The G4 PPCs have AltiVec as well, so I don't see how that would make the difference in comparison to G5s. All I know is that my 400-MHz G4s seem to get through WUs in a similar time to Intel & AMD hosts with several times the clock speed—or, as is more often the case, two or three times as long as other systems with seven or eight times the MHz rating.
that's where you're partially wrong....
G4s have better Altivec instructions than G5s, and with the latest-yet-unofficial x86 optimisations, Intel & AMD hosts are able to trump the G5s of similar clock speeds at an incredible rate.
The G4 PPCs have AltiVec as well, so I don't see how that would make the difference in comparison to G5s. All I know is that my 400-MHz G4s seem to get through WUs in a similar time to Intel & AMD hosts with several times the clock speed—or, as is more often the case, two or three times as long as other systems with seven or eight times the MHz rating.
that's where you're partially wrong....
G4s have better Altivec instructions than G5s, and with the latest-yet-unofficial x86 optimisations, Intel & AMD hosts are able to trump the G5s of similar clock speeds at an incredible rate.
Sorry, I don't follow: what's wrong in what I said? I was countering a claim that only G5s benefit from AltiVec optimization, and you seem to be going even further by stating that they benefit less than G4s.
Sorry, I don't follow: what's wrong in what I said? I was countering a claim that only G5s benefit from AltiVec optimization, and you seem to be going even further by stating that they benefit less than G4s.
[pre]
Time (h) CPU Benchmark (MFlops)
7.6 Intel/1.7GHz 862
3.8 AMD Athlon 1216
12.7 G4/400MHz 267[/pre]
I'd be wrong to conclude my Mac is doing any better, relative to the others, than its speed rating or benchmark would suggest?
Sorry to have you confused.... I was merely implying that G4 Altivec stands a better instruction than a G5 if not equal, because of different processor manufacturers.... but then again this will no longer be the case as for the past weeks x86 optimisations have been rampant.... with those excessive optimisation and benchmarking, even a Quad G5 got "smoked" by an AMD X2.... Imagine if such optimisation goes public, plus with participants taking advantages of the optimised BOINC client superbenches. And Bruce (Einstein founder, correct me if I'm wrong) had announced that he has already got those optimisations implemented into new science apps pending official release.
For the past months, Altivec-optimised hosts have had an edge over other host systems minus excessive benchmarking, but from what I've seen on the x86 optimisation, excessive benchmarking alongside the considerately-improved processing time is why this thread is taking shape.
Odysseus, you might want to edit that link, while there's still time.
Whoops ... too late to edit. It was supposed to go to WU No.5347493. I don't know why I put the URL in quotes, unless it was mental leakage from working on some web pages recently:
"I am tired of all this sort of thing called science here... We have spent
millions in that sort of thing for the last few years, and it is time it
should be stopped."
-- Simon Cameron, U.S. Senator, on the Smithsonian Institute, 1901.
RE: The G4 PPCs have
)
that's where you're partially wrong....
G4s have better Altivec instructions than G5s, and with the latest-yet-unofficial x86 optimisations, Intel & AMD hosts are able to trump the G5s of similar clock speeds at an incredible rate.
RE: RE: The G4 PPCs
)
Sorry, I don't follow: what's wrong in what I said? I was countering a claim that only G5s benefit from AltiVec optimization, and you seem to be going even further by stating that they benefit less than G4s.
Or are you saying that when I look at [url="http://einsteinathome.org/workunit/5347493"]this WU[/url], for example, selected at random:
[pre]
Time (h) CPU Benchmark (MFlops)
7.6 Intel/1.7GHz 862
3.8 AMD Athlon 1216
12.7 G4/400MHz 267[/pre]
I'd be wrong to conclude my Mac is doing any better, relative to the others, than its speed rating or benchmark would suggest?
Ahem, Odysseus, you might
)
Ahem,
Odysseus, you might want to edit that link, while there's still time.
I stand corrected on G5-only AltiVec.
Michael R.
microcraft
"The arc of history is long, but it bends toward justice" - MLK
RE: Sorry, I don't
)
Sorry to have you confused.... I was merely implying that G4 Altivec stands a better instruction than a G5 if not equal, because of different processor manufacturers.... but then again this will no longer be the case as for the past weeks x86 optimisations have been rampant.... with those excessive optimisation and benchmarking, even a Quad G5 got "smoked" by an AMD X2.... Imagine if such optimisation goes public, plus with participants taking advantages of the optimised BOINC client superbenches. And Bruce (Einstein founder, correct me if I'm wrong) had announced that he has already got those optimisations implemented into new science apps pending official release.
For the past months, Altivec-optimised hosts have had an edge over other host systems minus excessive benchmarking, but from what I've seen on the x86 optimisation, excessive benchmarking alongside the considerately-improved processing time is why this thread is taking shape.
I'm saving my credits for a
)
I'm saving my credits for a new car!
lolx....
)
lolx....
RE: Odysseus, you might
)
Whoops ... too late to edit. It was supposed to go to WU No.5347493. I don't know why I put the URL in quotes, unless it was mental leakage from working on some web pages recently:
[pre][/pre]
RE: I'm saving my credits
)
Ahh.. very ambitious! I'm simply saving for that cute pink flamingo on the top shelf. :)
"Chance is irrelevant. We will succeed."
- Seven of Nine
Give me the blue-skinned
)
Give me the blue-skinned alien beauty.
RE: Give me the
)
Gold plated hedgehog please.
"I am tired of all this sort of thing called science here... We have spent
millions in that sort of thing for the last few years, and it is time it
should be stopped."
-- Simon Cameron, U.S. Senator, on the Smithsonian Institute, 1901.