Sidebar Sidebar
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.

a challenge to all pc users who prowl these boards.

The below link has a pretty nice comparison of both price and performance between Power Mac G5s and high-end x86s. Note that this was prepared by someone who certainly could not be called an Apple enthusiast, but I consider it to be a fair comparison. In terms of performance the 2.7GHz G5 Power Mac came in just above middle of the pack. It posted the second best results in the Photoshop tests (losing to a dual Opteron system that sold for about $1000 more than the Power Mac). In the Cinebench and After Effects tests the Power Macs did not do quite as well. Note, however, that the single-core Penitum 4 system (Velocity Micro 840EE) was overclocked to 4GHz.

However, if you look at the prices you will see that the G5s are actually on the low end of the scale (except in the case of the new dual-core Pentium Extreme, that is where Apple's real problem will be over the next year or so -- the new dual-core Pentiums and Athlons).

http://episteme.arstechnica.com/eve...24/m/867008203731/r/895009013731#895009013731

Finally, as for your tag line about Cray. I believe that the story is the following ( http://en.wikipedia.org/wiki/Seymore_Cray ) :

When in 1986 Apple bought a Cray X-MP and announced that they would use it to design the next Apple Macintosh, Seymour Cray replied, "This very interesting because I am using an Apple Macintosh to design the Cray-2 supercomputer."
Click to expand...
So, it was "design" of the Cray-2, not simulation of the Cray 3.
 
fpnc said:
The below link has a pretty nice comparison of both price and performance between Power Mac G5s and high-end x86s. Note that this was prepared by someone who certainly could not be called a Mac enthusiasts, but I consider it to be a fair comparison. In terms of performance the 2.7GHz G5 Power Mac came in at (or just above) middle of the pack. It posted the second best results in the Photoshop tests (losing to a dual Opteron system that sold for about $1000 more than the Power Mac). In the Cinebench and After Effects tests the Power Macs did not do as well. However, if you look at the prices you will see that the G5s are actually on the low end of the tested systems (except in the case of the new dual-core Pentium Extreme, that is where Apple's real problem will be over the next year or so -- the new dual-core Pentiums and Athlons).

http://episteme.arstechnica.com/eve...24/m/867008203731/r/895009013731#895009013731

Finally, as for your tag line about Cray. I believe that the story is the following ( http://en.wikipedia.org/wiki/Seymore_Cray ) :



So, it was "design" of the Cray-2, not simulation of the Cray 3.
Click to expand...

i got my sig off an exhibition of a cray 1 in the london science museum which btw is a sixth of the speed of my G4 cube :p, meh it needs a change

anyway, those are some pretty interesting numbers, remember that the opteron has four cpu's on two dies and costs the earth and the velocity micro is an overclocked pc and would send your electricity bills through the roof with the cpu probably consuming about 200w and the vapochill keeping it sub zero, and as you can see with some quick calculations the dual 2.7GHz G5 gives pretty good bang per buck (there is a pricing error on the dual dual core opteron) i dont think apple dose have to worry about the extreme edition, for the cost of the dell you can get a dual 2.3 which will probably beat it in the benchmarks, apple has to worry if ibm dose not get the 970MP out by MWP, the real competition is the xeon and that wont go dual core for a while yet.
 
Hector said:
...i dont think apple dose have to worry about the extreme edition, for the cost of the dell you can get a dual 2.3 which will probably beat it in the benchmarks, apple has to worry if ibm dose not get the 970MP out by MWP, the real competition is the xeon and that wont go dual core for a while yet.
Click to expand...
The dual-core 3.2GHz Dell Gen 5s are frequently on sale for just under $2000 (U.S.), so on a price comparison they are often nearer to the 2GHz Power Mac. But I do agree that in some areas a dual 2.3GHz Power Mac would be a near equivalent in performance to the Dell Gen 5.

In any case, with the new dual-core Pentiums and Athlons I think that price is going to become a bigger problem for the Power Macs. The dual-core x86 systems do very well in areas where the dual-processor Power Macs use to have clear advantage on a price/performance basis. Frankly, Apple needs the dual-core 970MP as soon as possible or they need to consider reductions in their pricing (or maybe a combination of both, since even a single-processor, dual-core 970MP system will likely begin at fairly high prices with little performance benefit over the existing dual-processor G5s).

If Apple/IBM can deliver a dual-core 970MP at 2.5GHz before the end of summer then things should be okay for the Power Mac (but they also have to add PCI-Express for video). However, I think there is very little chance that we will see a 3GHz 970MP this year, so Apple is still going to have to watch their prices if they want to remain competitive with the dual-core x86 systems.
 
I use both PCs and Macs daily. I "greatly" prefer Mac's but do not consider anyone who has different taste and prefer's PC to be any less intelligent then I. After all, we have one thing in common...we both love computers. Why would I feel a need to change them? I can only explain my preference for Mac's and listen to their preference's for PC's. I can admit that there are things that PCs do better then Macs (Games) without feeling I've lost anything. With open minds, perhaps we can learn from each other. Both sides are pursueing a common interest, just using different tools.

I'd like to see the day when we get beyond labeling; "PC Users" or "Mac User". I don't feel the need to do the "mind is bigger" scene (leave that to Gates and Jobs)...but then I've been in this business a long time.

OK....I'm off my soap box now! :p
 
thats what i'm saying i just hate it when pc users seem to think that pc's are way way faster than macs.
 
Heh.

Every time I read one of these threads they seem less and less relevant. So much that people use their machines for doesn't require the full power of today's processors. Much of what does, it often ends up back in that <10% difference range, which hardly ever matters to the user anyways. It's like people arguing US vs. Japanese car engines. They get all yakked about horsepower, but hardly ever consider issues like the overall power curve, transmission gearing and tire limitations. Let alone whether they want to drag race or do rough-surface rally racing.
So many factors come into play when looking at how well a computer performs. Sure Doom 3 runs better on Windows. Chances are that Final Cut Studio does faster transitions on high-definition DV streams than the equivalent Winderz software. As has been previously touched on, having a machine that's relatively quiet due to good thermal design may make the price/power ratio worse, but for many people is still a good thing. Oh, and Halo was originally written for the Mac until M$ subverted Bungie. ::shrugs:: As long as your machine still plays it, why care? In the same vein, if gaming is the standard, why don't we start comparing these PCs to consoles? I bet that my PS2 does polygon mapping better for less money than any PC or Mac. That's why I own both. It's just like the cars... what's better often depends on what sets of conditions you're starting with.
Incidentally, that's a good argument for excluding custom-built computers from benchmarking. We're not talking about testing chips here, we're talking about testing computers. If you assemble a computer from parts, it's now a unique creation. Regardless of how superior you can make it, you're not selling them by the hundreds of thousands. So saying your unique modified computer does certain things better than my off-the-shelf unit for the same cost is pretty meaningless. I can put my hands on a Honda Civic that'll beat most any production vehicle in most any test you can set up. That doesn't change the fact that most Civics are econoboxes. Just like most Dells. Well, except they *are* built better...
 
Dreadnought said:
Sounds a lot like Folding@Home, but you can't compair hte result of F@H of one platform with the other because of the optimized codes for windoos, unfortunately for the mac the code isn't completely optimized, but with every revision, they tweak it a bit.
Click to expand...

That's a bad argument. If I buy a computer with a PowerPC, I'll have to use the compilers that are available. Therefore, if the compilers for x86 are better, then this is an advantage vor x86. Arguments like "the PPC would be faster if there were better compilers for it" are nonsense.
 
how well cross platform apps are optimized is an issue as most the apps i use are mac only and one cant to fair benchmarks between different apps, but one can with cross platform apps, if those few cross platform apps favor x86 thats not much of an issue as most apps you will use (safari, mail, iapps, keynote, FCP) are properly optimized.
 
Hector said:
how well cross platform apps are optimized is an issue as most the apps i use are mac only and one cant to fair benchmarks between different apps, but one can with cross platform apps, if those few cross platform apps favor x86 thats not much of an issue as most apps you will use (safari, mail, iapps, keynote, FCP) are properly optimized.
Click to expand...

Even testing cross-platform apps won't give you results that are incredibly accurate. If you want to do a fair and balanced comparison of CPU's, the benchmarks should be run on a linux distro that runs on PPC and x86 like Debian or Gentoo. Otherwise you still have differences due to memory management, process scheduling, driver optimization, etc.

Even if the apps are cross-platform, it might mean that the Windows version is written in C++/MFC, while the Mac version is written in Objective-C/Cocoa. Platform independent optimization is definitely a big issue. If you want to compare the CPUs, the simplest test would be the most relevent in my opinion.
 
you know ppc linux is far far far less optimised than X86 linux.

no this is a test of the mac as a platform if one were to compare cpu grunt you would use the FP performence but in the real world is dose not mean jack all.

if one looks at this site http://www.top500.org/lists/plists.php?Y=2004&M=11

and compare the 2.2GHz G5 cluster to a 2.2GHz opteron cluster and work out the flop/cpu the g5 wins by a fair bit

2.2GHz opteron: 3.149Glop per cpu
2.2GHz 970: 5.76Gflop per cpu (thats with a bigger cluster and the bigger the cluster the slower per cpu added)
3.4GHz xeon: 4.282Gflop per cpu
 
Well, just to offer a benchmark...

This is a little bit old (about a year ago), but it provides a comparison using a cross-platform application in a more-or-less real world computing environment.

VectorWorks benchmark comparison

One thing to note, while this article shows some nice things about the G5, it really shows some of the weakness of the G4. This is one significant reason that I'm awaiting a G5 PowerBook.
 
This looks like an interesting and relevant article on AnandTech that compares the G5/PowerMac to x86 hardware and compares OS X to Linux. I haven't had time to read it yet, but I'm looking forward to reading it later on, once I'm out of work.

Someone else on MR started a new thread about this, but I thought it would be applicable to this discussion as well.
 
You must log in or register to reply here.
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.
Top Bottom
Back