Apple's Snow Leopard bests Windows 7 in speed tests

[AppleInsider]

When both Mac OS X 10.6 and Windows 7 were tested on a MacBook Pro, Apple's new operating system clearly beat Microsoft in terms of speed, a new test has shown.

Both operating systems were tested on a 2008 MacBook Pro machine by CNet, and each was given its own, separate, clean hard drive. The 64-bit version of each OS was included in the test, which measured a variety of speed and performance related tasks. Snow Leopard was given true, full 64-bit support with most of its native applications taking full advantage of modern processors.

Each OS had the same software installed: iTunes 9, QuickTime, Call of Duty 4: Modern Warfare, and Cinebench R10. In the test, Snow Leopard booted and shut down significantly faster than Windows 7.

"In time-based tests, Snow Leopard consistently outdid Windows 7," the study found. "It took only 36.4 seconds to boot up, while Windows took 42.7 seconds. In a shutdown test, Snow Leopard took only 6.6 seconds, while Windows needed twice the amount of time: 12.6 seconds. Both computers, however, took just about 1 second to return from sleeping. For this reason, I didn't actually test the wake-up time as it was too short in both operating systems to produce meaningful numbers or even allow me to measure the difference."

The Mac software also unsurprisingly ran Apple's own native applications more efficiently. Converting a movie from M4 format to iPod in Quicktime X on Snow Leopard took 444.3 seconds, while Windows 7, with QuickTime 7 (the latest version available) took 723 seconds. Similarly, converting 17 songs in iTunes from MP3 to AAC took 149.9 seconds in Snow Leopard, while Windows 7 required 162 seconds.



The test also found that Mac OS X 10.6 had better battery life on the MacBook Pro than Windows 7. The 2008 model has a removable battery. But author Dong Ngo said he believes Boot Camp drivers were mostly responsible for the Windows 7 battery life, as many PC laptops fared much better than the 77 minutes the Microsoft OS fared.

One area where Windows 7 was able to easily trump Snow Leopard was in graphics performance. The system's 512MB Nvidia GeForce 9600M GT graphics card helped the system score much better in the latest version of Windows, earning a 5,777 3D rendering score in Cinebench R10. Snow Leopard scored 5,437.

In testing Call of Duty 4: Modern Warfare, Windows 7 again came out on top, with an average 26.3 frames per second performance, compared to 21.2 frames per second within Snow Leopard.

Ngo's conclusion: Unless you are a gamer, get a Mac.

"If you can get by with just software designed by Apple and if money is not a big issue, you will be happy with a Mac," he said. "Examples of these software choices are iTunes, iLife, QuickTime, Safari, iChat, and so on (and you probably won't need much more than those for daily entertainment and communication needs). Finally, if money is not an issue--and it definitely is for most of us--you should get a Mac anyway. It's the only platform, for now, that can run both Windows and OS X."

See also:

Windows 7 vs. Mac OS X Snow Leopard

Exploring Windows 7 on the Mac

Inside Mac OS X Snow Leopard

[Quadra 610]

Hardly surprising. OS X is designed for Apple hardware.

[iekozz]

A Apple OS is faster on a Apple machine.. WOW I did NOT expect that! /sarcasem I'd say that a Windows 7 is faster on a Windows laptop. But oh wait, os x can only run on Apple stuff.

I find this test a waste of time for something we already knew.

[AppleStud]

Originally Posted by AppleInsider

Unless you are a gamer, get a Mac.

Groundbreaking work they're doing over there at CNET.

[zeasar]

actually, windows runs faster on my mac then on a regular pc.

but i think they really need the same version of itunes for a fairer result.

[Hattig]

This is barely a test! There are so few test points, and the ones that are chosen are dubious because they favour one manufacturer (iTunes, Quicktime, using MacBook Pro with suspected dodgy Win7 drivers).

If iTunes used Grand Central to dispatch encoding tasks, or OpenCL for the tasks themselves, I could see why it would be faster on Snow Leopard. Clearly it doesn't.

The gaming result isn't unexpected, and is probably due to lack of platform optimisation in the game itself, and possibly slightly less performant drivers and using OpenGL instead of DirectX (I don't know what this game uses on Windows, but if it is DX, then porting to OpenGL quickly would be less than optimal). We can't expect improvements here rapidly however, however the rise of console gaming vs windows gaming is allowing the Mac's graphics hardware to be competitive with the fixed hardware over lifetime consoles, and hence the games that are written.

How long is the wake from sleep, when it includes reconnecting to the wireless network?

[Cransy]

...that historically Apple Hardware has been shown to run windows more quickly than "PC" laptops. Let's also not forget that for all intents and purposes a Mac is essentially a PC, down to the intel processor. I think this is a fair test between the two OS's given the afore mentioned items.

[bdkennedy1]

Anyone with common sense can see this is biased. How about comparing Office versions?

[Hutcho]

Originally Posted by zeasar

actually, windows runs faster on my mac then on a regular pc.

You have a PC with exactly the same specs for as your Mac? Unlikely.

What would be more fair is getting a PC running Windows 7 for $2000, and a Mac running OSX 10.6 for $2000 and see which is faster. The Windows machine would win convincingly.

[Lorre]

Originally Posted by bdkennedy1

Anyone with common sense can see this is biased. How about comparing Office versions?

Indeed. Or iTunes encoding on Snow Leopard vs WMP encoding on Windows 7. WMP o, modern hardware is much more snappy than iTunes on the Mac is...

[BenRoethig]

Originally Posted by Quadra 610

Hardly surprising. OS X is designed for Apple hardware.

No, it thrives in spite of Apple Hardware. Also, there is no such thing as Apple hardware anymore, they died with the PowerPC. They design ways to cram other people's parts into pleasant looking forms.

[Shookster]

This report assumes that the applications are fully optimized for each operating system and the only difference is the OS itself, but in the real world this is obviously not the case. There are a lot of additional factors here that are being ignored.

[doyourownthing]

it would be interesting if they gave these computers to actual users and ran the same tests after 1 year

would like to see how the windows computer performs with a bloated registry and viruses

these tests are a joke

[doyourownthing]

Originally Posted by Hutcho

You have a PC with exactly the same specs for as your Mac? Unlikely.

What would be more fair is getting a PC running Windows 7 for $2000, and a Mac running OSX 10.6 for $2000 and see which is faster. The Windows machine would win convincingly.

i can tell you first hand, a windows on a 4-core processor will not outperform snow leopard in a 2-core processor...this is just from my everyday use...

[Hudson1]

Originally Posted by Hutcho

You have a PC with exactly the same specs for as your Mac? Unlikely.

What would be more fair is getting a PC running Windows 7 for $2000, and a Mac running OSX 10.6 for $2000 and see which is faster. The Windows machine would win convincingly.

OK, but make sure that Windows machine has equivalent software installed (including the version of Windows itself) for a grand total of $2000. You're not looking at a "$2000 Windows computer" anymore.

[ascii]

Just goes to show how hard it is to compare the two platforms. You have to use Apple hardware, since you can't legally use anything else, and you have to use apps common to both, many of which are written by Apple. It will inevitably be called unfair to Windows but how else could you do it?

Edit: actually a better way might be to make a common list of tasks a home user might perform and then do them on both platforms using whatever apps are available on both platforms.

[teckstud]

Originally Posted by Lorre

Indeed. Or iTunes encoding on Snow Leopard vs WMP encoding on Windows 7. WMP o, modern hardware is much more snappy than iTunes on the Mac is...

...or surfing the internet using Flash sites.

[Hutcho]

Originally Posted by doyourownthing

i can tell you first hand, a windows on a 4-core processor will not outperform snow leopard in a 2-core processor...this is just from my everyday use...

What speed processor and running what? Your blanket statement means nothing. Of course, in apps that don't take advantage of multicore processors, then a 3ghz Dual Core is going to be much faster than a 2.4ghz Quad Core. But when the program does, the Quad Core is going to kill the Dual Core (in CPU intensive tasks).

[DKWalsh4]

Originally Posted by teckstud

...or surfing the internet using Flash sites.

Actually, using click2flash ignoring all the pointless ads, my Mac runs Flash sites much faster then my windows machine. And YouTube is using my built in H.264 capabilities, not the resource hog that is your beloved Flash.

[buceta]

Despite the naysayers, this is a fair test. In fact it is biased towards PCs because it has been PROVEN windows and applications for windows runs faster on a Mac than PC.

It is a fair test for the PC. It still does not answer the bigger questions: i) which is the most productive and satisfying computer to use?

For instance, the test does not take into account the ridiculous amount of time you will be spending upgrading each and every 3rd party app including antivirus on your windows box. Something that does not happen in a mac since applications are fewer (since they accomplish more with less) and the upgrade path is unified (via Software Update...)

Ngo's conclusion: Unless you are a gamer, get a Mac.

Correct again. I have been saying this all along.

[suckerpunch86]

what would the results be if Apple delivered Windows-7 optimized and compatible drivers?

[Povilas]

Originally Posted by buceta

Despite the naysayers, this is a fair test. In fact it is biased towards PCs because it has been PROVEN windows and applications for windows runs faster on a Mac than PC.

It is a fair test for the PC. It still does not answer the bigger questions: i) which is the most productive and satisfying computer to use?

For instance, the test does not take into account the ridiculous amount of time you will be spending upgrading each and every 3rd party app including antivirus on your windows box. Something that does not happen in a mac since applications are fewer (since they accomplish more with less) and the upgrade path is unified (via Software Update...)



Correct again. I have been saying this all along.

Or you can run Geekbench and everything will start to make sense. Another problem is that Call of Duty 4 is not optimized for Snow Leopard.

[teckstud]

Originally Posted by DKWalsh4

Actually, using click2flash ignoring all the pointless ads, my Mac runs Flash sites much faster then my windows machine. And YouTube is using my built in H.264 capabilities, not the resource hog that is your beloved Flash.

I tried click2flash for 1 week and couldn't stand the way my web pages looked- like something out of George Orwell's "1984". It looked like I was banned from viewing 15% of the web. I un-installed it last night. I hate to impose censorship just because Apple and Adobe can't solve the problem. I'm not running any other Apps (usually) when I surf at nioght anyway. But I'm glad you can tolerate the look of censorship.

[Povilas]

Originally Posted by suckerpunch86

what would the results be if Apple delivered Windows-7 optimized and compatible drivers?

Apple provides drivers for Windows 7. They are included in Snow Leopard DVD.

[Asherian]

Originally Posted by buceta

Despite the naysayers, this is a fair test. In fact it is biased towards PCs because it has been PROVEN windows and applications for windows runs faster on a Mac than PC.

It is a fair test for the PC. It still does not answer the bigger questions: i) which is the most productive and satisfying computer to use?

For instance, the test does not take into account the ridiculous amount of time you will be spending upgrading each and every 3rd party app including antivirus on your windows box. Something that does not happen in a mac since applications are fewer (since they accomplish more with less) and the upgrade path is unified (via Software Update...)

This really doesn't make any sense at all.

Windows Update will update all MS software on a Windows computer, Software Update will update all Apple software on a Mac. Just claiming that the Mac's software update is significantly different than Windows goes to show how very little experience you have with the platform, and not in a position to judge.

As for "proven windows" applications running faster on Mac than PC -- which ones would those be? None of those on the list are really examples of Windows Native apps. Just more nonsense...

[Asherian]

Originally Posted by Povilas

Apple provides drivers for Windows 7. They are included in Snow Leopard DVD.

They're not optimized drivers. In fact, they're frequently broken (eg, using the microphone port on a MacBook Pro doesn't work in Vista or 7) or very old (like the packaged Nvidia drivers). On my MBP I've had to hack in more modern drivers, and got noticable speed gains out of it.

[Povilas]

Originally Posted by Asherian

They're not optimized drivers. In fact, they're frequently broken (eg, using the microphone port on a MacBook Pro doesn't work in Vista or 7) or very old (like the packaged Nvidia drivers). On my MBP I've had to hack in more modern drivers, and got noticable speed gains out of it.

So they are intentionally cripling those drivers?

[Asherian]

Originally Posted by Povilas

So they are intentionally cripling those drivers?

I wouldn't say that. They're just not going out of their way to ensure the Windows experience is the best it could be. It's probably laziness or lack of caring more than malice.

[Povilas]

Originally Posted by Asherian

I wouldn't say that. They're just not going out of their way to ensure the Windows experience is the best it could be. It's probably laziness or lack of caring more than malice.

How do you explain this:

http://gizmodo.com/317060/macbook-pr...vista-notebook

[S8ER01Z]

Originally Posted by zeasar

actually, windows runs faster on my mac then on a regular pc.

but i think they really need the same version of itunes for a fairer result.

Do you think about your comments prior to posting them? Tell me...what is the difference between the Mac hardware and the PC hardware? Does Intel ship Apple different i7 920s than they do everyone else? Does apple use better DDR3 memory than Corsair offers for PC? Does Apple get higher performing Nvidia cards than what Nvidia offers PC owners? Please... educate me here, what is giving the Apple hardware it's advantage?

[Povilas]

Originally Posted by S8ER01Z

Do you think about your comments prior to posting them? Tell me...what is the difference between the Mac hardware and the PC hardware? Does Intel ship Apple different i7 920s than they do everyone else? Does apple use better DDR3 memory than Corsair offers for PC? Does Apple get higher performing Nvidia cards than what Nvidia offers PC owners? Please... educate me here, what is giving the Apple hardware it's advantage?

I don't know what kind of magic is involved, but Windows 7 boots faster on my iMac compared to friends Dell.

2 Ghz C2D vs 3.06 Ghz C2D
4GB RAM vs 4GB RAM
1TB Hitachi vs 1TB WD

Go figure.

[vachi]

""Apple's Snow Leopard bests Windows 7 in speed tests"""

shoud be ""Apple's Snow Leopard beats Windows 7 in speed tests"""

[jimerl]

i'm a bit puzzled why windows is able to perform graphics operations significantly better.

[S8ER01Z]

Originally Posted by Povilas

I don't know what kind of magic is involved, but Windows 7 boots faster on my iMac compared to friends Dell.

2 Ghz C2D vs 3.06 Ghz C2D
4GB RAM vs 4GB RAM
1TB Hitachi vs 1TB WD

Go figure.

Boots faster? Both machines have zero software installed and are of the same exact specifications in hardware? Please tell me you don't think 'boot time' is a performance number that shows an Operating systems dominance over another?

Do you compare cars by turning the key and timing how long it takes the engine to start up?

[S8ER01Z]

Originally Posted by jimerl

i'm a bit puzzled why windows is able to perform graphics operations significantly better.

Driver maturity for one...DirectX maturity...and it would probably have a lot to do with the developers of the game that was tested. If they initially wrote the COD series for OSX and not Windows I'm sure it would be the other way around by now. They have had a long time to figure out how to get the most from the PC platform running under Windows. As far as I know the OSX platform is relatively new for hardcore gaming and the COD series. (I could be wrong on that though)

[Povilas]

Originally Posted by S8ER01Z

Boots faster? Both machines have zero software installed and are of the same exact specifications in hardware? Please tell me you don't think 'boot time' is a performance number that shows an Operating systems dominance over another?

Do you compare cars by turning the key and timing how long it takes the engine to start up?

Dell with better hardware boots longer than iMac which is 2 years old. So to you it's completely irrelevant? Of course Dell will outperform my iMac in most of tasks I have no illusions about that, but the fact is it boots slower than my iMac. Is it so hard to believe?

[Povilas]

Originally Posted by S8ER01Z

Driver maturity for one...DirectX maturity...and it would probably have a lot to do with the developers of the game that was tested. If they initially wrote the COD series for OSX and not Windows I'm sure it would be the other way around by now. They have had a long time to figure out how to get the most from the PC platform running under Windows. As far as I know the OSX platform is relatively new for hardcore gaming and the COD series. (I could be wrong on that though)

OS X Nvidia drivers are quite bad. I guess it's one of the reasons.

[kpluck]

I think all you really need to know about this test is in the graphic. They only show the tests where OS X wins and conveniently leave out the two tests where Windows 7 wins.

Basically, all the article proves, at best, is that Apple hardware runs Apple software better then Microsoft software.

Here is a test for you. I just compared over 100 different benchmarks tests on my HP Vista Laptop. In every test, Vista out performed OS X. In fact, OS X wouldn't even boot the machine. Thus, the only conclusion that can be drawn is that OS X has major compatibility issues and should be avoided.

-kpluck

[Povilas]

Originally Posted by kpluck

I think all you really need to know about this test is in the graphic. They only show the tests where OS X wins and conveniently leave out the two tests where Windows 7 wins.

Basically, all the article proves, at best, is that Apple hardware runs Apple software better then Microsoft software.

Here is a test for you. I just compared over 100 different benchmarks tests on my HP Vista Laptop. In every test, Vista out performed OS X. In fact, OS X wouldn't even boot the machine. Thus, the only conclusion that can be drawn is that OS X has major compatibility issues and should be avoided.

-kpluck

Looks like BS to me

[mytdave]

I like Macs, and the tests showed Mac in a good way, but those "tests" were pointless and all over the map. Really, there is a large difference between SL and Win7. These types of tests need to be revamped, and conducted later when both OSes have a little time to mature.

Keeping in mind no test will ever be perfect, you have to make some decisions about what you want to test. Do you want to test SL & Win7 performance on Apple hardware, or do you want to test performance on *equivalent* hardware? Are you testing the hardware or the OS? Are you going to test the OS or test Apps running on the OS? Testing with Apps shows the performance of the App on a particular platform, not the performance of the OS. Choose what you want to test for.

If you want to test the OS, then boot/shutdown times are good. Disk access, network performance, I/O, graphics performance and other subsystems are what you look at.

Separately you can test App performance... but seriously, you need to find 3rd party Apps, but those where the developer has optimized for BOTH platforms, not just ported generic code from one to the other.

Based on Windows to Windows testing conducted elsewhere, Win7 is faster than Vista but slower than XP. MacOS X has been shown to be consistently faster in some functions than Windows, but behind in some others (like graphics performance ironically).

What do you need your computer to do? Games will likely always be faster on Windows cause developers optimize for DirectX. Other Apps such as video editing run exceptionally well on Mac. When serious developers optimize for multi-core processing that MacOS does very well, those apps run better on Mac. When developers start to take advantage of the full 64bit, GCD, OpenCL, etc. features of SL, then those particular apps will really spank similar/equivalent apps on Win7.

There is no blanket "faster" system. It's case by case, over time.