Geek Patrol has pitted an Apple Mac Pro against a Power Mac G5.
The configs of the two test machines:
Mac Pro
• Intel Xeon 5150 @ 2.66GHz (two dual-core processors)
• 1024MB RAM
• Mac OS X 10.4.7 (Build 8K1079)
• Geekbench 2006 (Build 190)
Power Mac G5
• PowerPC G5 @ 2.5GHz (two dual-core processors)
• 1024MB RAM
• Mac OS X 10.4.7 (Build 8J135)
• Geekbench 2006 (Build 180)
Geek Patrol reports, “Overall, the Mac Pro is 7% faster than the Power Mac G5. In individual tests, the Mac Pro is between 34.9% and 280.4% of the performance of the Power Mac G5. The Mac Pro outperforms the Power Mac G5 when it comes to integer calculations and scalar floating point calculations, while the Power Mac G5 outperforms the Mac Pro when it comes to vector floating point calculations and standard library memory performance. Of course, we’ve tested a mid-range Mac Pro ($2500 US) against a top-of-the-line Power Mac G5 ($3000 USD), so even though the Mac Pro is only a little bit faster than the Power Mac G5, it’s significantly cheaper!”
Full article with benchmark details here.
MacDailyNews Take: Obviously, we’d like to see the top-of-the-line Mac Pro (with two 3.0GHz Dual-Core Intel Xeon “Woodcrest” processors, US$3,299 as configured above) vs. the old top-of-the-line Power Mac G5 (Original price: US$3,299) that Geek Patrol has used in their testing.
that when the G5 was the latest and the greatest it truly was an outstanding processor.
What may be more helpful is to time the two machines doing the tasks that customers would use them for…CS2, video work, etc.
What I’m wondering is how the faster processors, etc., would compensate for the Rosetta effect on non universal applications like CS2.
The G5 rocked the HOUSE when it came out. It still kicks some serious butt.
Regardless of what people would like to see, a year old Quad G5 standing up to a brand new Core Duo 2 proves that the PowerPC architecture was solid.
IBM failed PowerPC. When IBM wrote off Apple as not being important enough to keep PowerPC and the G5 a cutting edge processor they seem to have doomed PowerPC. Even the G5 was a lackluster effort on their part.
What is done is done and for reasons outside of performance I like the move to Intel processors. Dual Boot, Virtualized OS Environments and possibly a Wine type implementation in OS X is reason enough to make the switch.
as a video geek – i’m not 100% in awe of the new intel machines – they’ve always put integer perf ahead of fp and vector perf.
of course, in the end, the intels always beat the snot out of the PPCs after a time… so i expect this trend to also continue.
So… one way to summarize is in terms of alternate OS support – the G5 Quad lets you keep your Classic support; Mac Pro gives you boot camp.
The Pentium was not. End of story.
The G5 proves again that Apple did NOT lie in the megahertz myth, but now Intel is again the future.
I’ve nothing to add. Just repeat: waiting for the performance you want will cause it to become available sooner and at a good price.
Resist Apple’s marketing machine by being a smart consumer.
Interesting that the 2.66GHz Dual Duo is 7% overall faster than the 2.5GHz Quad G5, and that 2.66GHz is 6.4% faster than 2.5GHz. Seems clock-for-clock they are the same.
Not bad for a chip that came out a year ago to be on par with one of Intel’s latest & greatest, huh?
Now where’s my 3GHz G5????
Altivec is the difference, otherwise both look pretty equivalent. Which is impressive for the G5 since Woodcrest is supposed to the best ever… so far.
>> Now where’s my 3GHz G5????
In my xbox 360!
I have a 2.5 with 2 gig of ram. I really would have thought the new model it would have been alot faster even as a mid-line model. I wasn’t in the market to upgrade, but now I know even the 3Ghz probably wouldn’t be worth a $3,500(or more) hit to upgrade, for me at least.
So the old G5 is 7% slower than a Woodcrest that is 6% faster. Sounds like a draw to me. And to top it off, you can keep your coffee warm on the G5.
There is no indication that the GeekBench test suite was compiled to take advantage of the “128-bit SSE3 vector engine” that is part of the new chips but un-mentioned with the older chips. Given that the Quad PM has long used G5s with their Velocity Engines, the test compiled for them likely takes advantage of that technology … thus a huge advantage in processing vector floating point calculations. Had they upgraded the memory in the new model they would have seen a jump in memory performance as well.
The message is:
If you have a G5 quad keep and have fun. If you want buy a new one then pay less and have the same amount of fun.
Overall, the Mac Pro is 7% faster than the Power Mac G5…
oooooooh not impressed.
I expect to see that lead get larger after Leopard arrives. OS X is not yet optimized for Core anywhere near as well as it is for PPC. More efficient compilation can make a big difference in a lot of situations. And as DLMeyer said, NOTHING is optimized for the new SSE unit in Core 2 yet.
Nevertheless, it remains true that the G5 is a beautiful and powerful processor.
But it’s also true that IBM is making a whole lot more money selling PPCs to all three new consoles than they did selling PPCs to Apple.
“MacDailyNews Take: Obviously, we’d like to see the top-of-the-line Mac Pro (with two 3.0GHz Dual-Core Intel Xeon “Woodcrest” processors, US$3,299 as configured above) vs. the old top-of-the-line Power Mac G5 (Original price: US$3,299) that Geek Patrol has used in their testing.”
Obviously, I’d love to see a new top-of-the-line Mac Pro with Woodchuck, er Woodcrest, processors go up against a top-of-the-line PPC system running two quad-core Cell processors at 3.0 GHZ. Of course, its about “performance per watt”, so PPC got dumped and we’ll never get to see that in a Mac. And yes – it would be cheaper – a single quad-core cell will be in the PS3 priced out at $500.
Once 33% or so of the Mac installed base moves over to Intel, look to Apple to start weaning you over to Vista, or whatever Microsoft will have by then, since the features will be “seamlessly the same”. The moment Apple went over to Intel, the clock began ticking. Remember the hype about EFI and how a Mac could NEVER boot Windows? Funny how that feature-set got turned off. Keep drinking the Kool-Aid MDN…you’re only slightly better than Rob Enderle – at least he knows the gig is up.
On december or eary 2007, intel will realese the “Quad Core” chip, so there will be 8 cores mac by simply replacing the chips in the actual Mac Pro, can you imagine a quad core iMac or Mac Mini?
I believe taht Steve Jobs Plans never was to have the “Ultimate Proccessor”, G5 with a few more MHz or the Cell proccessor will be far more power that any Intel chip, but Only intel chips will allow Apple to directly take over the PCs market, that is the real reason of Why Apple Move to Intel. Besides that, remember that Apple help Motorola and IBM to desing the PPC chip, so may be can help Intel to make a better, but far more better chip some time.
Once 33% or so of the Mac installed base moves over to Intel, look to Apple to start weaning you over to Vista, or whatever Microsoft will have by then, since the features will be “seamlessly the same”. The moment Apple went over to Intel, the clock began ticking. Remember the hype about EFI and how a Mac could NEVER boot Windows? Funny how that feature-set got turned off. Keep drinking the Kool-Aid MDN…you’re only slightly better than Rob Enderle – at least he knows the gig is up.
You’re delusional. OSX is the Mac. Without it, I would just go buy the cheapest box I could get to run Windows like most other people. I love Apple’s designs, but OSX is the reason why I’m a Mac user. OSX Tiger was Apple’s most profitable software release ever. They’re not abandoning it.
The Power Mac G5 is quite a machine. I don’t have a Quad (it’s a 2.0 GHz dual-core), but performance is not an issue for me. It will serve well for many more years, I’m sure. Perhaps a “third party” can provide a 3.0 GHz Quad PPC upgrade, using future parts that do not require liquid cooling.
TeamZissou sounds like MacDude and I wish people would stop complaining that Apple didn’t go with the Cell. That would have been a horrible decision for several reasons. 1. The cell was designed to be a gaming chip from the ground up. It’s optimized to perform the same functions over and over again which is great for a video game that runs the same program for hours and needs intensive graphic processing (which can be vectorized). That power is wasted in a general purpose computer where you have several applications running and they are constantly requesting different things. 2. Even the things that the cell is good at are hard to access. Taking advantage of the vector power and multiple cores of the Cell is not simple. Coding for it requires specific strategies and from what I’ve heard the developers have found it to be very difficult and time consuming. Apple has always had a problem getting the developer numbers MS has or getting people to port over to the mac. Using the cell would have made that much worse, whereas using Intel makes it much better. 3. The cell is proving to be much more expensive than originally planned. The huge chip with 9 cores has lots of possible points of failure, so that when the chips come off the line at the fab many of the are unusable, meaning the remaining chips have to be more expensive to cover those costs. It also means yields are lower than expected and that’s another chronic Apple problem that they are trying to get away from. Sony can only sell the ps3 for $599 with the cell because they take a huge loss hoping to make the money back in game sales. The actual cost of the ps3 is probably closer to $1000. Add more RAM and RAM slots, harddrive and HD bays, more networking options and the fact that Apple needs to make a profit on the machines and it becomes clear that a cell based PC is not any cheaper than what Apple released on Monday, plus this is available starting two days ago. Cell isn’t available and it isn’t a good choice for Apple.
The new PowerMacs are, on average, a whopping 7% “better” than the G5? This is almost within the range statistical error! I had expected a minimum 30% increase in average performance. I suppose that I’ll have to wait until next year and hope that the next generation of Intel chips will have the power I want.