[Silky Voice of The Gorn]

This article details some inconsistencies with Apple's rating the G5 as the "World's Fastest Personal Computer"

http://www.haxial.com/spls-soapbox/apple-powermac-G5/

Thoughts?

[Pierre B.]

Originally posted by Silky Voice of The Gorn:
This article details some inconsistencies with Apple's rating the G5 as the "World's Fastest Personal Computer"

http://www.haxial.com/spls-soapbox/apple-powermac-G5/

Thoughts?

This subject is being discussed in AI. Look here, there are some interesting comments.

[neilxu]

I only care the real world apps.

[Silky Voice of The Gorn]

Originally posted by Pierre B.:
This subject is being discussed in AI. Look here, there are some interesting comments.

So becauses a subject is being discussed at another forum it precludes discussion here?

[InterfaceGuy]

AMDZONE.COM
I listened in to Apple's presentation from Steve Jobs on the new PowerPC 970 powered G5 systems. I found some claims downright false, and benchmark numbers quite suspicious. Not to mention one must wonder why no comparisons to Opteron or Athlon were made, and Apple doesn't even mention AMD despite their usage of HyperTransport and the AMD PCI-X chip borrowed from Opteron. Read my thoughts here, and see a comparison of Apple's SPEC scores to those of Opteron, Athlon, P4, Xeon, Itanium 2, and Power4. The results seem troubling for Apple, and tarnish the PowerPC 970 which is a much better CPU design than Motorola's G4.

http://www.amdzone.com/articleview.cfm?articleid=1296

[jyvin563]

Originally posted by InterfaceGuy:
http://www.amdzone.com/articleview.cfm?articleid=1296

That article is so one-sided it about made me lose my breakfast. Give me a break! Much of that argument is opinion based. Face it PC world, APPLE HAS A VERY COMPETITIVE MACHINE NOW!

[Pierre B.]

Originally posted by Silky Voice of The Gorn:
So becauses a subject is being discussed at another forum it precludes discussion here?

I don't said that, simply we can take some points and elaborate just to start.

[MacOSR]

Any conjecture on the part of the article would be just as bad as the article claims that Apple is.

The article is making way too many assumptions and then making conclusions based on those assumptions.

I can tell you, from someone that has MULTIPLE Xeon machines that Hyperthreading is NOT necessarily faster when running multi-threaded apps. Some are slower!

The only responsible thing to do is to realize the following:
1. The new G5 machine IS VERY fast.
2. The only real test is real-world comparisons running the same apps.
3. Numbers can be made to say ANYTHING!
4. This machine IS in the league of the PIV Xeon DP 3.06 from what we have SEEN.

I am ordering one as soon as they are shipping and will put it up against a Dell 2600 with dual 3.06 Xeon's running RH AS. We will do a real-world mySQL comparison!

[slider]

Yep, read this guys page and he has some major issues beyond what any Mac Zealot has. He seems like a poster that just can't seem to let things go. He needs to be right all the time and goes to the extreme of producing his website were he has complete control. A control freak. Hey maybe I'll make his "hate mail" page.

Anyways the point of the G5's is that the mac is back and is competative again. I am thinking that this bold statement may be a marketing ploy to get people to look at the new mac.

[slider]

Originally posted by jyvin563:
That article is so one-sided it about made me lose my breakfast. Give me a break! Much of that argument is opinion based. Face it PC world, APPLE HAS A VERY COMPETITIVE MACHINE NOW!

True, I love the statement were he says he doubts that the G5 with hit 3Ghz within 12 months. Obviously he would be in the know.

[Mad Hatter]

True, I love the statement were he says he doubts that the G5 with hit 3Ghz within 12 months. Obviously he would be in the know

Well, look how long it took them to get the G5 out. It's been years and years. So the assessment that a G5 will not hit 3Ghz within 12 months is a correct one, IMHO.

Mad Hatter

[[APi]TheMan]

Originally posted by Silky Voice of The Gorn:
This article details some inconsistencies with Apple's rating the G5 as the "World's Fastest Personal Computer"

http://www.haxial.com/spls-soapbox/apple-powermac-G5/

Thoughts?

Oh geez. A co-worker of mine (PC geek) was reading this article this morning and rubbing it in my face. Then I return to MacNN to find the forums back up and I see this thread... and a link to that article, where I see it's that haxial guy again. Man, this guy is relentless, I'm surprised he hasn't written trojans in his MacOS X software to ruin us all.

I don't care what the benchmarks say. I am still impressed with the dual 1.42 Ghz G4 line of PowerMacs. When I saw the specs on the new PowerMac G5 I was overcome with glee. It's a fast machine. Period.

Bah, I just upgraded my Pismo 400 to 900 mhz... G3! wtf, these new computers are hot sh!t.

edit: By the way, the official test results for the benchmarking are available here from Veritest's website: http://www.veritest.com/clients/repo...erformance.pdf (950k PDF file)

[Scotttheking]

Here's the short version.
The official spec scores are submitted using the best compilers for that platform. Apple used gcc 3.3 for both machines. The scores are lower, but that's a more used compiler.
I asked our programmers here, we use MS' compiler, not the intel compiler.
If Apple has used IBM's Visual Age compiler for the mac (if that's possible), the mac would have been much "faster" too.

[superfoo]

The benchmarks for the P4's on the SPEC site were done using ICC... Obviously, Intel's processor is highly optimized for their CPUs.... likewise, it would have been interesting to see a comparison for the 970 using IBM's C compiler as well... But Apple's benchmarks for both the P4 and the 970 were done using GCC on both platforms... that it is at least one less variable to worry about in the comparison... and it is also the C compiler that the majority of the apps I would be running (for MacOS or Linux) are compiled under.

Otherwise, what's he comparing Apple benchmarks to? Dell benchmarks and Intel benchmarks? I'm sure Dell and Intel are going to pull the same BS as Apple does - all vendors do. And, a big part of the 970 performance gain is on SMP systems.... I'm not surprised that Apple fudged around with the PCs, but this guy hardly even touched on the SMP stuff. The dual 1GHz CPU busses are a huge deal, the 8GB of RAM is a huge deal... I seriously doubt IBM would've sunk $3 billion into building a new FAB for these chips if they were hardly going to compete with the current Intel or AMD offerings... and the "LESS than $2000 Dell" that he mentions (Which actually comes out to about $2300 on Dell's site.. funny coming from someone who would complain about a mfg. having prices for their systems $1 off from the next $1000 increment.) does not have the 64-bit 133MHz or 100MHz PCI slots, does not have FireWire 800 or 400, only supports 2GB of RAM, and does not have a serial ATA. And it's still a uni-processor system.

Regardless of what the guy on that site says, there's a notable difference between running on uni-processor and dual processor.... I say that from personal expereience.

I would gladly pay $800 more for a system with that much more superior hardware, OS, build quality, and aesthetics... and dual CPUs... and that whole 35dBA sound rating is pretty damn impressive too...

Real world apps? It's kind of hard to get an idea on something that isn't readily avail. for folks to test, but the demos on the video at http://stream.apple.akadns.net/ ,while surely slanted, are pretty DAMN impressive, even so.

Benchmarks are going to be skewed from any angle... for anything... go look at the disk I/O benchmarks for the Xserve on Apple's site... YEAH, RIGHT... but in real world scenarios, the new PowerMacs are AMAZING... it's not just CPU that matters, it's I/O in general... and these things are just amazing so far as that goes.

If someone could find a [cheaper] PC with specs and an I/O system to rival those of the new PowerMacs, please give me a URL, as I would very much like to see it - I'm thinking of upgrading my [Linux-based] fileserver sometime this year.

[Spliffdaddy]

Well, you hafta figure that Apple used whatever made their hardware look the best. If there was another tweak - they would have jumped on it. Everyone seems to agree that Apple certainly didn't go out of their way to use the best available methods for benchmarking the x86 machines. A score of 11 is hard to explain when you personally know that similar machines score 15+. Heck, my 2.8 P4 does better than the 3.06 Apple used. Why didn't Apple simply re-print the SPEC scores that already existed for the Dells? Because the numbers were too high.

Bottom line: The dual G5 is competitive with high-end peecees.

[slider]

Originally posted by Mad Hatter:
Well, look how long it took them to get the G5 out. It's been years and years. So the assessment that a G5 will not hit 3Ghz within 12 months is a correct one, IMHO.

Mad Hatter

"They" Mad Hatter, who is "they". Apple does not actually make the chip, and Moto was not putting the money into R&D. This is a completely different chip which is being produced by IBM. I have always thought it was funny how so many have blamed Apple for not putting faster chips into their Macs. Apple does not make chips. Your reasoning, 'look how long it took them to get the G5 out', is misleading. If we were talking about the Moto G4 chips and they had just reached 2Ghz, then yes, you would be correct. Why do you think so many of us are so happy Apple is moving away from Moto? It's a completely different ball game, I thought that was obvious.

[idyll]

If that's a fact, I wish Apple would have just admitted it..

It won't stop me from buying one, though. They're great systems, and finally something to brag about.

[nvaughan3]

overclockers chimes in:

http://www.overclockers.com/tips00408/

[365]

These guys can diss the Mac hardware all they like, the thing they have no answer to is OSX and that just got a whole lot better with Panther.

To argue with these people is futile, I prefer to sit here with a smug grin on my face.

[The Ancient One]

Originally posted by Mad Hatter:
Well, look how long it took them to get the G5 out. It's been years and years. So the assessment that a G5 will not hit 3Ghz within 12 months is a correct one, IMHO.

Mad Hatter

Disagree. IBM just built the facility where they make these chips. The current chips are 130nm and the 3GHz will be 90nm. Jobs mentioned something about next generation samples existing now, so I don't expect them to have any problems at all. The G5 is only the catchup chip, after all. Remember, IBM is moving to the Power5 and the corresponding desktop chip will probably be called the 980. I'm sure Apple will rename it the G6. Look for it in '05 starting at 4GHz minimum. That should put Apple decisively ahead, at least for a while.

[Eriamjh]

I don't see the point in discussing benchmarks until real world, side-by-side, same software to same software comparisons are made with real shipping hardware.

I'm MORE interested in how the new 1.6 G5 compares to the dual 1.4 G4 and the Dual 2.0 G5.

I will never own another PC for everyday use, but I'd like to see how much of a leap Apple has made in total speed comapred to the last revision. Don't you all agree?

[itai195]

Computer companies manipulating benchmarks? Wow, I'm shocked! The funny thing is that people for some reason expected better. Apple has never been a great source for benchmarks, this isn't really news. The Intel compiler everyone's wringing their hands about is optimized specifically for high SPEC benchmarks, so it doesn't offer a "real world" environment either.

[[APi]TheMan]

Originally posted by nvaughan3:
overclockers chimes in:

http://www.overclockers.com/tips00408/

From the overclockers article:

No Credibility

If Apple were part of the x86 market, they'd be ripped to shreds in a second presenting this kind of data. It would be a huge scandal. PC companies get torn a new one for attempting even a small fraction of what we have here.

Heh. Weren't nVidia and ATI both just busted for majorly fibbing on benchmarks for their latest cards? Gah. PC zealots/Mac zealots... we're all the same.

We make a point, they make a point, they call us cheaters, then they cheat. That's how the market works, you have to cheat to get ahead. It makes me sick.

then:

But it's a different world in Macdom. The average Macster is woefully ignorant of hardware, and tends to believe anything and everything Apple tells him. For many, it is guru and flock.

Man, this guy should wander around these forums for an hour or two... Yikes, it must suck to be as uninformed as this gentleman. I guess we're not "average macsters"... but then "average macsters" wouldn't really care about all this "dual 64bit 2 gigahertz G5s with 1 gigahertz frontside bus and 8 gigs of RAM" stuff... they're just computers to those average fellows... and the newest PCs are still... just computers to those average folks too.

[nforcer]

Hmm, it is becoming increasing obvious that posts need to have multiple topics to save the mods time from closing/merging/redirecting posts to others already listed.

But anyway, my reply to the subject is here

[nvaughan3]

Heh. Weren't nVidia and ATI both just busted for majorly fibbing on benchmarks for their latest cards? Gah.

i think that's his point.

[[APi]TheMan]

Originally posted by nvaughan3:
i think that's his point.

"PC companies get ripped a new one" for cheating on benchmarks? Yet... ATI and nVidia and AMD and Intel are all still... big names... doing big business... Look at Microsoft... biggest cheaters of them all, yet Bill Gates makes more money than I'll make in a lifetime in one day.

[Graymalkin]

This "Apple lying" crap is getting old and is only a few hours along. Apple used a specific testing pattern to show its computers are faster than a competitor's. If you followed their testing paradigm to the letter you would find the same results. Other tests might show the G5 falling behind a Dell, these are other tests. As far as Apple's tests are concerned, the G5 is the fastest desktop in the solar system. In a more objective test or another subjective test this might not be so. It is called marketing, everyone does it, that is how people get exposed to products.

Regardless of objective tests the G5 is an extremely fast system. The POWER4 it is derived from is extremely fast. It is highly likely a G5 will get done what you need to do in very short order. It will do so using OSX and while looking really sweet. If you don't want to use OSX or look sweet doing whatever it is you need to do, buy a Dell or some other computer. I am not going to care, it is likely no one else besides yourself will care.

There is no reason to get offended by marketing, it is all shullbit from start to finish. A good marketeer can find a way to make my 66 Mustang beat a turbocharged Supra in a quarter mile. Would it happen all the time? Hell no. Might it happen in some particular case? Maybe. Will the marketeer ever mention the cases where my Mustang doesn't beat the Supra? Hell no.

[snerdini]

Originally posted by Graymalkin:
A good marketeer can find a way to make my 66 Mustang beat a turbocharged Supra in a quarter mile. Would it happen all the time? Hell no. Might it happen in some particular case? Maybe. Will the marketeer ever mention the cases where my Mustang doesn't beat the Supra? Hell no.

And it sure doesn't mean your '66 Mustang isn't cool as hell either.

[KidRed]

Originally posted by [APi]TheMan:
From the overclockers article:

Heh. Weren't nVidia and ATI both just busted for majorly fibbing on benchmarks for their latest cards? Gah. PC zealots/Mac zealots... we're all the same.

We make a point, they make a point, they call us cheaters, then they cheat. That's how the market works, you have to cheat to get ahead. It makes me sick.

then:

Man, this guy should wander around these forums for an hour or two... Yikes, it must suck to be as uninformed as this gentleman. I guess we're not "average macsters"... but then "average macsters" wouldn't really care about all this "dual 64bit 2 gigahertz G5s with 1 gigahertz frontside bus and 8 gigs of RAM" stuff... they're just computers to those average fellows... and the newest PCs are still... just computers to those average folks too.

I find this curious too-

Finally, in the Apple testing, the code for the Apple machines were tweaked to the max with all kinds of libraries installed..

What libraries? X needs additional libraries to operate an app? Adding libraries can make photoshop run faster? Or adding libraries make the spec benchmarks better?

I have long heard the spec #s are unfair because it's an x86 type test and isn't optimized for the mac.

Also, where are all these PC whiner's remarks concerning Logic, Photoshop and mathmatica?

[KidRed]

Someone at macrumors pointed out this guy's claims to be a mac lover are BS. His screenshots of Apple's benchmark graphics are taken from a PC because of the crappy windows non anti-aliased text. So he's full of **** before he even begins typing his full of **** opinions.

[KidRed]

I am a Mac user. I have been using Macs for years. I am writing this article on a PowerMac G4. I enjoy using Macs.

Yet, this is a screenshot from a PC?

[Nebrie]

Originally posted by Mad Hatter:
Well, look how long it took them to get the G5 out. It's been years and years. So the assessment that a G5 will not hit 3Ghz within 12 months is a correct one, IMHO.

Mad Hatter

You're confusing building an entire processor from scratch with simply tweaking an existing processor. Totally different things.

[[APi]TheMan]

Originally posted by KidRed:
I find this curious too-

What libraries? X needs additional libraries to operate an app? Adding libraries can make photoshop run faster? Or adding libraries make the spec benchmarks better?

Veritest installed said libraries :

"Installed a high performance, single threaded malloc library. This library implementation is geared for speed rather than memory efficiency and is single-threaded which makes it unsuitable for many uses. Special provisions are made for very small allocations (less than 4 bytes)." From the Veritest results PDF (page 5)

When I saw that all this ruckus was being stirred up by the same old haxial.com/soapbox URL as the "Mac bigotry" article a few weeks back, I couldn't help but laugh.

[graphics84]

So when can we get some real world speed comparisons?

photoshop
quake
etc...

what's that web site that does all that?

seems like they would have a rep at the show...

[Nebrie]

Originally posted by graphics84:
So when can we get some real world speed comparisons?

photoshop
quake
etc...

what's that web site that does all that?

seems like they would have a rep at the show...

They did the Photoshop, Mathematica, etc. ones at the keynote. Watch it at http://www.apple.com/quicktime

You can get the Quake ones here.

http://www.apple.com/powermac/graphics.html

[klinux]

Originally posted by The Ancient One:
The current chips are 130nm and the 3GHz will be 90nm.

I like the G5. In fact, I love the G5! However, by the time it ships, Intel will be ready with its Prescott chips (90nm). Not that it matters but I guess that's what Steve gets for comparing a system (G5) to be shipped 3 months later to a system (Xeon) that is available 3 months ago!

[power142]

When push comes to shove, the new G5 machines will be fast, but who cares if it takes an extra 5 milliseconds to open a Word document?

Now if you're talking about *real* apps that's a different story. I've had the 'delightful' experience of tweaking scientific software to run on a variety of machines. For what it's worth, Hyperthreading don't make a lot of difference either way once you start giving the processor a lot of work to do, but you sure as hell can't run 4 threads and expect them to run as fast as 2 will (far from it in fact). Intel's compiler is another issue. A number of apps I've had the pleasure of compiling end up running at best 5% faster, and ironically run better on an Athlon. In my opinion, if I have to spend weeks with a million different compiler options to get an app to run any faster, it's a waste of my time!

[aloner]

I trust Apple more than I trust the author of that article.

His opinion on the new PowerMac is as relevant to me as Homer Simpson´s

[MindFad]

I was more excited to see the G5 kickin' ass in the demos against the dual Xeon in Photoshop, Luxology, etc., not the SPEC numbers. It's very fuggen fast, and nicely priced. Untangle your panties, people. My SPECs are bigger than your SPECs, nyaaaaaah. "Dude, I just got the dual 2GHz G5 -- BBEdit *flies* on this puppy!"

Me, hopefully I'll have saved up by the time the dual 3GHz ones come out, or 980s.

[rm199]

I want to see some bmarks of VPC running on the dualie 2Ghz. If there is one application that will make a step forward it is likely to be that...

RM

[bigmat]

Apple respond to the claims

http://apple.slashdot.org/article.pl...26&tid=181

[Eug Wanker]

Originally posted by bigmat:
Apple respond to the claims

http://apple.slashdot.org/article.pl...26&tid=181

"Apple Hardware VP Defends Benchmarks

Posted by pudge on Tuesday June 24, @07:11PM
from the i-blame-florida dept.

Greg Joswiak, vice president of hardware product marketing at Apple, in a phone interview today, defended Apple's performance claims for its upcoming Power Mac G5, after they came under fire in the wake of yesterday's announcement. Read on for the details.

Joswiak went over the points in turn, but first said that they set out from the beginning to do a fair and even comparison, which is why they used an independent lab and provided full disclosure of the methods used in the tests, which would be "a silly way to do things" if Apple were intending to be deceptive.

He said Veritest used gcc for both platforms, instead of Intel's compiler, simply because the benchmarks measure two things at the same time: compiler, and hardware. To test the hardware alone, you must normalize the compiler out of the equation -- using the same version and similar settings -- and, if anything, Joswiak said, gcc has been available on the Intel platform for a lot longer and is more optimized for Intel than for PowerPC.

He conceded readily that the Dell numbers would be higher with the Intel compiler, but that the Apple numbers could be higher with a different compiler too.

Joswiak added that in the Intel modifications for the tests, they chose the option that provided higher scores for the Intel machine, not lower. The scores were higher under Linux than under Windows, and in the rate test, the scores were higher with hyperthreading disabled than enabled. He also said they would be happy to do the tests on Windows and with hyperthreading enabled, if people wanted it, as it would only make the G5 look better.

In the G5 modifications, they were made because shipping systems will have those options available. For example, memory read bypass was turned on, for even though it is not on by default in the tested prototypes, it will be on by default for the shipping systems. Software-based prefetching was turned off and a high-performance malloc was used because those options will be available on the shipping systems (Joswiak did not know whether this malloc, which is faster but less memory efficient, will be the default in the shipping systems).

As to not using SSE2, Joswiak said they enabled the correct flags for it, as documented on the gcc web site, so that SSE2 was enabled (the Veritest report lists the options used for each test, which appears to include the appropriate flags)."

BTW, I asked a some computer geek friends which compiler they use for Linux on x86. Very few actually use the Intel compiler. Why? Because each licence costs several thousand dollars. So they use GCC, which is free. Ironically, one IBM friend is switching over to the Intel compiler on his Linux boxes because IBM is paying for it.

[Coroner]

Murder per 100,000 residents in counties won by Gore 13.2 by Bush 2.1

Is this supposed to imply that people who live in these counties don't deserve a vote? Or is it supposed to imply that all the 'good' people voted for Bush? Or maybe you are implying that blacks, who largely vote Democrat, are all murderers, is that it?

Why is this racist allowed to have this sig?

[allformac]

Here's where I think the new Power Mac G5 will crush what Intel offers...

I don't know if anyone else has mentioned this yet, but what about how powerful Mac OS X is at multithreading.

Imagine having tons of applications all running at the same time with no affect on performance at all.

I mean, not only will we have a much more powerful processor (with a HUGE FSB), but we will be able to add a lot more RAM. And we all know how much Mac OS X loves RAM.

This is what I feel the new G5 will bring to us.

And let's face it. I don't care how fast Intel's offerings are. When running Windows, you will always take a performance hit -- especially when you start wanting to have more than one application open at the same time.

[Stratus Fear]

Originally posted by allformac:
When running Windows, you will always take a performance hit -- especially when you start wanting to have more than one application open at the same time.

Ugh, myth and zealot misconception. I use Windows. I use many different programs at once. I also run multiple instances of some of those programs. I don't experience any slowdowns or performance hits until I use all available physical memory. This is how it works with any system.

[Eug Wanker]

Originally posted by allformac:
Here's where I think the new Power Mac G5 will crush what Intel offers...

I don't know if anyone else has mentioned this yet, but what about how powerful Mac OS X is at multithreading.

Imagine having tons of applications all running at the same time with no affect on performance at all.

I mean, not only will we have a much more powerful processor (with a HUGE FSB), but we will be able to add a lot more RAM. And we all know how much Mac OS X loves RAM.

This is what I feel the new G5 will bring to us.

And let's face it. I don't care how fast Intel's offerings are. When running Windows, you will always take a performance hit -- especially when you start wanting to have more than one application open at the same time.

OS X will slow down the OS compared to Linux. Windows will slow down the OS compared to Linux.

G5 Power Macs have an 800 MHz memory bus. Intel machines have an 800 MHz memory bus.

G5 Power Macs have up to a 1 GHz FSB. Intel machines have up to an 800 MHz FSB.

OS X is multithreaded. x86 Linux and Windows are multithreaded (with hyperthreading available on the Intel side).

Power Mac G5s will support up to 8 GB RAM. Wintels will support up to 4 GB RAM (or possibly more with certain setups).

Thus, as far as I can tell, they're pretty closely matched. Both are screamingly fast for most consumer usage.

[Stratus Fear]

Originally posted by Eug Wanker:


OS X will slow down the OS compared to Linux. Windows will slow down the OS compared to Linux.

G5 Power Macs have an 800 MHz memory bus. Intel machines have an 800 MHz memory bus.

G5 Power Macs have up to a 1 GHz FSB. Intel machines have up to an 800 MHz FSB.

OS X is multithreaded. x86 Linux and Windows are multithreaded (with hyperthreading available on the Intel side).

Power Mac G5s will support up to 8 GB RAM. Wintels will support up to 4 GB RAM (or possibly more with certain setups).

Thus, as far as I can tell, they're pretty closely matched. Both are screamingly fast for most consumer usage.

Exactly, damnit I think the G5s might be eeking past the current P4s slightly, but the P4 3.2 was announced the other day, and by the time the G5s are available, the Athlon 64s will be out. It's a neverending cycle, we shouldn't get all bent out of joint about it, right?

[allformac]

You have the wrong idea with my post.

I'm not some Mac zealot. I have used Windows all my life. From 3.1 all the way up to XP. And I really don't know what you're talking about.

Windows being multithreaded and able to handle multitasking with ease?

My experience with Windows has always been sort of bad. Programs crashing all the time. System slowdown after running heavy duty applications.

I have just recently switched to the Mac platform and couldn't be happier. I am impressed at the way Mac OS X handles multitasking.

I can listen to music, burn cds, surf the Internet, and do graphics work with Photoshop all at the same time.
---------------------

And I feel that the new G5 along with the ability to add up to 8GB of RAM with greatly improve all of this.

Will this be true?

Maybe. Maybe not. I don't know. I don't own one yet. These are just some of my opinions. I could be wrong.

But hey, that's what these places are for.

[Stratus Fear]

That argument I usually hear from Mac people who've barely used Windows. My assumption wasn't a good one, my bad. Regardless, some people do have bad experiences with either OS. That doesn't mean everyone has the same one as you though -- most people don't have problems with XP. The reason it seems like they do, though, is because only the people with problems really talk about it.

[clarkgoble]

Where the benchmarks are helpful is in regards to compilers, video drivers and so forth. Further while I think the G5 is more than fast enough, one must admit that the Opteron is almost certainly faster. Likewise while I think the dual G5 is fairly priced, I tend to think the single processor 1.6 and 1.8 GHz boxes are about $300 overpriced as compared with PCs.

There are good cases to be made about gcc on both sides. However realistically most software on Windows uses the MS C compiler or the Intel compiler. Linux users use gcc, but it isn't as fast for most things. For real world apps, however, gcc isn't used. And the newer version of MS C is much improved over other versions and the Intel compiler, while tweaked unfairly for Spec, is also typically faster than the MS C.

Likewise on the Mac side Codewarrior is typically faster than gcc. And indeed gcc hasn't been that great although I hear 3.3. is much improved.

Given that, I really think they should have compared the Mac gcc to the MS C. But that's me.