Going to 2.4 is a 20% increase in speed. The question is - does that relate to a 20% increase in performance? I've read that a 10% speed increase on a Pentium results in only a 2 - 3% increase in performance. How will the G5 compare?
To me the upgrade is really about moving tot he 90 nm G5's and I have a feeling that has required a bit of engineering, plus probably some revisions to Panther - read 10.3.3. The other issue is will this engineering work also cover the next revision this summer when the magic 3.0 is reached?
As for Steve's comment on hitting 3.0 in a year, remember that an IBM VP came on stage and said 3.0 in a year. IBM's commitment for 12 months is rather impressive and, I think, dependable.
As for Steve's comment on hitting 3.0 in a year, remember that an IBM VP came on stage and said 3.0 in a year. IBM's commitment for 12 months is rather impressive and, I think, dependable.
I gotta agree here. I am thinking, again, that Apple and IBM have some kind of deal where Apple gets to announce the faster chips. Maybe they let IBM tell about the 2.4Ghz chips but not the faster ones???
I gotta agree here. I am thinking, again, that Apple and IBM have some kind of deal where Apple gets to announce the faster chips. Maybe they let IBM tell about the 2.4Ghz chips but not the faster ones???
Apple has learned theri lesson, and IBM is too smart (and conservative) to let Steve announce they will be at 3 Ghz without being almost absolutely sure it will happen. Really I wouldn't be surprised if they made it to (or at least announced) 3.2 or higher in that time frame.
So... when will revisions be made? When 10.3.3 is out?
Going to 2.4 is a 20% increase in speed. The question is - does that relate to a 20% increase in performance? I've read that a 10% speed increase on a Pentium results in only a 2 - 3% increase in performance. How will the G5 compare?
Chances are that the 970FX will perform better than the 970 at the same clock speed thanks to the SSOI technology. I am not sure though how performance is affected here by clock speed scaling.
Chances are that the 970FX will perform better than the 970 at the same clock speed thanks to the SSOI technology. I am not sure though how performance is affected here by clock speed scaling.
No, the 970FX will not perform better per clock because of SSOI. SSOI enables it to run more efficiently - at lower power at the same frequency. To predict how a faster G5 will run compared to a slower, just check out benchmarks comparing the dual 1.8 and 2.0, as well as single 1.8 and single 1.6.
Everything indicates that performance scales pretty well with clock speed increases, for the 970(fx).
The observation made in 1965 by Gordon Moore, co-founder of Intel, that the number of transistors per square inch on integrated circuits had doubled every year since the integrated circuit was invented.
I spoke to an Apple "insider" today. He told me that Apple WILL release new G5s the week of March 15th-19th (on Mon or Tues probably). I know I, know, you dont believe me, but...
I spoke to an Apple "insider" today. He told me that Apple WILL release new G5s the week of March 15th-19th (on Mon or Tues probably). I know I, know, you dont believe me, but...
Ya but this is what makes the week bearable, knowing that another exciting announcement from apple is just around the corner.
I spoke to an Apple "insider" today. He told me that Apple WILL release new G5s the week of March 15th-19th (on Mon or Tues probably). I know I, know, you dont believe me, but...
I think this why the online stores went down Friday evening (EST). They were testing for the update on Monday or Tuesday. This also coincides with the 10.3.3 update, which is said to be almost complete, and probably required for the new Power Macs, etc.
I remember when apple went from 33mhz to 66mhz on the 68040.
Well, I can remember when Apple Marketing decided to retroactively double the 040 Megahertz rating for no good technical reason. But I don't think Apple ever shipped a 66Mhz 040. Only 33Mhz and 40Mhz (Q840AV).
Comments
To me the upgrade is really about moving tot he 90 nm G5's and I have a feeling that has required a bit of engineering, plus probably some revisions to Panther - read 10.3.3. The other issue is will this engineering work also cover the next revision this summer when the magic 3.0 is reached?
As for Steve's comment on hitting 3.0 in a year, remember that an IBM VP came on stage and said 3.0 in a year. IBM's commitment for 12 months is rather impressive and, I think, dependable.
Originally posted by kenaustus
As for Steve's comment on hitting 3.0 in a year, remember that an IBM VP came on stage and said 3.0 in a year. IBM's commitment for 12 months is rather impressive and, I think, dependable.
I gotta agree here. I am thinking, again, that Apple and IBM have some kind of deal where Apple gets to announce the faster chips. Maybe they let IBM tell about the 2.4Ghz chips but not the faster ones???
Originally posted by tfworld
I gotta agree here. I am thinking, again, that Apple and IBM have some kind of deal where Apple gets to announce the faster chips. Maybe they let IBM tell about the 2.4Ghz chips but not the faster ones???
Apple has learned theri lesson, and IBM is too smart (and conservative) to let Steve announce they will be at 3 Ghz without being almost absolutely sure it will happen. Really I wouldn't be surprised if they made it to (or at least announced) 3.2 or higher in that time frame.
So... when will revisions be made? When 10.3.3 is out?
Originally posted by Jubelum
How does that affect AppleCare coverage?
iDon'tCare
Originally posted by kenaustus
Going to 2.4 is a 20% increase in speed. The question is - does that relate to a 20% increase in performance? I've read that a 10% speed increase on a Pentium results in only a 2 - 3% increase in performance. How will the G5 compare?
Chances are that the 970FX will perform better than the 970 at the same clock speed thanks to the SSOI technology. I am not sure though how performance is affected here by clock speed scaling.
Originally posted by PB
Chances are that the 970FX will perform better than the 970 at the same clock speed thanks to the SSOI technology. I am not sure though how performance is affected here by clock speed scaling.
No, the 970FX will not perform better per clock because of SSOI. SSOI enables it to run more efficiently - at lower power at the same frequency. To predict how a faster G5 will run compared to a slower, just check out benchmarks comparing the dual 1.8 and 2.0, as well as single 1.8 and single 1.6.
Everything indicates that performance scales pretty well with clock speed increases, for the 970(fx).
Originally posted by cowerd
What does that mean? The size of technology doubles? The weight of technology doubles? The number of technologies double? What is a technology?
tech·nol·o·gy ( P ) Pronunciation Key (tk-nl-j)
n. pl. tech·nol·o·gies
1.
a) The application of science, especially to industrial or commercial objectives.
b) The scientific method and material used to achieve a commercial or industrial objective.
2. Electronic or digital products and systems considered as a group: a store specializing in office technology.
That help?
That help?
No, however this may help you:
Moore's Law
The observation made in 1965 by Gordon Moore, co-founder of Intel, that the number of transistors per square inch on integrated circuits had doubled every year since the integrated circuit was invented.
Originally posted by cowerd
No, however this may help you:
Is that still true today? I know intel is around 100 million... g5 is around 58 mill?
Originally posted by emig647
Is that still true today? I know intel is around 100 million... g5 is around 58 mill?
May be you missed per square inch ?
Moore's Law is about the density of transistors on a chip not the quantity!
But to your question: yes, sometimes they are behind and sometimes ahead of Moore's prediction. Chip technology advances in leaps.
For Motorola it was quantum leaps
And quantum leaps will be the end of Moore's Law...
Originally posted by dstranathan
I spoke to an Apple "insider" today. He told me that Apple WILL release new G5s the week of March 15th-19th (on Mon or Tues probably). I know I, know, you dont believe me, but...
Ya but this is what makes the week bearable, knowing that another exciting announcement from apple is just around the corner.
Originally posted by dstranathan
I spoke to an Apple "insider" today. He told me that Apple WILL release new G5s the week of March 15th-19th (on Mon or Tues probably). I know I, know, you dont believe me, but...
I think this why the online stores went down Friday evening (EST). They were testing for the update on Monday or Tuesday. This also coincides with the 10.3.3 update, which is said to be almost complete, and probably required for the new Power Macs, etc.
Originally posted by Beige_G3
Hmmmm....If Apple announces new machines this week how long until they actually ship one?
Well lets see... we'll do the math.
The XServes still haven't shipped. They were announced on Jan 6th. They have a shipping prediction of April 4th. We'll just call that 3 months even.
The PowerMac sells about 6 times more then the XServe... soo
3*6 = 18 months.
Assuming the PowerMac is released next week... we're looking at: Sept 23rd '05
Thats assuming that some college like VT doesn't buy all of them
Originally posted by emig647
Assuming the PowerMac is released next week... we're looking at: Sept 23rd '05
That'll be a nice b'day present for me, finally getting the computer I've ordered
Originally posted by emig647
I remember when apple went from 33mhz to 66mhz on the 68040.
Well, I can remember when Apple Marketing decided to retroactively double the 040 Megahertz rating for no good technical reason. But I don't think Apple ever shipped a 66Mhz 040. Only 33Mhz and 40Mhz (Q840AV).