Apple MacBook Pro 'Fastest Windows XP Notebook'?

rgraham writes "The Register has a great opening line in a recent article, "Want the fastest Windows XP Core Duo notebook? Then buy a Mac. According to benchmarks carried out by website GearLog, Apple's MacBook Pro running Windows XP is a better Adobe Photoshop rig than any other Core Duo laptop on the market." GearLog ran the same tests that were run by PC Magazine with the Mac coming out on top."

Comments

Best tool for the job

[networkBoy]

Now all I want to know is which is faster: Photoshop on XP or OSX?
-nB

Re:Best tool for the job

[TheRaven64]

Some parts of OS X are much slower than others. System calls are quite expensive (roughly 10x the cost on a conventional UNIX system), for example. The slowest part of the system I have found is the VM subsystem, which absolutely crawls. I wrote some fairly I/O intensive code with a number of back ends. The aio back end is about half the speed on OS X as on FreeBSD on similar hardware. The mmap backend is an order of magnitude slower on OS X than the aio back end, while they are both about the same speed on FreeBSD. This means that anything that causes page faults is going to slow the system down to a painful speed, which is why Mac users always recommend that you buy a lot of RAM.

Re:Best tool for the job

[bunratty]

Differences in cache size, cache speed, disk access time, and disk throughput, among other things, would cause two computers with exactly the same CPU, RAM, and bus to run at different speeds. This is part of the MHz myth -- there's more to how fast a computer is than the speed of its individual parts.

Re:Best tool for the job

[ivoras]

This is actually a nice indication of a (now not so) subtle shift in the industry. Ten years ago a company that produced a unix-like OS with so much lag in a core system would be loughed down and burn. Now, people don't even notice (for the most part). You can make a system with wicked clever algorithms, and still it wouldn't matter because what people are drawn to are pretty colors of the hardware and the UI.

Re:Best tool for the job

[pammon]

Some parts of OS X are much slower than others. System calls are quite expensive (roughly 10x the cost on a conventional UNIX system), for example

I'm not disputing this, but I'd like to provide some context so people aren't left with the impression that "Apple's programmers are st00pid n00bs." There's at least three decisions that negatively impact OS X's system call performance, but that provide wins in other areas.

1) Mach/FreeBSD system call disambiguation. OS X has to support both Mach and FreeBSD system calls through the same trap interface. Determining which you have isn't cheap, but the win is apparent - how many Mach messages per second does your conventional UNIX benchmark at? Features don't come for free. This is fixed overhead which will be especially apparent with "fast" system calls.

2) 4/4 memory split. A system call requires a context switch to and from the kernel's own address space. I'm not sure about other UNIX flavors, but Linux in particular (usually) maps the kernel's address space into each process with a 3/1 split, which is faster but has an obvious downside - 25% less address space for the process and 75% less for the kernel!

3) Dynamic library binding. OS X is unusual in that every library is always dynamically bound, which adds overhead for every call, but gives you all the benefits of non-static libraries (code sharing, security, etc.) Benchmarks often don't take this into account.

The slowest part of the system I have found is the VM subsystem, which absolutely crawls. I wrote some fairly I/O intensive code with a number of back ends.

There's a few things I've found that impact OS X's I/O negatively:

1) Spotlight wants to index any file you opened for writing and then closed. That's obviously going to incur a cost.

2) Unified buffer cache - cacheing reads in the VM system. For a linear read of a huge file, this only hurts; it can be turned off on a per descriptor basis, but code compiled naively for OS X won't have bothered to do that.

3) Bugs. There seems to be a bug where a program doing linear I/O can monopolize the I/O system, which improves performance for that process but decreases apparent responsiveness.

Re:Best tool for the job

[Kadin2048]

True; tests of the OS for server-type tasks have already been done, at least on the PPC side -- people have compared database performance on the Apple XServe when running OS X Server versus Linux, with the same pieces of software and (I think) the same test data, the only thing that was changed between the two was the OS. Result was ... Linux wins, although I've never heard a really conclusive explanation as to why. It apparently has something to do with process scheduling, I've heard.

I suppose you could do the same thing with Apache or MySQL on a Mactel, in order to compare OS X versus Windows, but I'm not sure how useful or interesting a comparison it would be, since those machines (consumer laptops) aren't really used for those kinds of tasks very often.

Frankly I think the best measures of desktop performance are ones having to do with UI latency: I don't care how fast a system crunches numbers, if I have to use it and it takes more than a tenth of a second for it to register that I've pressed the mouse button, it's too slow. Program launch times, UI latency, window scrolling and resizing ... these are all far more important to real desktop users than how many simultaneous threads it can handle in Apache, or how many queries it can process in a MySQL DB.

Re:Best tool for the job

[Breakfast Pants]

Is there a similar switch to turn off VM caching on a per descriptor basis in linux? Right now I have an ugly script for some things which sets "swapiness" to a very low value, does all of the job (right now I just use this for updatedb), and then sets swapiness back. If I code could do this on a descriptor level that would rock (think extracting a large tar, etc.).

Re:Best tool for the job

[communikatsiglobale]

Gentlemen, Wa wa wa wa wa, wa wa wa OSX wa! Wa wa, wa wa wa XP. Wa wa wa wa wa wa wa. Wa wa wa wa, wa wa wa wa! Wa wa OSX wa XP wa wa wa? Wa! Wa wa wa wa wa wa. Wa wa wa wa wa wa wa, wa: 1 Wa wa wa wa wa wa wa. 2 Wa wa XP wa wa wa. Wa wa wa wa wa? 3 Wa wa wa wa wa wa? Wa. Wa wa, 4 Wa wa wa wa? IMHO wa wa wa wa wa OSX wa wa wa wa wa. Wa wa? Wa wa wa wa wa wa XP, wa wa wa wa wa wa wa, wa wa wa wa.

Re:Best tool for the job

[megabeck42]

1, How is supporting Mach and FreeBSD system calls an advantage? Apps should not be making direct system calls and instead invoke the appropriate c library wrapper. This doesn't seem like a feature, it sounds like a bad design decision. What was the reasoning by basing OSX on Mach in the first place? What is the justification for this excessive, per system call overhead?

How many Mach messages per second does my conventional UNIX benchmark at? None. It can't. Does this preclude me from doing anything I need to do? No.

2, The 4/4 memory split only applies to 32 bit environments. Haven't the G3/G4/G5 been 64 bit? I know the current Core Duo's are 32 bit, but the next generation are supposed to support AMD's x86-64 (called EM64T in Intel language.). There's no advantage to seperate virtual address spaces when you have 64 bit architectures. Even on 32bit architectures, a 2/2 or 3/1 split is only a problem if you have more than two or three gigabytes of physical ram or if your application wants to memory map more than two or three gigs of virtual address space. If you need to do that, then get an architecture that supports it. Most people still have less than 2 gigs of ram in their computers, why optimize for those that do when the computers people will be buying when four gigs of ram or more is common will already be 64 bit.

3, Are you suggesting that FreeBSD, Linux, Windows, or any other modern operating system doesn't use dynamic libraries? What is so special about OSX's dynamic loader that Benchmarks need to take it into account?

1, Agreed.

2, What? Please. Yes, there is an advantage to not using the buffer cache in some cases, something you can do in linux with the O_DIRECT filedescriptor flag, but in most cases, it's not worth the trouble. Bypassing the buffer cache spares you the cost of a copy, but if you ever need the block again, you are at a clear advantage if it could have been read out of the buffer cache.

3. Don't know anything about it, but it's quite possible.

Re:Best tool for the job

[rsborg]

You can make a system with wicked clever algorithms, and still it wouldn't matter because what people are drawn to are pretty colors of the hardware and the UI.

Isn't it a bit deceptive to label "Good UI" as "pretty colors"? It's been proven that the OSX UI guidelines, look and feel, is MUCH better than both Windows and especially Linux.

It doesn't matter a damn that a computer or program is 50% faster, as most of the time the process waits on user input... it's making the users more productive and happy that really makes a computer/program solution *faster*... if that means "pretty colors" or "good UI guidelines" or "system stability" makes this possible, then those should really be what matters in benchmarking.

Re:Best tool for the job

[pammon]

How is supporting Mach and FreeBSD system calls an advantage?

There's a lot of historical decisions of dubious validity in retrospect, but there's also an excellent technical defense that can be made for Mach. In short, Mach is really cool. Mach IPC makes signals, sockets, pipes, shared memory, SysV IPC, etc. look positively clumsy. What's FreeBSD's answer to the Mach Inteface Generator? CORBA?

So OS X gets a lot of mileage out of Mach messaging - AppleEvents, distributed notifications, run loops, etc. If OS X processes seem good at talking to one another - think VoiceOver, Spotlight, the window server, iLife's media sharing, even copy and paste - it's due in part to the fast, flexible IPC mechanisms enabled by Mach.

The 4/4 memory split only applies to 32 bit environments. Haven't the G3/G4/G5 been 64 bit?

In principle, yes; in practice, OS X has a 32 bit kernel even on 64 bit machines, not least of all for driver binary compatibility. You want to know the win here - take a look at the binary compatibility driver story on 64 bit Linux or 64 bit Windows. Apple allows 64 bit processes on Tiger without breaking everyone's hardware.

(Incidentally, only the G5 is 64 bit.)

Are you suggesting that FreeBSD, Linux, Windows, or any other modern operating system doesn't use dynamic libraries?

Yes. Benchmarks typically compare statically linked libraries on Linux (because they're faster) to dynamically linked binaries on OS X (because that's all Apple ships).

Yes, there is an advantage to not using the buffer cache in some cases, something you can do in linux with the O_DIRECT filedescriptor flag

Thanks, I wasn't familiar with that flag on Linux. From googling, it looks like it does somewhat different things, in particular, not speeding up sequential file access.

In any case, I'll certainly agree that there Linux-specific filesystem optimizations; I was just commenting on a technique I found to give a substantial boost to OS X programs with sequential access patterns.

Re:Best tool for the job

[localman]

It's not pretty colors, it's total utility. That rarely correlates with raw speed or cleverness of the algorithm. My guess is that most slow bits in the system were conscious choices where the benefits outweighed the costs. I'm sure there's a thousand things they could improve, still, but overall it's the most productive system I've worked on.

And don't hate it just because it's beautiful :)

Cheers.

Re:Best tool for the job

[Kalak]

time to market is probably correct, as OS X is a continuation of NextStep, not a write from scratch. Now NextStep chose Mach for different reasons than time to market I suspect, but in speaking strictly of OSX you have to remember its original origins (part of the reasson Intel compatability is a simple issue for the OS, since the dual life is quite historical for NextStep).

Re:Best tool for the job

[Halo1]

man fcntl -> F_NOCACHE

AMD

[Eightyford]

It would be nice if they tested AMD notebooks.

Re:AMD

[hawkbug]

Yeah, you're right. I only build AMD systems because they run cooler, faster, etc, etc - when compared to the Pentium 4 that is. With this new Intel CPU, things have changed. AMD needs to respond quickly with a good dual core notebook product. Unfortunately, they still haven't mastered the 65 nm process yet - so it will be a while. My question is, is the CoreDuo 64 bit? If not, it seems rather stupid to buy one of these right now.

Re:AMD

[jonnythan]

AMD doesn't make any dual-core notebook chips...

Re:AMD

[IronTek]

AMD will release their Turion64 X2 dual core processor pretty soon. But I am impressed with the Intel CoreDuo. It does not seem to be (and is not) the POS that their first "dual core" processor was.

Re:AMD

[hattig]

Current Core Duo (Yonah) is 32-bits only.

AMD will be releasing 25W Dual-Core Turions in May, running with DDR2 memory (which will save a few Watts over DDR memory).

Yonah is 31W (TDP, actual power consumption is lower. Same goes for AMD of course.). AMD includes half of a northbridge on their processor as well.

Of course, AMD's 25W Turion X2s only come in 1.6GHz and 1.8GHz variants. The 2.0GHz and 2.2GHz versions are 35W, but still comparable in power consumption to Yonah. The interesting thing is that this is at 90nm. If AMD has any of the hi-speed, low-power-consumption features of IBM's 65mn process, then next year could be very interesting however.

Doesn't negate the fact that Intel was there first, nor that AMD isn't overtaking them but merely having a competitive offering in the mobile arena.

Re:AMD

[loony]

Not really - AMD doesn't have dual core laptops out yet and even if you look at just one core, the Intel Core architecture beats anything AMD has right now hands down...

Peter.

Re:AMD

[Eightyford]

AMD doesn't make any dual-core notebook chips...

That doesn't make comparisons impossible. Who cares how many cores there is. People want speed.

fastest in one test

[gEvil (beta)]

Fastest WinXP notebook for the Photoshop test. It doesn't look like it fared so well in the Windows Media encode test.

Re:fastest in one test

[MustardMan]

Am I the only one who feels a little dirty reading about encoding windows media on windows xp running on apple hardware? It just feels so wrong.

Re:fastest in one test

[drinkypoo]

The graphics card isn't involved in media encoding. Well, there ARE schemes that involve it, but it's not normal. Of course, if they were actually displaying the clip while they encoded it, that could possibly do it - but that's a silly thing to do unless you're doing a very short clip and you want to see what the compression artifacting looks like as-you-go.

Re:fastest in one test

[gEvil (beta)]

And please tell us what portions of the video encoding task are handled by the GPU.

Re:fastest in one test

[rob_squared]

Well, creating .asf or .wmv files would make anyone want to shower repeatedly.

I'm just waiting for the .asf/.wmv support groups* to start popping up.

And please mods, realize that suggesting a support group for a file format is indeed a joke.

Re:fastest in one test

[megabeck42]

You wouldn't need to display the clip while encoding it to take advantage of the graphics card. The whole gpgpu system, general purpose graphics processing, uses off screen rendering targets such as the OpenGL pixel buffer object or frame buffer object. It would be straightforward, with an OpenGL 2.0 compliant graphics card, to offload the colorspace transforms, fast/discrete fourier transforms, and a few other compute intensive algorithms.

But, you are right when you said its not normal. Hopefully, as OpenGL 2.0 becomes more ubiquitous, we'll see more useful offloading.

This is weird...

[__aaclcg7560]

Why did it had to be Microsoft confirming that Steve Jobs was correct that the Intel Mac was a lot faster than the PPC Mac?

Re:This is weird...

[soft_guy]

"Your decision to use Intel's chips instead of PowerPC chips was a wise one" - MS

"Your decision to use our chips instead of PowerPC chips was a wise one" - Intel

"Don't the door hit you on the way out" - IBM

Bah! Plugins are not a real benchmark...

[tpgp]

Fastest at running certain photoshop plugins :-/

Still - yet another reason to not dismiss windows-on-mac-hardware efforts.

Find out next year

[Midnight Thunder]

Now all I want to know is which is faster: Photoshop on XP or OSX?

That will have to wait until next year, sine Adobe has stated that the Intel version of Photoshop for MacOS X won't be available until next year.

Re:Find out next year

[man2525]

Adobe does not want to make a MacIntel version of Photoshop CS2. They want to release Photoshop CS3 as a Universal Binary. They are trying hard to have a release by the end of this year, but they need to move their development environment from Metrowerks Code Warrior to Apple XCode.

Re:Find out next year

[consonant]

Hm. I was thinking of the year-long wait too, but only because Photoshop takes frickin' ages to open...

Re:Find out next year

[ceoyoyo]

Does Carbon port as well as Cocoa? I know it runs fine, but I had the impression Carbon allowed a little more leeway in how you did things... leeway that wouldn't necessarily port quite correctly.

I wonder if Adobe was quite as fanatical about doing SSE optimizations on their code on Intel as they were with Altivec on PPC. As others point out, Apple always used them as a benchmark so their code was definitely optimized. I wonder if they did that to the same extent on Windows, where they got considerably less attention (though possibly more sales).

You're probably right though, copy and paste job. How come it's taking so long? ;)

Re:Find out next year

[statusbar]

I think you are confusing carbon with 'classic'.

--jeffk++

Commercial Offering for Dual Boot

[RunFatBoy.net]

Now that the Mac is showing off it's quality hardware and such, as the Intel models become commonplace, I wouldn't be surprised to see a couple of commercial offerings for dual boot between Mac and Windows.

There's an opportunity for business to finally transition to a quality hardware platform/OS, and I hope someone steps up to the plate to make a formal solution in this area (not that I don't appreciate the current hacks offered).

-- Jim http://www.runfatboy.net/

Re:Commercial Offering for Dual Boot

[MindStalker]

Why would you pay for an ipod when you can build your own MP3 player with a altoids can and some electronics parts. Geez Doesn't sounds like a a great buisness plan to me.

Thats really not a fair comparison I know. But people will pay a premium for a preconfigured system with good support. Hell I quit building my own machines are work because I just have the time to support them, and just order from dell (work for a university grant, so dell sees us as the university which means we get top tier support)

Why photoshop?

[Locke2005]

Because photoshop is one of the few applications out there that is actually designed to take advantage of multiple CPUs by splitting up the work.

Re:Ummm...

[MustardMan]

Ya know, I'm pretty much an apple zealot... but the biggest thing they do to piss me off is include far too little ram in their systems. I bought a powermac dual g5 that came standard with 512 megs of ram. This is supposedly a top of the line powerhorse, and I paid a price premium for it. The LEAST they could do is throw in a couple sticks of ram to get the thing up to par. Applications on an imac launched every bit as fast as those on my top end dual processor beast. After I threw an extra gig in there, the machine started really smoking - like it should have off the factory floor.

Re:How things change.

[ccollao]

That's not completely accurate. The fisrt company into make USB massive was apple. Regarding PCI, several years ago they introduced it into their PPC market. Gee, even Sun did (altough, PCI was slower than other buses but just plain cheaper).

The other Technologies before mentined, AGP, PMT, SMP Protected memory never said so. About intel, well different story, but with your comment you are just trolling (me thinks).

Macbooks are also the most expensive

[Enrique1218]

I have been shopping around for a notebook for a family member. I found that Lenovo and Apple have the highest price dual core. Dell is of course the lowest. But looking at the specs, the lower price ones tend to have GMA or ATI Hypermemory GPU, slower memory, and are pretty bulky. Apple does put in the best stuff available at the launch. I would even venture to guess that the Macbooks are gaming quality.

The Reg sexed up our dossier

[saschasegan]

Just wanted to preemptively strike out and mention that the Reg "sexed up our dossier" a little, to use a British reference.

Over here at PC Mag/Gearlog (it's the same thing - Gearlog is the blog of PC Mag) we like to say that our tests show Apple makes a "fast" Windows machine, not "the fastest." As somebody else pointed out, while the MacBook squeaked out a win on the Photoshop test, it came in behind other Core Duo laptops on the Windows Media Encoder test. But the news in my mind isn't a one-second difference in this or that. It's that Apple's machines run Windows comparably to the best designed-for-Windows machines. That bodes very well for folks who want to have the best of both worlds by running both OSes natively.

We couldn't run 3DMark, Sysmark, etc. because of the missing video drivers - wouldn't have been fair. The Photoshop and Windows Media tests were the only ones of our standard benchmark suite we thought would generate results that made any proper sense, because they hit processor/disk/RAM rather than video.

Also, for the AMD fanboys, we haven't tested any AMD dual core notebooks yet, so we didn't have the data to compare those.

If you haven't already, read our original story: http://gearlog.com/blogs/gearlog/archive/2006/03/2 1/8212.aspx

Re:The Reg sexed up our dossier

[saschasegan]

Apple has a tendency to heavily customize their machines, and one of their selling points is a tight coupling between hardware and software (namely, OS X.) So we wanted to make sure there was nothing in the Macs that would have prevented XP from running to the limits of the performance of the hardware, and to prove that a dual-boot solution could be both viable and desirable. I'll personally wait for the video drivers to call it "desirable," but we're safely within the realm of viable.

Running these benchmarks also allowed a direct comparison between Apple hardware and other manufacturers' that always used to be cloaked a little by the difference in OSes. Now of course you can argue that the driver situation may have affected our results, but I hope this will be only the first of many data points. It's a start.

Re:How things change.

[99BottlesOfBeerInMyF]

A few years ago, the Mac crowd said there was no need for stuff like PCI, AGP, PMT, SMP, protected memory, Intel, USB, etc. etc....

Ummm, what? More than a few years ago macs already shipped with USB and PCI by default. Heck macs had USB before anyone else was producing a significant number of peripherals for it. The only item on this list I ever heard people argue against was Intel (as in processors).

But just how is a Mac running x86 and Windows XP, a Mac?

Macintosh is a brand name. How is a Dell Inspiron running Linux still a Dell Inspiron? The answer to both questions is that is the name under which it is sold.

Re:Apple no happy

[Midnight Thunder]

I bet Apple is PISSED right now. They're handing all their technology over to Microsoft.

But Apple is get paid $$ for the hardware, so they can't be that annoyed.

Re:hmm is it released now

[jinushaun]

Here is the official Windows XP on Mac website: http://onmac.net/

The patch is available here: http://download.onmac.net/

Re:Ummm...

[ceoyoyo]

Minimum memory configuration on a MacBook Pro is 512MB and 1GB for the 2.0GHz model.

Re:Ummm...

[ceoyoyo]

That's GOOD. Nothing like ordering a G5 with 256 MB, throwing that away and putting in 8 GB or whatever from a commodity memory place. Oh, and saving $1000 while you're at it. I think Apple puts the base configs on their workstation machines really low on purpose because they don't want to be bothered running a big memory business (no profit) so they're tacitly encouraging you to go buy your own memory.

Re:Why?

[hahiss]

Dude, its because the Apple has speed holes

Re:Why?

[99BottlesOfBeerInMyF]

Why should the MacBook be any faster then any other DuoCore notebook out there.

Because each laptop uses slightly different hardware. They use different brands, with different specs, and in different configurations. For any given test, one will win. If you read the article you'd know Macbook Pros scored about the same as the best other Duo Core notebooks out there. Sure they took first in a given photoshop test, but not by a really significant margin. They did worse in some other tests. There are no conspiracies here.

People willfully misinterpreting this test should be ashamed of the FUD they are spreading. This does not prove MacBooks are the "fastest" laptop. It proves they are (aside from the non-existant video drivers) as good as anything else out there for running Windows. This is good news for people who plan to dual boot. This is a good sign for those interested in emulating/VMing Windows. It is just trivia to anyone else.

Currently

[Upaut]

The solution of many problems, by having a Windows partition on ones Macbook, does have a few issues that will both effect preformance, and ones comfort. With the GPU not having any drivers yet, the CPU is doing all the work. So slower animations, more heat (massive amounts) being generated, and an inability to play any games. Now, I am still glad that I have this partition, so I can use a lkot of "Windows only" software my work/school wants me to be able to run, but until the graphics chip is running, I don't think most benchmarkes will be really reliable. That and while running Windows, until a driver is written, I really recomend that you don't have the machine in your lap, unless its a really cold day...

Other issues that are less important are:
*Trackpad does not work
*That little camera doesn't work

Adoption

[Truman Starr]

Ok, I am all for the Mac - my first 'puter was a II-GS back when I was 6 or so. But I haven't been following the hardware "jam XP into a Mac" soap opera like some have, so I'm not sure of the complexity of getting both OSes to share one living space (Talk about "The Odd Couple").

I am, interested, however, in hearing about it as it pertains to adoption by non-techies. I read /. but I've never had a dual-boot system myself. I have a Powerbook, an Ubuntu box, and my company thinkpad, so I've never needed to. Each box does its particular tasks, and does them pretty well (with the exception of the T23 my company insists is SOTA).

However, the specs from this article look quite promising. Like many of you, I salivate at the thought of running not only WoW on my MacBook, but games from developers who don't touch OSX. I'm not foolish enough to presume I'm in any kind of majority on that, but I think it has ramifications beyond the hardcore. I think when the new intel iBooks come out, they will be the perfect computer for just about any non-technical person; i.e. students, moms, grandmoms, whomever. If you can give them something familiar, adoption is going to be 1000 times easier. I'm not asking that Apple blow away other OEM's while running windows. The fact that it comes close (in all of the tests so far) is good enough for me. And grandma too.

Benchmarking the compiler

[Rob Y.]

Ultimately, a Windows vs. Mac CPU benchmark on the same hardware would amount to a comparison of the code generated by the respective compilers.

Don't know how fast the code generated by the Visual C++ compiler is, but I've read that the proprietary Intel compiler generates much faster code than gcc, which (I think) is the default compiler for OS/X apps these days. Does that bode poorly for the Mac in any benchmark wars?

What a retarded article

[greg1104]

It's been widely noted that the basic hardware in the MacBook pro is nearly identical to that in the Acer model mentioned in TFA; see http://www.everymac.com/systems/apple/macbook_pro/ faq/technical_performance_2.html for a rundown. So it's no wonder the run-time is the same.

The appropriate conclusion here is "Macbook Pro runs XP as fast as the fastest PC with the same CPU and chipset", to which I would say, duh!

Photoshop Test

[chowhound]

I've always felt the photoshop tests were an absurd measure of a computer's speed. I run Photoshop CS1 on my G4/400 1GB at home. The only time I ran into a problem was attempting to work on a backlit movie poster for a theatre - 3x5 foot by 300 dpi, with layers, effects & filters. But that is an absurdly huge file. As a designer for 10 years, I never encountered a file that big.

The point is that today's computers are overpowered. The now-deprecated Quad 2.7 G5 is vastly more powerful than any Photoshop jockey needs. Unless you're rastering 3D shiz or crunching a full length DVD-quality movie (neither of which requires Photoshop) it's just gonna be an issue for most users.

Re:Photoshop Test

[MacBoy]

I completely agree.
Photoshop used to be a great benchmark, because computers used to be sllllooowwwwww. Remember when Photoshop power users could drop a few thousand on a fancy DSP card for their Mac? In fact a couple of Macs (the Centris 660Av and the Quadra 810AV ca. 1992 or so) came equipped with a 25/33 MHz DSP on-board to handle certain realtime stuff, like softmodem. Adobe didn't waste any time supporting this DSP to accelerate Photoshop, with a pretty sizable improvement. The point is, people used to waste so much time waiting for Photoshop, that anything, absolutely anything, that could improve its performance was a godsend.

These days, modern CPUs are real powerhouses. I have an older Mac (Dual 866 MHz G4, 1 GB RAM), and I have never cursed at it while waiting for it to complete a PS operation. I have never had to wait. Today, any CPU is pretty much fast enough for PS work... It's the RAM you have to worry about. The instant PS starts hitting the disk during an operation, you might as well have a P100 in there.

Re:How things change.

[grand_it]

Ummm, what? More than a few years ago macs already shipped with USB and PCI by default. Heck macs had USB before anyone else was producing a significant number of peripherals for it.

Ooops, they're doin' it again :(

Apple dumped Pc Card in favour of ExpressCard way too early, imho.

Sure, for many purposes (Ethernet, modem, Usb, Firewire...) Pc Cards are quite obsolete today, because all this features come free with every notebook (except for the modem in MacBook Pro)

Here in Old Europe, however, GSM/Umts connect cards are quite popular among execs and road warriors. As of today, there is no such thing as an ExpressCard GSM/Umts modem.

Heck, the PCMCIAssociation lists a whopping-fifteen-items-list of available modules in his website...

Why a laptop?

[cloudmaster]

So I'm curious, why does Photoshop being faster on one laptop than another mean anything? Surely if you care about all-out photoshop performance, you'll have a desktop machine with a real power supply to drive real processors, room for real memory, and a real display? This laptop's slower for almost everything else, and not appropriate for the onething it's faster at.

Yay benchmarks. :( I'd be more imperssed if they laid the laptops out on a table at a college library and timed which one got stolen fastest. That'd test the *real* value of each laptop...

Re:Ummm...

[temojen]

oh good god. a 8MP image can be printed at 20X30 and look better than 35mm film.

And a 35mm image can be printed at 20x30 and look better than an 8MPixel digital sensor image. Untill you specify the specific cameras, lenses, tripods, subject, lighting, scanner, scanner software, raw converter, film, processing lab, camera settings, postprocessing steps, printing technology, and intended use, it is you that's trolling. In the mean time I'll continue using what works for me.

Most pro photographers do not shoot at more than 6MP because THERE IS NO USE for higher res right now.

It really depends on the intended use of their photos. Newspaper and portrait photographers shoot low-rez because that suits their needs. You can be sure though that PlayBoy's feature photographers are shooting full-frame digital at least (although I suspect medium-format Kodak Portra NC judging by the contrast, tonality, and colour balance).

I often choose 35mm Print film because it gives me resolution slightly better than I'd get with a 1Ds, but much nicer exposure lattitude. Plus I get smaller depth of field than with a sub-frame digital, without having to shell out $20,000 for a 1Ds and a bunch of new lenses.

Pros are not rushing out to buy new digitals they are getting FANTASTIC results with 6mp right now.

You seem to be confusing newspaper photographers with all pro photographers. It depends on their intended use. Fashion photographers are just starting to go digital (from MF) with the introduction of full-frame digitals and digital backs for MF.

you weenies that think that megapixels are everything are getting on my nerves.

People who think that how they use their camera is how everyone uses their camera, and what they expect from prints is what everyone expects from prints get on my nerves.

Re:Apple no happy

[Gilmoure]

Apple just wants to sell boxes. Whether they're little music players or laptops or a Jonathon Ives designed toaster, they get happy when people buy their stuff. What you do with it afterwards is your concern.

Re:Apple no happy

[Anonymous Coward]

No, nobody has ever noticed how vendors charge more for parts that you could buy separately by yourself. Thanks for letting us know. Seriously.

Re:How things change.

[rizzo320]

You are correct. The iMac G3 was the first to have only USB ports. The Blue & White Power Mac G3 was the second, followed by the "Lombard" PowerBook G3.

Although Apple may not have been the first to use USB, they were the first to remove the legacy ports to force peripheral and accessory manufacturers to introduce USB based devices. They were also one of the first computer manufacturers to encourage the ports use. I remember installing multiple labs of Dell Optiplex Gn+ and GXi workstations with USB disabled by default in the BIOS. It was until a year or two later that USB was enabled by default on all of their Optiplex models. Plus, Microsoft's OS USB support really didn't work well until Windows 98 (for DOS based) and Windows 2000 (NT based OS) were released.

Re:How things change.

[soft_guy]

Yeah, the Mac only got PCI what...12 years ago? (PowerMac 9500 was the first PCI Mac.) Prior to that, they had NuBus which was basically the same thing, but it lost out to the PCI standard.

Apple was the vendor that really caused USB to take off...8 or 9 years ago.

And let's lump Intel in there with protected memory.

The Reliability Factor

[lullabud]

I really don't think that Apple's are overpriced when you consider how reliable they are. I'm still using the original 12" powerbook which I purchased over 3 years ago and it still runs fine. Being an IT guy, I've seen numerous Gateway, Dell and Alienware laptops fall apart in that time. I'd personally take an IBM or a Dell if I couldn't go with Apple, but I go with the higher priced Apple hardware because it's more reliable. I'd rather have a machine that doesn't require replacement parts every 6-12 months of 24/7 usage. Also, target-disk mode (only available on Apple hardware, afaik) is invaluable if you ever need to back up your entire hard disk, repair it, or do any other maintinence on it that requires full read/write permission to the filesystem.

This just in!

[HitScan]

In a news story released earlier today it was noted that 2.16GHz notebooks are indeed faster than 2.0GHz notebooks. Citing aspects of their construction such as "This one appears to have the bigger number," and "I guess they used a faster chip in that one," critics rained high praise upon the faster equipment. It is currently unclear whether the testing method employed will encourage more individuals and business to drop what they're doing and purchase their own faster laptops from this rebel manufacturer, or if they will demand that their current supplies "get them some of those faster ones."

Apples and oranges, truly

[SilentChris]

One should also note that the machines they compared didn't even have the same hardware. The Mac is a dual-core 2.16 GHz machine while the PCs were 2.0 GHz. Not to mention discrepancies with hard drive speeds, video cards (including the non-existant XP drivers on the Mac), etc. It's just not a good comparison by any stretch of the imagination.

A more valid comparison would be SPEC tests between the MacBook and other machines. What you'd likely see is, given the same hardware, they perform exactly the same -- which is the point.

As someone pointed out, most geeks would be interested in a box that runs both XP and Mac OS equally well. Apple is in a big transition year: with Vista delayed and the switch to Intel, they finally have means to court a massive number of geeks to their platform. Some random people claiming the MacBook is somehow "faster" than PCs with different hardware damages this. Geeks will look at the specs and know it's not a valid comparison. Mac fans just need to sit tight and let the benchmarks speak for themselves.

Re:Apples and oranges, truly

[n8_f]

And yet the Apple was 37% faster on the Photoshop test against 3 of the laptops. That is much better than the 8% advantage it has in processor speed. And with the crappy drivers, it has one hand tied behind its back. I am very skeptical of that result, but it is possible that further tests will bear these results out.

A more valid comparison would be SPEC tests between the MacBook and other machines. What you'd likely see is, given the same hardware, they perform exactly the same -- which is the point.

Umm, if they were the exact same hardware, then yes, they would perform the exact same. But the point is they aren't the same hardware. If they were all using the same processor then you would expect them to perform similarly on a processor-limited test like SPEC, but the rest of the hardware could have been customized. So some subsystem could perform better and lead to better real-world performance on tests that use that subsystem. As you'll notice, the Acer scored almost the same as the Mac on the Photoshop test. And yet, they use different processors. Is it possible that that test isn't processor-limited and instead stresses some other part of the system? Maybe it is the Acer's better video card, but if the Mac is using generic graphics drivers and getting similar, valid results, it would point to something else. Perhaps they both have 7200 RPM drives. But if they both have that as an option from the manufacturer and the others don't, that has some value. Yes, you can always add one after you buy, but testing manufacturer's configurations is a valid test. The point is that the Macs could be faster than other laptops in real-world XP performance. This isn't an ideal test, but it is an interesting result.

Re:Apples and oranges, truly

[saschasegan]

The conclusion we came to was that Apple makes fast Windows machines, that compete well in several tests with a range of other consumer purchaseable, recently released PCs.

That's very interesting, because it tells folks that they have a good chance of buying machines that will dual-boot the two OSes, getting market-competitive performance in both.

That's never been the case before, and even once Apple went to Intel there were murmurings that either Apple or Microsoft would do something to hurt Windows performance on the new Macs. Our results go some distance towards disproving that.

How is that conclusion invalid?

Re:Did I miss something in TFA?

[quakeroatz]

Apparently I am blind:x

The Macbook pro w/ 2.16ghz Duo ran a photoshop script in 1:10
The Acer Duo w/ 2.0ghz Duo ran a photoshop script in 1:11

The Mac Duo laptop with a 8% higher clocked CPU ran a benchmark 1.4% faster than the Acer Duo.
If anything, this looks bad for the Mac Duo.
Why is it seeing such a tiny improvement when it is 160mhz faster?

I run photoshop ONLY on my laptop

[drhamad]

Laptops have reached a point where they're fast enough for most anything. Heck, if you look at Apple's current offerings, their desktops and laptops use basically the same hardware! (of course, the PowerMac excepted)

Some people such as myself need Photoshop on the go. Others, also like myself, only have 1 license. Third, I have two systems: a Mac mini (G4) and a Thinkpad T40 (1.3 GHz Centrino, I believe). Should I therefore not use Photoshop, since both are (basically - the Mac mini is an iBook) laptops? Should people with iMac's not use Photoshop either, since those systems use Core Duo's?

Low end systems and laptops both passed the point years ago where they were fast enough for almost anything. Sure, Photoshop is faster on a high end G5 or P4 or whatever system, but it's very useable on any modern laptop or low end system.

Re:Dumb question? Dual mouse buttons anyone?

[akhomerun]

control + click

it's been like that on macs forever.

anyone with a laptop should probably get a small wireless mouse anyway, they are way better than using a track pad anyway.

An explanation

[Captain Tripps]

Say you're applying some filters in Photoshop. And say each pixel requires a whole bunch of CPU instructions to calculate. Then you might have what's called a CPU-bound task. But you know your new MacBook has an upgradable processor socket, so you try swapping in a faster one, and indeed your filters finish faster. Good, you say. Let's try an even faster CPU. And you stick in a really high end chip, but now you don't get any more improvement.

What's going on? Each pixel is still taking the same number of cycles, but those cycles are flying by so fast that the memory bus can't move pixel data in and out fast enough. So the CPU is idling part of the time waiting data to process. The point at which this happens depends on the task. Maybe your Folding@Home client will still be CPU bound with even the fastest process.

Now say you've created a bunch of images, and your boss wants them as JPEGS, instead of PNGs like he told you last week. So you run a batch conversion. These are big files, and your JPEG encoder is really fast, so now it's the hard disk that's the bottleneck. Your conversion won't run any faster on a multi-Opteron server, unless it's got a faster disk. And of course, everyone's familiar with the network being the bottleneck.

Does that make more sense? I guess my point is that upgrading a component will only make things faster if it's the bottleneck. And bottleneck component will be different for different tasks.

Mac?

[brjndr]

Ok, I am probably not the first to think of this, but it just hit me...

Powerbooks have PowerPC processors. MacBook's have Intel processors.

So when the PowerMac's are switched to Intel, are they called MacMac's??

Re:How things change.

[djkuhl]

*Bzzzt* Nice try, but wrong.

In 1996, before Steve Jobs came back to Apple, he owned this little company called NeXT. Through this company he developed a relationship with Intel, who were one of the NeXT operating system's chip makers. The next year after Steve's August '97 return, the first new Mac (better known as the original iMac) shipped only with USB. No ADB, PS/2 or any other junker port from the late 80's. I remember many new PCs that my friends brought to college with them still had required PS/2 ports. Around this time Apple also dropped the floppy disk, but had an external drive available via the USB port. Now fast forward through time and you'll also find out that during the second Steve Jobs reign Apple adopted USB and Bluetooth before any other major manufacturer. You see, ever since 1997 there has been a shift to Intel that nobody was really paying attention to until last year.

By the way, Firewire wasn't introduced until after USB was standard on all Macs.

Re:Dumb question? Dual mouse buttons anyone?

[toddestan]

control + click

it's been like that on macs forever.

Yes, that's great. We all know that.

However, control+click doesn't do squat in Windows XP.

Battery life

[solarbob]

Whilst its all very nice having these great process which make for desktop replacements I still can't feel there is a market for a 1Ghz laptop, 512Mb RAM, 40GB storage and a good 12 hours of battery life in a relativly slimline case. The Sony TX1's come pretty close with 6 hours but a full 12 hours would mean you could just not have to charge up during a normal working day

Re:Make up your mind, damnit!

[MysteriousPreacher]

I'll admit I had to look up the exact definition and correct spelling of synergistically (save us from the psuedo-intellectual bullshit words oh jebus) but the rest of it seemed pretty clear.

Overall a perfectly cromulent post.

Re:Dumb question? Dual mouse buttons anyone?

[Eljay451]

Shift-F10 simulates a right mouse click in Windows.