giving us a conservative estimate with the "3GHz in 12 months" statement. Also, in the x86 world it is more like 2x in 18-24 months. Since the 970 matches and exceeds the P4 and Xeons in the test results given I think we can take from Mr Job's statement that the product line is not going to fall way behind again.
MM
The g6? 4ghz+? The future is bright indeed.
Originally posted by Booga
Considering that the rest of the industry doubles performance every 18 months, a 50% increase in 12 months isn't exactly good news. Why the "wow"?
That statement refers to process technologies not performance and it's closer to 2 years anyway.
On the topic of scaling as it happens the PPC970 will scale faster than the Athlon 64 despite the fact the Athlon 64 is going to launch at virtually the same frequency.
Originally posted by applenut
i believe its double the transistors every 18 months no?
It's not really set in stone, but that's about right.
And as far as jumping from 2 to 3 GHz (+50%) in 12 months is concerned...
Since AMD can't keep up, Intel has gotten lazy. They only plan on reaching 3.4 GHz by the end of the year...whereas they released 2.53 GHz P4s in May 2002. A 34% jump in ~18 months.
AMD has been even worse off. They 've increased clockspeeds by 32% since January 2002, going from 1.67 GHz to 2.2 GHz. They briefly had a 2.25 GHz chip but they couldn't make an appreciable number of them.
Originally posted by PB
Someone please clarify, but I am not sure that overall performance will scale linearly when both the processor and the bus take a 50% increase.
Well, no, performance won't scale lineraly. Hopefully Apple will tweak and tune the system to get as much performance as they can out of the computer. So the total performance will be (much) greater but not at the same level the clock scaling would indicate.
To get much better performance at that clock speed they would need to adjust the architecture of the chip: give it larger caches (which I suspect they will), greater execution resources (won't) or add something like SMT which uses idle resources much more efficiently (coming in the 980).