aeberbach
Headphoneus Supremus
The third party tests need to by somebody who isn't paid to do them, especially not by Apple!
Quote:
BoyElroy this is wrong. You have presented no evidence or sources for this. Why can't you understand that market share does not matter as long as the business you do brings in more money than it costs you to do it?
Nobody here is claiming that Apple's benchmarks are reliable. As I suggested (and MacDEF agreed) there will be no reliable indication of G5 performance until a non-partisan 3rd party like Arstechnica does some real testing and compares it directly in real-world applications.
Joswiak's comments are the kind of rubbish you would expect from a VP of Marketing. Either he knows as much about what he sells as Best Buy staff of he's just a liar.
"gcc has been available on the Intel platform for a lot longer and is more optimized for Intel than for PowerPC"
No. Microsoft and Intel compilers outperform gcc in almost any test. GCC is acknowledged as about the worst compiler available.
"Apple numbers could be higher with a different compiler too"
But they just couldn't be bothered doing that before the launch of this flagship new line of machines - wouldn't want to look TOO good.
Ed Stroligo's article is mostly reasonable but he misses a really interesting test that could have been performed - x86 Darwin on dual 3.06 Xeons. I say mostly reasonable because he misses a few more things too. I expect that would make a G5 look pretty weak. But it's not just about benchmarks. If I knew the G5 would be as snappy as Windows XP on a P4, and had an anti-aliasing technology as good as Cleartype I would probably buy one.
Quote:
Regarding the first item, clearly, Apple Inc. is in financial trouble |
BoyElroy this is wrong. You have presented no evidence or sources for this. Why can't you understand that market share does not matter as long as the business you do brings in more money than it costs you to do it?
Nobody here is claiming that Apple's benchmarks are reliable. As I suggested (and MacDEF agreed) there will be no reliable indication of G5 performance until a non-partisan 3rd party like Arstechnica does some real testing and compares it directly in real-world applications.
Joswiak's comments are the kind of rubbish you would expect from a VP of Marketing. Either he knows as much about what he sells as Best Buy staff of he's just a liar.
"gcc has been available on the Intel platform for a lot longer and is more optimized for Intel than for PowerPC"
No. Microsoft and Intel compilers outperform gcc in almost any test. GCC is acknowledged as about the worst compiler available.
"Apple numbers could be higher with a different compiler too"
But they just couldn't be bothered doing that before the launch of this flagship new line of machines - wouldn't want to look TOO good.
Ed Stroligo's article is mostly reasonable but he misses a really interesting test that could have been performed - x86 Darwin on dual 3.06 Xeons. I say mostly reasonable because he misses a few more things too. I expect that would make a G5 look pretty weak. But it's not just about benchmarks. If I knew the G5 would be as snappy as Windows XP on a P4, and had an anti-aliasing technology as good as Cleartype I would probably buy one.