@WolfpackN64

Quote:

WolfpackN64 wrote: @cdimauro Just to clarify, I didnt write that article.

I've personal feelings that you did it (and you're Jon L. Aasenden, of course), based on what you've written on the article and the contents of your comments here.

On FB, Jon Aasenden never denied it, even when I explicitly stated the correlation.

And here you didn't denied when I've first stated it. Only now.

So, I keep my opinion.

@pavlor

Quote:

pavlor wrote: I have good source for X5000 and 970MP benchmark comparison. Some benchmarks simply don´t show well average performance of the machine.

Can you show them? Because I remember only the onces which were shown here some weeks ago, reporting the situation which I cited.
Quote:

Tabor...

I know, but this forum has serious issues. I tried to edit to correct this typo, but it's not possible.

It has also problems with quoting (sometimes the quoted text is reported, other times no), and once you preview a message you loose it.

It's really difficult to write something: it's should be perfect at the first writing! :(
Quote:

Comparing this one to G5 is ridiculous.

I never did it. Which should be obvious to you, if you know me.

Anyway, it ought to be enough reading my comments, which report my exact words, and verify yourself.

@cdimauro

Quote:

if you know me.

That´s the reason.

@pavlor: and unfortunately I know you.

@cdimauro

Because I thought you made a mistake first. Is Jon L. Aasenden 21, a ginger with a beard? Otherwise I think this is a case of mistaken identity.

@WolfpackN64

Quote:

WolfpackN64 wrote: @cdimauro Because I thought you made a mistake first. Is Jon L. Aasenden 21, a ginger with a beard? Otherwise I think this is a case of mistaken identity.

It can be, but as I said before, I saw that you have said several things in this thread which match with the Aasenden's article, which just by case was written just yesterday.

I might be wrong, but at least I've some reasons.

@pavlor: haven't you found something interesting in my FB comments? There's something which Jon Aasenden said that, I think, deserves.

Last edited by cdimauro on 06-Aug-2017 at 12:52 PM.

@cdimauro

I can imagine people saying similar things, but trust me, we're not the same person. There would be no reason for me to be lying about it.
I have other reasons to be invested into PowerPC and I know my stuff well enough.

@WolfpackN64: fine for me. We should be adult enough to avoid playing hide & seek.

@cdimauro

There we go.
I'm involved with the PowerPC laptop project, just to play open cards.

@pavlor

Quote:

pavlor wrote: @WolfpackN64 G5 has faster FPU (and of course AltiVec), e5500 is faster with integer code.

Even the G4 is faster (or about the same) than G5 in integer, clock by clock. But the G5 wins by having a much higher clock frequency (2.7GHz).

So far there has been no competitor managing to knock the G5 off of the PPC throne in performance. And given the EOL state of the PPC platform, and the shift towards embedded stuff long before that, it seems very unlikely that this will change. I think it's safe to proclaim that the G5 was the undisputed king of the PowerPC era. While running hot and noisy it was as far as the PowerPC platform ever went regarding sheer CPU performance.

@TRIPOS

EOL is still disputable. If NXP would ever cough up the T5xxx platform, it would finally knock the socks off the G5.

But no-one is served with speculation.

@TRIPOS

Quote:

I think it's safe to proclaim that the G5 was the undisputed king of the PowerPC era. While running hot and noisy it was as far as the PowerPC platform ever went regarding sheer CPU performance.

Exactly. Even PC magazines admited it was - after long time - the first CPU in Apple computers competitive to x86.

@WolfpackN64

Quote:

WolfpackN64 wrote: @cdimauro There we go. I'm involved with the PowerPC laptop project, just to play open cards.

Nice to know. I appreciate a lot, thanks!

@TRIPOS

Quote:

TRIPOS wrote: @pavlor Quote: pavlor wrote: @WolfpackN64 G5 has faster FPU (and of course AltiVec), e5500 is faster with integer code. Even the G4 is faster (or about the same) than G5 in integer, clock by clock. But the G5 wins by having a much higher clock frequency (2.7GHz).

Processors performances shouldn't be compared only clock-for-clock, because one can be designed for weak IPC but higher frequencies, and viceversa.

What really counts is the performance at the maximum frequency reachable, at a certain productive process used.

@pavlor

Quote:

pavlor wrote: @TRIPOS Quote: I think it's safe to proclaim that the G5 was the undisputed king of the PowerPC era. While running hot and noisy it was as far as the PowerPC platform ever went regarding sheer CPU performance. Exactly. Even PC magazines admited it was - after long time - the first CPU in Apple computers competitive to x86.

Absolutely not. It was exactly the opposite: Athlon 64 vs. Apple G5 Systems: Not Even Close

The G5 was a really weak processor, compared to the x86 counterparts of the time. Even a P4 was doing much better.

And if you take a look at AnandTech benchmarks, especially on the server market, it was a complete disaster.

@cdimauro

"Regardless, we can finally say that Apple has introduced a worthy competitor to the PC platform."

Maximum PC, November 2003, p. 55.

@cdimauro

Steve Jobs said "Motorola chips suck" before he went with the IBM chip for the G5. Guess he regretted that afterwards. Though, since he always kept an x86 built of Mac OS X behind his back, that might have been his plan all along.

@pavlor

Quote:

pavlor wrote: @cdimauro "Regardless, we can finally say that Apple has introduced a worthy competitor to the PC platform." Maximum PC, November 2003, p. 55.

The numbers speak differently.

Of course, if you want to read them.

@WolfpackN64

Quote:

WolfpackN64 wrote: @cdimauro Steve Jobs said "Motorola chips suck" before he went with the IBM chip for the G5. Guess he regretted that afterwards. Though, since he always kept an x86 built of Mac OS X behind his back, that might have been his plan all along.

Jobs plan was to release MacOS X 1.0 for x86, because of what you reported (Motorola wasn't able to scale the processors as he was expecting, and at the end of the 90s Intel and AMD were able to catch and became much more convenient).

He regretted only because an IBM rampant manager (which lately became the Freescale CEO) promised him the above G5, with much better performances.

We know that it didn't happened, so the move was just delayed.

@cdimauro

Motorola had some problems, but their designs were still competitive. Some dual core 2Ghz G4 accelerator cards by Sonnet even beat out some G5 systems. Ultimatly, I think Apple wanted to focus picking and designing chips on the ARM architecture for their iPhones and went x86 as the easy way out (because the argument that Intel's chips were faster as a then 3 year olf PowerPC chip were just an excuse). Coincidentaly, their exit of PowerPC, buyout of PA Semi and the iPhone all came in a one year timeframe.

@cdimauro

Quote:

The numbers speak differently.

You disputed my claim, I provided source (die-hard PC magazine).