Apple Dual 64bit 2ghz vs. Single 2.2Ghz Athlon 64

[FrYGuY]

The results may suprise you! Unless, like me, you didn't believe Apple's overinflated numbers...

http://www.pcworld.com/news/article/...49,pg,8,00.asp

[ClawTovski]

sure compare apples to pears...

nice to compare two programs which got nothing to do with eachother

why don't they compile and run a Java program?

[Cirga]

Still rather interesting. Maybe the Apple PC might be a good way to go for high end work?

For gamig.. not sure if an Apple would be a good choice only because of compatibility... and the fact that I can break and fix everything in my PC which I do all the time.

[FrYGuY]

Originally posted by ClawTovski
sure compare apples to pears...

nice to compare two programs which got nothing to do with eachother

why don't they compile and run a Java program?

Considering Graphics editing is Mac's strong point, and it didn't even win the Photoshop benchmark...

[ClawTovski]

First, the Apples are 1.8 and 2.0, there are no 1.8 pc's only 2.0 and 2.2 GHz models.

Second, only the Intel Opteron with 2 GHz and the same amount of memory / HD-setup managed to be faster than the Apple 2 GHz on Photoshop. The rest is faster or has RAID, which gives a huge performance boost when swapping.

Third: "In Adobe Photoshop we timed the operation of ten filters on a 50MB image file and a 150MB file." which filters? The Opteron has good support for rough processing tasks (like encoding etc), but does this say anything about how smooth the application works? Does it say anything about how it 'feels' when you drag the image around your screen?

Fourth: I can make a benchmark in which the Apples have the highest scores, the problem with these benchmarks is that it is not clear how the test was taken. Word for Windows and Word for OSX are two completely different programs, and each comparisation between those two will give a meaningless comparisation.

[FrYGuY]

Originally posted by ClawTovski
First, the Apples are 1.8 and 2.0, there are no 1.8 pc's only 2.0 and 2.2 GHz models.

Second, only the Intel Opteron with 2 GHz and the same amount of memory / HD-setup managed to be faster than the Apple 2 GHz on Photoshop. The rest is faster or has RAID, which gives a huge performance boost when swapping.

So, because the SINGLE 2Ghz CHIP opteron beat the DUAL 2Ghz CHIP Power Mac G5, the test was unfair?

Again, I point out that Graphics editing is Mac's strong point.

And, by the way, just because Java runs on multiple OS's, doesn't mean Java runs equally well on the different OS's, so your claim that that should have been the test is moot. Running Photoshop as one of the tests was actually a bias towards Mac, because like I've said, Mac's strong point is graphics editing.

[ClawTovski]

You got a point there, but all I was trying to say is that measuring how fast a car can drive is not a measure of how 'good' that car is.

I didn't see the dual thing at first. I agree with you that you would expect the dual 1.8 be faster than a single 2.0.
Unfortunately this is untrue, because probably only one processor will be used when photoshop is doing it's processing. This is not due to OSX neither the hardware, but because software has to be written with support for multiple processors (nodes). Java for instance is written to support multiple nodes. It still is unknown if the two nodes are used in such case, because using two or more nodes is a complex task and not always possible too.
A car with two engines doesn't have to run faster too.

Yes I think it ain't a bad idea to have that the Apples perform very bad, in the end that's what the article was meant to say.

[FrYGuY]

Originally posted by ClawTovski
You got a point there, but all I was trying to say is that measuring how fast a car can drive is not a measure of how 'good' that car is.

Oh not at all. If you like Mac's interface and think that's the most important thing, go ahead and buy it. There are plenty of people who feel that way. I, frankly, hate Mac's "Somethings wrong, but you don't need to know what" and "You don't need to be able to upgrade at all, just buy a new one!" philosophies. And with Windows XP being roughly comparable to OSX as far as interface, stability, and speed, and toss the extra availability of software for Windows (I exclude Linux from all of this, as Linux is available for both) and I'll choose Windows every time. At least that's me.

I didn't see the dual thing at first. I agree with you that you would expect the dual 1.8 be faster than a single 2.0.

Actually, the 1.8Ghz was single. The 2.0 Ghz Mac was dual. The Single 2.0 PC was still faster.

Unfortunately this is untrue, because probably only one processor will be used when photoshop is doing it's processing. This is not due to OSX neither the hardware, but because software has to be written with support for multiple processors (nodes). Java for instance is written to support multiple nodes. It still is unknown if the two nodes are used in such case, because using two or more nodes is a complex task and not always possible too.

Very true, but the single 2. Ghz Opteron still won.

A car with two engines doesn't have to run faster too.

Um. Yes it does, if the two engines are the same horsepower as the single. Have you ever seen unlimited class tractor pulls? Some of them have 4 or 5 engines, and for a reason. More power is created than a single engine alone would. More power equals more acceleration, and with a proper transmission it also means more speed. But that's physics, and not computers. Your analogy just doesn't extend from computers to cars in this area.

Yes I think it ain't a bad idea to have that the Apples perform very bad, in the end that's what the article was meant to say.

That was kinda the point of my posting the review. And I wouldn't even say "very bad", more accurately would be "Nowhere near the performance levels their marketing department attributed to them". It's faster than MY box, that's for sure.

[ClawTovski]

Originally posted by FrYGuY
Oh not at all. If you like Mac's interface and think that's the most important thing, go ahead and buy it. There are plenty of people who feel that way. I, frankly, hate Mac's "Somethings wrong, but you don't need to know what" and "You don't need to be able to upgrade at all, just buy a new one!" philosophies.

This is a myth. Apple has like Microsoft and RedHat also an up-to-date feature, which has every few months an update or two.
OSX tells you in the same way like UNIX what's wrong this time. (If there is something wrong...)

Um. Yes it does, if the two engines are the same horsepower as the single. Have you ever seen unlimited class tractor pulls? Some of them have 4 or 5 engines, and for a reason. More power is created than a single engine alone would. More power equals more acceleration, and with a proper transmission it also means more speed.

I don't agree with you here. I said "doesn't have to". I didn't claim a car with two engines doesn't run faster.
Tractors are typical vehicles which need power, not speed. Two engines each connected to one wheel don't have to go faster, and surely won't go 2x faster than one engine connected to both wheels.

Main point: not vehicles, but adding a few processors doesn't mean the speed increases with the same multiplier. It can even decrease due to lock/wait and/or communication.

That was kinda the point of my posting the review. And I wouldn't even say "very bad", more accurately would be "Nowhere near the performance levels their marketing department attributed to them". It's faster than MY box, that's for sure.

Did you notice the Intel outperforms the AMD's here? I guess you have read all those roaring stories about AMD's being faster and cheaper as well too?
My main point: don't trust/rely on benchmarks. You can create a benchmark so it concludes what you like to see as a result

Microsoft managed to have a benchmark result of Windows being cheaper than Linux on the long term. They simply let all problems Linux can have occur at once and compared that to
Windows systems having no tricky situations at all...
RedHat can do (and probably has done already) exactly the opposite by choosing and applying all the weak spots of Windows... What is the final value of such benchmarks?

[FrYGuY]

Originally posted by ClawTovski
This is a myth. Apple has like Microsoft and RedHat also an up-to-date feature, which has every few months an update or two.
OSX tells you in the same way like UNIX what's wrong this time. (If there is something wrong...)

That's great and all, but my point was Hardware. My favorite story about Mac comes from a friend of mine. He had a Mac, and he bought a game for it. He puts it in his computer, and discovers that the OS is too old. So he upgrades. The new OS, however, discovers a slight glitch in the connection to the CD-ROM. Rather than being a $10 problem (IDE cable) like it would be on a PC, Mac in its infinite wisdom decided to hard-wire it. So now he has a computer without a CD-ROM (Which, ironically, he used to install the OS). So, he takes it to the local Mac repair shop, and they replace it for $400 or so.

The kicker? The processor, which Mac claimed would be fast enough to run the game, wasn't actually fast enough. He then swore off Mac for life.

My point? Mac's main advantage, that it has complete control over it's hardware, is my main pet peave. That is why I will *NEVER* own a Mac. But that's besides the point. Apple still lost a head to head benchmark with a comparable system, when Apple claimed it was 'miles ahead'.

I don't agree with you here. I said "doesn't have to". I didn't claim a car with two engines doesn't run faster.
Tractors are typical vehicles which need power, not speed. Two engines each connected to one wheel don't have to go faster, and surely won't go 2x faster than one engine connected to both wheels.

Um. Yes, it does. Again, physics. More power == More speed. Cause, effect. The only way it wouldn't is in the unlikely event that the Transmission wasn't geared for it, and that's only top speed. Acceleration would flat out be faster.

Main point: not vehicles, but adding a few processors doesn't mean the speed increases with the same multiplier. It can even decrease due to lock/wait and/or communication.

That's great and all, but I already said that. I just said that your analogy didn't extend that well. Cars are a different beast than computers.

Did you notice the Intel outperforms the AMD's here? I guess you have read all those roaring stories about AMD's being faster and cheaper as well too?
My main point: don't trust/rely on benchmarks. You can create a benchmark so it concludes what you like to see as a result

Um, except that AMD's are cheaper. That's been AMD's main selling point since day one. They are, however, currently slower than the relative Intel chips, something which AMD has ceded to Intel in the last few months. AMD's big selling point on the Athlon 64 is that it's Cheap 64 bit computing which is backwards compatible with 32 bit computing, Neither of which can be said about the Opteron (And even less so on the point of price about the G5, though it too is backwards compatible).

Microsoft managed to have a benchmark result of Windows being cheaper than Linux on the long term. They simply let all problems Linux can have occur at once and compared that to
Windows systems having no tricky situations at all...
RedHat can do (and probably has done already) exactly the opposite by choosing and applying all the weak spots of Windows... What is the final value of such benchmarks?

They managed to do it by factoring long term savings from the glut of MCSE compared to qualified Linux System Admins. The 'tricky situations' they applied to both. The value of such a benchmark is valuable to a company looking at the overall cost of running said system. That same test, however, (Which was Microsoft SPONSORED, not Microsoft run) concluded that there were applications which would be cheaper for Linux, such as running an eMail server. But that's neither here nor there, and certainly isn't a performance benchmark, but a theoretical expense report.

[ClawTovski]

- end of discussion, would end up in repeating the same over and over again.

[mofen375]

The expecting Warhammer online gongce is coming. Tens of thousands of MMORPG players wanna a try. At the same time, each game studio and website wants to make a big profit. Warhammer online cd key is on hot sell. Warhammer online gold is even preselling weeks before September 18, 2008. And war power leveling? Not one less. Although warhammer power leveling serves the players and make money for the game product company, game studios and websites, which is the same as war cd key and gold, it is the headache of game product company, game studios and websites. Why and why not? All of them need, yet they are not standing on the same line. Game studios and websites wanna the profit. For them, the number of power leveling users, the more, the better. While for the game product company, they wanna war leveling to flourish their game but do not wanna it to disturb even destroy the balance of the game itself. Game studios are suffering force-out. And some websites are suffering force-out too. Safe war leveling websites are in need now. There are still many websites which provide safe Warhammer leveling. Friends, do not miss WoW Power Leveling | WOW Powerleveling | Warhammer Online Power Leveling | World Of Warcraft Power Leveling | Honor Power Leveling --YOYO GameStore. It provides the best service.---------------------------------------------------------------------------------------------------------------------------------Please keep an eye on Warhammer leveling!!!We have free Warhammer Gold and cheap RS Gold.LWB