@AGROOM:

EXACTLY! Mac Fanboy's can whine about the 4 GB all they want but what matters is COST!.

If you want a true economic comparison of the power of a PC Hackintosh v/s the overprice fanboy mac then lets do this:

Pay $2000 for a MacBook Pro and Pay $2000 for a Hackintosh.

Then let's see what happens.

Thanks, but I'll be sticking to my hackintosh.

arsenik

@chris-mcc:
Nope.

OmegaRed59

I personally don't think a hardware-to-hardware benchmark is the best test, but cost vs. cost. Take the cost of the Mac Pro or MacBook Pro and then purchase the equalavent in "PC" hardware. The benchmark shows the native Mac hardware doing the better performance, though at cost probably close to 300% more. So of course it's going to out perform.

agroom

@bitfactory: If OSX really requires 2 GB of RAM to work properly, then maybe Apple, the company that everybody tells me prides itself on selling configurations that "just work", should sell them with 2 GB of RAM by default, rather than selling them with 1 and making people upgrade.

Meh, people will buy into whatever party line sounds good, I guess. Me, I'm thankful for the hackers, crackers, and DIYers that make computers so much more interesting and fun.

I'm going to be building myself a rig based on an overclocked Pentium Dual Core chip soon; maybe I'll take a shot at making it into a Hackintosh.

HeartBurnKid

@Aneurism: Because the famed stability and reliability of a Mac comes in a large part due to the locked hardware.

The only way Apple could do this is to release a 'hobbyist' edition of OS X that is unlocked and unsupported.

onidavin

Macintosh should just enable their OS software for PC's, they could make more money this way. Why pay 3000 for a MAC when you could build a PC system with better specs and performance for under 1000 and simply run the famous MAC software on it??? Sure they have a business model of snobery, but in the end there are a lot more average joes with $200 in their pocket to buy a stand-alone OS, then there are rich idiots who are willing to pay a $2000 premium for a certain brand name.

Aneurism

Man people here are really picky. This was not meant to be some kind of definitive guide on how well hacked-up PCs running OS X run compared to true Macs, just information to help people considering the change after his last article.

onidavin

First off, he neglected that the Hackintosh is a "DUAL CORE" processor aka 2 Chips in one. And if that is a new Mac Pro, it would be equal to a Quad (2 Dual core's).
Ram, ram, ram *sigh*
Also the Hackintosh has a newer video card than a year old Mac Pro, if it is the newest model out.

I won't comment on the Macbook, most laptops don't perform as well as desktops!

Next time perhaps he can put a little more thought before he writes.

Loremaster101

While I would love it if Apple sold the OS to anyone with compatible hardware, but they would be crazy to do that. Their market share as the "elite" computer manufacturer would shrink to previous levels, and they would have to charge over $200 (US) for the OS.
And as we all know, there are precious few people willing to pay that much for an OS, no matter how shiny it might be.

btgoss

Has anyone made a dual dual core (2xdualcore or even quad core) Hackintosh?

I'd like to see it go against the Mac Pro, and see if it comes out for cheaper.

I'm remembering the days of those Mac clones. :D

fuzzycuffs

A reader sent in his Xbench marks for a 2.4 GHz iMac with 4 GB RAM:

Below are the Xbench results from my work iMac (~$2700) that seems to be
pretty similar to your Hackintosh at 3X the cost. Thanks for the building
article and the follow up benchmarks, very interesting. I love what
Lifehacker does!

My Mac isn't quite built-to-order, but does feature the following:

CPU: 2.4 GHz Intel Core 2 Duo
GPU: 256 MB ATI Radeon HD2600
RAM: 4GB (2x2GB) DDR2 SDRAM 667 MHz

Xbench testing:
Results 153.60
System Info
Xbench Version 1.3
System Version 10.4.10 (8R4031)
Physical RAM 4096 MB
Model iMac7,1
Drive Type WDC WD3200AAJS-40RYA0
CPU Test 124.49
GCD Loop 282.72 14.90 Mops/sec
Floating Point Basic 136.27 3.24 Gflop/sec
vecLib FFT 100.19 3.31 Gflop/sec
Floating Point Library 88.70 15.45 Mops/sec
Thread Test 240.81
Computation 218.65 4.43 Mops/sec, 4 threads
Lock Contention 267.97 11.53 Mlocks/sec, 4 threads
Memory Test 136.12
System 119.81
Allocate 91.99 337.81 Kalloc/sec
Fill 137.01 6661.57 MB/sec
Copy 145.56 3006.51 MB/sec
Stream 157.57
Copy 145.33 3001.83 MB/sec
Scale 150.80 3115.55 MB/sec
Add 168.67 3593.13 MB/sec
Triad 168.21 3598.52 MB/sec
Quartz Graphics Test 173.97
Line 151.16 10.06 Klines/sec [50% alpha]
Rectangle 184.79 55.17 Krects/sec [50% alpha]
Circle 175.84 14.33 Kcircles/sec [50% alpha]
Bezier 160.81 4.06 Kbeziers/sec [50% alpha]
Text 207.98 13.01 Kchars/sec
OpenGL Graphics Test 124.37
Spinning Squares 124.37 157.77 frames/sec
User Interface Test 466.70
Elements 466.70 2.14 Krefresh/sec
Disk Test 79.07
Sequential 118.11
Uncached Write 128.80 79.08 MB/sec [4K blocks]
Uncached Write 116.27 65.79 MB/sec [256K blocks]
Uncached Read 99.73 29.19 MB/sec [4K blocks]
Uncached Read 133.81 67.25 MB/sec [256K blocks]
Random 59.43
Uncached Write 24.50 2.59 MB/sec [4K blocks]
Uncached Write 115.96 37.12 MB/sec [256K blocks]
Uncached Read 91.25 0.65 MB/sec [4K blocks]
Uncached Read 144.81 26.87 MB/sec [256K blocks]

Adam Pash, LH Senior Editor

1GB on the Mac Pro? That's a crime. OS X needs at least 2 GB to run correctly on those machines. Geesh.

bitfactory

Cloudscout hits it on the head, though the 'fairness' comments should really examine the effect that 1Gb has in the MacPro. I would put money on the MacPro performing noticably (though perhaps not significantly ;-) better with 2Gb of ram vs 1Gb -- purely based on the VM/mem caching techniques that os x uses. that being said. the machines really do need to be running the same OS. (tiger and leopard have differnet memory and VM management code)
I'd love to see you run the tests again with the OS change, and 2GB in the macpro. if you do, please KEEP the original test results, for comparison - im sure it will show some interesting effects that memory and swap usage have on performance.

as for updates killing your hackintosh - who cares. itsonly ever a matter of time before the osx86 community gets the patch out there...

great first pass on this kind of comparison, keep it up!

RogueGeer

Interesting results. If nothing else, it proves that there's a demand for OS X outside of Apple computers. People frequently ask why OS X is restricted to Apple hardware, and the answer is because Apple tries to match hardware and software for the best possible experience. However, it's still not quite perfect.

While I like keeping OS X restricted to Apple hardware and, likewise, refuse to install Windows on my Apple hardware, I don't believe Apple would suffer greatly if they sold OS X licenses for users to install on any machine. Increase the price some to compensate for the loss in hardware sales and slap a "Your Mileage May Vary" sticker on it. Let the users support it themselves.

At least it would give those who want to run OS X without purchasing a Mac a legal way to do so.

EchoD

I gotta say (i'm not the first one)..

Why would you buy a 3k MacPro and only have 1 gigabyte of ram.

It does make sense to buy it that way from Apple, and add memory purchased elsewhere... between two and eight gigabytes.

buckyreal

well, and I fairly certain I am right here, but normal 32bit architecture will not allow for the addressing of more than 4gb of ram, with some ram reserved for the GPU (proly 512mb)and whatever on board RAM it has, so the Hackintosh in this case really only has access to around 3.5gb of ram, assuming that the "Hacked" version of Leopard doesn't support addressing extensions, which in that case would actually hurt the Performance - but help in accessing more RAM. SO the Ram issue really is irrelevant.

shadhimself

I predict your hackintosh will be biatchslapped by an OSX update just like the iphone.

Why even bother when hackjobs are constantly shut down by the man????

ironchef

Why is everybody complaining about the Hackintosh having 4GB RAM ... when the Mac Pro has a Double CPU and the Hackintosh a single CPU !!! 2x2.66 VS 1x2.21

These bench are worthless !

Put 2 CPU in both machine with the same amount of RAM, then redo everything. I bet the double CPU on the hackintosh will make a bigger difference than more RAM on the Mac Pro.

xaqc

I don't think the difference in RAM is as important as the fact that the Mac Pro probably has only two sticks of RAM in it. If you had four matched sticks of RAM in the Mac Pro (two on Riser A, two on Riser B) you get a performance boost.

I'd like to see the same benchmarks repeated with four matched DIMMs in the Mac Pro. I don't care if one machine beats another machine in this or that. I'm just really curious about the actual performance difference between 2-way interleaving and 4-way interleaving.

Cloudscout

I don't see what all the contention is about. It seems to me that OmegaRed59 is exactly right. I don't think Adam's point is to scientifically compare equal processing power but to show what one can pull off with significantly less money put into the mix. I personally wouldn't go this route because I don't mind paying more for style points, but I think he's got a solid experiment demonstrated for us. And whether most people using Mac Pros would want to build their own machines is kind of a moot point. The ones who WOULD want to can benefit from the article. Thanks, Adam!

questionmarke

If the 'hackintosh' wasn't built on the cheap, and instead spent the same budget on PC hardware as a real MAC, it would absolutely blow the MAC out of the water.

Aneurism

@The How-To Geek: I guess I'm poor. Ignore my comment then.

I think it would be interesting to see how the Hackintosh stacks up against computers in its price range. Maybe these Macs are but I don't keep up with Mac enough to know that.

chris-mcc

But part of the point of doing this is it used to take a lot of work and support to get OSX running on generic PC hardware, and it was always crippled in some way. Here we've gotten to a pretty easy, straightforward install on about $900 (pre-rebate) worth of equipment, and it performs about as well as a MacPro.
That's a pretty substantial bang for buck, and if it runs as stable, then I think I've found my next box (with an XP partition for some legacy software of course).

SomeAudioGuy

@OmegaRed59: And you assume that everyone on the internet is from the United States right?

chris-mcc

There is no generic "Hackintosh"
Well... you can built a Hackintosh even stronger than a Mac Pro and do the Benchmarks again

menuhin

@OmegaRed59: Better is somewhat subjective. Or at least you need to consider "better" in various dimensions.

Better in this case means better performance per $. But the Mac Pro has some nice things like those internal hard drive bays. And well support too. I'd venture to say that most people using Mac Pros don't want to build their own machines.

pantsonfireliarliar

@Cidinho:
Here you go:
[www.newegg.com]

That is without even looking for deals.

OmegaRed59

The Hackintosh only cost $800. So it is a very fair comparison considering it is the cheapest computer in the group.

The whole point of the Hackintosh is to show that it is possible to build a better machine for cheaper than Apple's.

OmegaRed59

@The How-To Geek: If I could get, where I live, 1gb for that price I'd get it.

Cidinho

In your hackintosh article you used a core 2 duo processor. Why then do you say that it only has one processor running at 2.21 ghz. Shouldn't both the macpro and the hackintosh be running on duo cores albeit at different speeds.

jadpower

The whole Hackintosh thing is very interesting but I would like to see a fairer fight. As someone else mentioned you could start by taking 2gb out of the Hackintosh. Also you didn't mention what type of processors are in these machines. They are all compared to "Ladd's G5" PowerMac G5 (June 2004) over at Xbench. We know the Hackintosh is an Intel Core 2 Duo but what is powering the Mac Pro and the MacBook Pro? Still I am interested in this project and would like to see your fix for the install DVD boot issue.

GrimWit

how about gaming? running a mac version of Unreal Tournament or World of Warcraft. How about BootCamp on the Hackintosh? how well does that work? i am really interested in making one of these, but if boot camp isnt going to work and gaming isnt up to par with the real deal, then i will either go with a true Mac or just use the machine for windows.

inthepit

Why in the hell does the hackintosh have 4GB of RAM? Not a very fair comparison at all.

dead_red_eyes

if you want to speed up the boot time, you might consider replacing the pc bios with open firmware.

zqfm

The discrepancy in RAM is huge. One gig of RAM in a hinderance new Macs they as well as any PC will run significantly faster with just 2GB of RAM.

I'd like to see a test with even RAM or run them all with 1GB of RAM and see the results.

metabold

The specs may not be competitive, but the price is. The point I think, is that you can do all a Mac can do on a PC (with hacked Lepoard of course) for a lot cheaper, despite the. More benchmarking on similar systems (try it laptops) would be more conclusive though.

sinenomen

Ok that was great, but can you make maybe a level 2 configuration? Maybe some thing in the $1100 range that would be faster or equal to the Mac Pro? Ram can stay the same, but maybe upgrade the processor and graphics card. I would be running Final Cut Pro on it so I would love to beef it up a little bit more. Also have you tried to do a software update yet? Keep us posted. Thanks

digitalnamo

I just bought 4gb of Patriot RAM from Frys for 99 after rebate - ridiculously cheap.

Anyway, you should show some sort of performance per dollar or percent improvement per dollar as that is obviously where the hackintosh really shines.

vsboxerboy

@chris-mcc: You can get 4GB of RAM for about $120 these days... that's really quite cheap.

The How-To Geek

You built a computer "on the cheap" with 4GBs of ram? I must have missed that when I was reading the "time travel with tinfoil" article.

chris-mcc

@jwszoid: It's not meant to be fair and it's not a fight. Or do you understand that and just think it'll be more useful if we compare 3 machines with exactly the same stats doing exactly the same tasks?

Collins1

i dont see any memory intensive test so there is no need to put 4gb to da macbook, well if you have 700$ in you pocket ok try it. The thing is that a mac is a pc now and the mac's fans boys dont want to realize it.

chocolim

yeah a lot of these tests kinda skew in the Hackintosh's favor because of the 4 gigs of RAM.

Falconfire

Yeah why in the world does the Hackintosh in a lot of cases seem much more stacked in terms of RAM? This doesn't seem like an accurate reading of the 3 machines together.

ImTheKing

I'd say pull 2GB of the Hackintosh sticks out to give better results. At least then it would be matched to the MBP evenly in memory.

UnStatusTheQuo

Big friggin shock - having a faster processor and double the RAM makes for a faster computer? How about outfitting the macbook with 4 gigs of ram for a fairer fight?

jwszoid

The cost differences here are too small for me to give up the support and aesthetics of my Macs. I tend to use my machines at lot and they occupy prominent places in my life (and that of my family as well).

For something that I depend upon and will have to look at for several years, a difference of a few hundred dollars is nothing. If I were still a college student in my 20s I might think differently, but for me the days of ugly PC boxes are as gone as the milkbox-and-plywood bookshelves.

The Apple hardware is somewhat more expensive, but IMHO they make it worth the trouble. And it works very well, thank you. Oh, and it is supported.

bradpdx

So all of the people above complaining about 4 GiB of RAM, unfair comparisons, etc. must have failed to read the second paragraph in the article:

"First, it's worth noting that none of these setups is a perfect match for benchmarking comparisons. ... These aren't necessarily the ideal test subjects, but they're the machines I have access to. That said, the results are still very interesting."

Come on, people. Open your eyes.

wffurr

this all goes to prove that if you have a choice awesome cpu or awesome memory you choose memory (unless you are doing video editing but then you need both)

gothika

P.S Apple charge $800 just to add 4GB to an iMac!!

justswitched

Valid comparison to compare a $800 DIY with two current offerings from Apple that are >>$800. It shows that equivalent performance can be attained for a << price.

BTW: I really want to see a DIY 4GHz overclocked Peryn dual quad core PC running OSX, benchmarked!!

justswitched

@Loremaster101: The Mac Pro I compared it with is just a couple months old.

And yes, the Mac Pro is running 2 dual core 2.66 processors vs. the single dual core 2.21 in the Hackintosh.

As for the RAM, the Mac Pro is sold in the default config with 2x512 sticks of RAM, and that's what my friend (who owns the Mac Pro) is running with.

Adam Pash, LH Senior Editor

@bitfactory:
Don't fault the author if Apple sells a Mac Pro for $2200 (base model) and it only has one gigabyte of ram. Doesn't Apple pride themselves on selling Macs that just work? It is stupid if the base model doesn't run well, and the base model is way overpriced.

Think how fast the Hackintosh (this one is $800) would be if they spent $2500 on it and compared it to the base Mac Pro (again, base model with lowest CPU is $2200).

OmegaRed59

@huntermunkey: Right but half the reason the iMac exists is form-factor. Something I truly doubt can be replicated with newegg sourced parts. I'd like to see someone do it for less than the cost of an iMac.

pantsonfireliarliar

This sets out to answer a question that gets a big old "duh" as an answer. Of course you can build a PC that's cheaper than a Mac and offers similar performance. You just get no support, no fancy case, not much sleep functionality, no OS etc.

Another question with the same answer: Can you build a PC using generic parts from newegg that's faster than a PC bought from Dell or Sony? Yes, duh.

Now comparing Apples to (well) apples, Macs are pretty price competitive with other manufactured PCs, excepting memory prices of course.

pantsonfireliarliar

@Adam Pash, LH Senior Editor: I think you should get some sort of LH award for getting everyone so rowled up! It's (mostly) fun to read everyone's comments...
Not to mention 150,000+ views on the original article.

What's the most viewed article on here anyways?

ColinC

if you werent worried about the price could you make a really nice hackintosh? like more ram, or a lot faster procceser?

takemetoyourtoaster

@monkeyboy

It's the best benchmarks tester out there.

huntermunkey

Hackintosh is looking better and better... A computer that runs better than an iMac, for under half the price. If I throw in $200 in extras my Hackintosh will be amazing, assuming I make one.

huntermunkey

BAH!

Xbench is next to worthless! I built a Hackintosh last year, and every time I ran Xbench, I would get a different result. I also ran Xbench on my current MacBook Pro, and the results vary even back to back. When I was shopping for my MacBook Pro, I ran Xbench on (4) different MacBooks at Compusa. The slowest/cheapest MacBook won according to Xbench. I never could figure out why...

monkeyboy

@HeartBurnKid: i know it wouldn't "brick" probably should have used crashed...but either way you can do that with the iphone too

tellahoohooo

@tellahoohooo: I don't know about that; there have been Hackintosh builds since shortly after the first betas of Intel-based OSX came out. Apple hasn't bricked any of them yet. And the nice thing about computers (as opposed to cell phones) is that, if they do "brick" it, you can always reinstall the software.

HeartBurnKid

looks like you still made your comparison graphs on a PC...office 07?

Either way...I'd like to see if the hacktosh bricks like the iphone after a system update

tellahoohooo

From what I read on other forums, this is not Illegal, assuming you have purchased a licensed copy of OSX from Apple. Rather it violates the Apple EULA which only means that Apple are not required to support you with this or future releases.
If you were selling these things maybe that would be a different story.

justswitched

@chris-mcc: There are a million places that do. Froogle is your friend.

HeartBurnKid

@OmegaRed59: Ok then. Just so you know, from my experience newegg doesn't ship to Canada so it probably doesn't ship over seas either.

chris-mcc

I've used Hackintoshes before. In fact, it was the bridge between me and my Macbook that I'm using to write this.

Hackintoshes are very very very good at what they do. They use PC hardware to run the Mac OS(Whichever you prefer, Tiger or Leopard) in a way as to make it run well. It accomplishes this. I had issues at first when installing, but communities help that in terms of fixes, helpin you, etc.

As of right now, I still use my OSx86 and my MB with Tiger regularly. They do NOT compare in hardware, yet my weaker PC still runs pretty well in comparison. It has 1 core vs 2 and 1GB Ram vs 2 in my MB. Aside from those hardware deficiencies, the PC is still very very very good.

And to quote the great Forrest Gump, "And that's all I got to say about that."

XphillyX

Your all missing another point. Beyond 2 Gigs, there is little to no difference in performance on benchmarks that are heavily CPU intensive. The differences are very minor. When it comes to memory sub-system, and only on certain benchmarks, more memory makes a difference. Otherwise, 2 Gigs is the sweet spot for most PC's (and macs).

And as others have said, its the fact that you can get an 800 dollar beige box (or actually get a box, like a Lian Li, which looks substantially better than the Mac Pro case, and has room for 12 hard drives and a crap load of other things) can compete with a 2500 dollar Mac. But go ahead, disregard that fact. Maybe you can finally understand that there is nothing special about mac hardware anymore, except that it can handle OSX natively and has a nice exterior.

Nothing about it being an Apple makes it faster. Nothing.

diabolusunknown

I think you guys are missing the big picture here. He built that hackintosh with about $850 and a Mac Pro on the site right now says it starts at about some $2000. See my point? He is showing you that with a hackintosh of $850 you can get almost the same result as $2000. I personally think its brilliant apart from being illegal unless you have the OS X DVD but even then it still has some legal issues UNLESS you were an apple developer in which case its legal. I have tried this but failed because i dont think my motherboard is supported. I have a $1000 homebuilt computer consisting of 2GB RAM, 2.3 GHz Core 2 Duo, and 8800GTS which leads me to believe that this OS would run lovely with my computer. Just need to wait for the right edition to come my way and then i think i would give this a try. I have been following this project before Lifehacker put this article up as i had seen it before but i'm still waiting for the right edition to come out. Uphuck (OS X patched DVD maker) is coming out with a new Leopard DVD thats said to be very good. I will give that a try probably. Wow, this is a fairly long article.

guaterickie

@bradpdx: You say the aesthetics thing like PCs are still stuck in the beige box era. There are plenty of cases available that are both stylish and functional. Unfortunately, there are also plenty of garish, ridiculous pieces of crap, but hey, you can get pretty much any aesthetic you want in the modern PC case.

And, as far as support goes, I suppose that one year of AppleCare is worth trading $600-$1000 for if you really don't know what you're doing, and I wouldn't recommend a home-built computer (especially a Hackintosh) for a complete newbie. But frankly, I build my own computers because I'm skilled to the point that the only time I ever call support is for warranty replacements. I have the option of foregoing support on the PC side of things; why not with a Mac?

HeartBurnKid

What I would like to see is a comparison of a couple identically spec'd PC's and Macs, get an iMac with an identical mini PC, a Macbook pro with an identical laptop and a Mac Pro with an identical PC tower, let price be irrelevant and see what actually performs better. Will the difference in some components (type of ram, type of video card, motherboard) speed up or slow down osx? That is what I would be interested in. Sure you can build a PC for much cheaper than a Mac, you can build a gaming rig for much cheaper than buying an alienware or voodoo, you dont get the aesthetic value and the support though. which is a primary reason why Mac does not release their OS to PC users, since they know all the hardware and it is a limited sample, they are able to offer very good and effective support and also able to optimize software for their machines better. Think of the Mac like a console, when a game or peice of software comes out for a mac it just works, like on a console, because the developers can design for set hardware and know the outcome, you rarely get hardware conflicts or issues like you do with a custom built PC. Of course the downside to this is the lack of choice and slight price premium.
I would definitely like to see a hobbyest version of OSX that it unsupported or say a reverse bootcamp that would let you boot osx on a regular pc and offer no support for the osx side, like how Apple will not support you on the windows side of bootcamp.

Bailen

Well you could have taken the 2GB of ram out of the hackintosh then it would be much more of a even comparison to the macbook pro.

urbanride

Do Apple fanbois on here have a patent on sillyness?! How about first reading an article CAREFULLY before you flap your virtual gums!

This was not about "fighting fair" - the comparison was one to see how far a hackintosh that costs much less than the Mac Pro and MBP (see first article) can get you via benchmark comparison -- so the 4 GB Ram you're all bleating about comes at a cheaper price in the total hackintosh package than RAM in both the Mac Pro nad mBP.

Seems like it makes way more sense to buy a hackintosh that benchmarks middle of the road for OS X performance at a considerably lower price!

mmfy

@pantsonfireliarliar:
The only problem is the form-factor of an iMac doesn't allow you to replace the monitor. So I would classify that as a huge downside.
Eventually I will replace my 19 inch monitor with a bigger widescreen. I won't have to buy a new computer when I do that.

Oh, and I love your screen name lol

OmegaRed59

@chris-mcc: I was answering someone who never specified where they are. I was just showing one quick example. How am I supposed to know where "where I live" is? I was actually trying to help him/her find a good deal. You don't have to be so negative.

OmegaRed59

To answer the whole "bricking" thing, as far as I know, or at least for the version I ran, you couldn't update via Apple anyways, so I don't see how they could possibly "brick" you. People would reverse engineer and redo the update for the Hackintoshes.

XphillyX

@pantsonfireliarliar: And I'd like to see an iMac owner upgrade his monitor and only his monitor.

There are trade-offs to everything, including the iMac's form-factor. I've never been a fan of all-in-one designs for this very reason.

HeartBurnKid

@ColinC: I believe it would be this one.

@tellahoohooo: Actually, it's Office for Mac 08.

Adam Pash, LH Senior Editor

@OmegaRed59:
Why would someone want to replace anything on his computer? If I wanted that, I would buy one of those shitty plastic Windows Dell things...

trat

If I wasn't laughing at these comments, I'd have to cry. But I'm laughing.

The Hackintosh thing is certainly interesting; I'm not sure I'd want to rely on it as my main machine, but the idea of building a farm of cheap headless boxes to run as Logic nodes or something is pretty compelling.

Adam - I'd be interested to hear about HackBooks, as well as desktop systems.

NineTailedFox

Uh oh, the Mac zealots are out in mass to defend their beautiful Mac princess from be smeared by this bastard "Hackintosh". Chill guys, they specifically noted that the comparison wasn't ideal nor all that balanced.

xxdesmus

You guys who are saying things like "He built that hackintosh with about $850 and a Mac Pro on the site right now says it starts at about some $2000. See my point? He is showing you that with a hackintosh of $850 you can get almost the same result as $2000" need to go and really look at what he built. (He isn't making this claim; it's the he-man-Apple-hater's club doing it.) And get a clue on Apple's prices. The exaggerations of Apple system prices here are laughable.

The computer he built is most comparable not to a $2000 Mac Pro (there isn't one; stock price starts at $2499), but to a $1499 iMac with a $100 upgrade on hard drive and a $135 upgrade on RAM. After adding all the stuff included on that iMac but not in the $850, the diff is a few hundred dollars cheaper on the Hackintosh side, and no single warranty, thrice the space, self-service, none of the aesthetics, self-construction, a hacked on OS with a DVD needed for booting. Unknown how they would perform comparatively, since equally speced systems weren't tested. But the most hardware-equivalent test would be: A $1499 iMac with 4GB RAM cf. the speced Hackintosh.

Where someone managed to get a $2700 iMac that's the closest to the Hackintosh, I'll never figure out. You'd have to ASK to pay extra to do that.

dstewart

@ironchef: Erm, you don't have to update.

Anyway, these benchmarks are totally meaningless. Unless you can run it on identical hardware (same CPU, same amount/speed of memory, same harddrive etc etc) - unless the hardware is identical, all these benchmarks prove is that a faster computer does stuff faster...

What I want to know is how reliable the "Hackintosh" is, how fairly-hardware-dependent stuff like Motion or any Quartz-based stuff is (Even Quicktime Playback). Does Perian work? NTFS-3G MacStumbler? TimeMachine?

dbr

Had to love this line too:

"As you can see, the only place where the Mac Pro blows my Hackintosh out of the water is when it comes time for the processor-intensive work..."

Translated: Disk bound tasks are still limited by the speed of the hard drive even on a Mac Pro!

If you're not doing processor intensive work why in the world would you buy a quad (or octo) xeon machine?

numbersix

Hey Adam, looks like they took your post to heart over at UneasySilence. They used your guide to load Leopard onto an eeePC. A $399 ultra portable macbook ftw!

[uneasysilence.com]

D-Bo

@onidavin: I'm totally with you on that! "you're paying for the support, the reliability, and the OS." I love the fact that I can click the software update button and 5 minutes later I'm running 10.5.1 and back to work :-p

I spend to much time fixing everyone else's home built computers to mess with my own.

ColinC

Also, if apple starts losing money from lack of poeple buying their "genuine" systems, you can probably bet to see the price of OS X rising

atomicrabbit

I have to TOTALLY agree with CunfusedAmused. These tests are absolutely USELESS and it was a waste of time to post and a waste of my time to read. The whole point of these tests was to see how a pc-built system can stand up to genuine mac system and they compared totally different spec'd computers. OF COURSE the quad-core mac pro will rape the dual core in processor intensive tasks. THAT'S FRICKEN OBVIOUS-- WE ALREADY KNOW THAT. Get similar (or preferrably identical) spec'd systems and do the benchmark, otherwise don't bother.

That being said, now that apple switched to intel based systems, they are going to lose a lot of money from ppl building so-called "hackintosh" computers with hardware that costs a fraction of the price. I priced out a decent spec'd mac pro the other day out of curiosity and it came out to almost $4000 -- the (generally) same hardware I can buy to build a PC for 1/3 of the price. Maybe the quality of the pc hardware is not to spec of the apple hardware, but I think i can live with that considering I'll be paying $1200 as opposed to $3700. And I'm pretty sure there are a lot of people out there that agree with me. Plus, in a year or two, we'll probably have 16 or 32-core processors or maybe even a totally different architecture altogether, so I see no point in spending a crapload of money just to find out it's outdated in 6-8 months. Everything is advancing so fast now.

atomicrabbit

I'm really shocked by the number of people who don't understand that the test was about "bang for the buck", not ensuring that each machine has the same amount of RAM.

Let's assume I have $100 to buy groceries. I spend it mostly on expensive ready-made items. I don't spend much time in the kitchen, but I run out of food on Thursday. By these peoples' rationale, I should complain that I have to be given more money to spend because the guy who bought cheaper stuff still has groceries left and I don't.

Adam took a budget which was a lot lower than the cost price of the other machines and designed one to compete with them on their own ground (an Apple OS). One of his design choices was a crap-load of cheap RAM. Apple clearly made different design decisions. Tough. Let 'em go hungry.

Get it now? The test is fair.

Counterglow

They are doing testing with various hardware because they all are around the same cost. It doesn't matter which computer is 'unfair' because guess what, you can get that hackintosh for the same price as that MBP or Mac Pro, or whatever. What they want to show right now is that the cost is similar, but how much you can get out of a Hackintosh with the same cost.

knvb1123

So many people completely missed the point of what Adam was trying to do. When you buy apple stock configurations you are stuck with the price. Here are some stock component price comparisions, based on newegg/Mac Store comparisons.

Apple vs. NewEgg
1GB RAM $300/$51
500GBHDD $260/$119
CPU Upgrade $1098/$190

Note: CPU Upgrade is the cost difference between switching from 2 2.66GHz 1.333FSB Core 2 Duos to 3.0GHz 1.333FSB Core 2 Duos.

So it is very clear how much "Premium" apple put onto off-the-shelf components.

Also another point of consideration is that stock macs aren't very overclockable. There are many sites (TomsHardware.com) for example that show the new Peryn chips being overclocked to 4GHz air-cooled. Intel have demonstrated Peryn overclocked to 5.9GHz.

This means that a cheap Hackintosh with similar CPU to mac-pro (assuming Peryn) can be very easily overclocked to provide %30 peformance boost.

justswitched

@justswitched:

RAM is user replaceable in Macs. There is no reason to buy it from Apple. Upgrading it yourself is easy, and does not void your warranty. The same is true for the hard drive in some models, including the MacPro and the MacBook.

I think you'll find that Apple does not use Core 2 Duos in the MacPro. They use Xeon processors.

UCF_Chris

@Adam Pash, LH Senior Editor: I ran Xbench on my MacBook (SuperDrive) 2.16 GHz Core 2 Duo 3 GiB RAM GMA 950 200 GB HDD.

The results:
CPU Test: 124.90
Thread Test: 164.14
Memory Test: 130.74
Quartz Graphics Test: 164.69
OpenGL Graphics Test: 267.74
User Interface Test: 270.81
Disk Test: 43.42

Overall Score: 120.12

That's 28% better performance than the MacBook Pro.
That's 12% off the performance than the Hackintosh.
That's 7% off the performance of the MacPro.

Of particular interest is the Open GL Graphics Test, in which the MacBook, with its Intel GMA 950 with shared system memory for graphics, trounced the MacPro, the MacBook Pro, and the Hackintosh, all of which have separate graphics cards with dedicated video RAM. The MacBook didn't do too badly in the Quartz Graphics Test, either.

So how much did the MacBook cost? With discounts, rebates and upgrades, it was right around $1200, and I still have the 120 GB SATA HDD and 2 x 512 MiB RAM. Now let's see how much it will cost to bring it up to spec with the MacBook.

Hackintosh (as built): $756
A copy of Leopard (we'll pretend that it can be legally installed) $109
iLife 08 $70
Bluetooth 2.0 + EDR $22
WebCam $75
802.11n adaptor $50
Monitor $150
Speakers $10
Microphone $5

Total: $1247

I shopped Newegg and Amazon for good prices. I had to estimate the price of the webcam, since I couldn't find an equivalent to the MacBook's. The cheaper ones were of lower spec, and the higher ones were of higher spec. The 802.11n wireless adapter interfaces to a USB 2.0 port, which will likely reduce its effective data rate. The Bluetooth 2.0 also connects via USB. And I have no idea how poor the quality of the ten dollar speakers are. The MacBook actually sounds pretty good for a laptop.

So, the Hackintosh costs almost $50 more than an equivalent MacBook (that's not even the current generation), and only performs 13% better. Of course, it isn't portable like the MacBook, and lacks aesthetic style. And the price does not include the time necessary to hack the Leopard DVD, or shop for and assemble the parts.

In conclusion, for less money than it takes to build a Hackintosh desktop on the cheap, it is possible to buy a Mac laptop that performs almost as well (maybe just as well if it is a current generation model). And that performance is with 1 GiB less RAM than the Hackintosh has!

So much for the overpriced Apple hardware myth.

UCF_Chris

Would you please tell us how to configure the BIOS?

Lone_Wolf

Strangely enough the mac store processor prices seem reasonable (except for that 2.66 to 3.0 Upgrade I mentioned) if not competitive with other online prices.

justswitched

However I do stand corrected somewhat, The price difference I quoted above is the difference between 2 2.0GHZ Woodcrests and 2 3.0GHz woodcrests. The actual difference between 2 2.66 and 2 3.0 processors on mac.com is $799, making the upgrade 3x more expensive at the mac store (rather than the 4x I showed before).

Funnily enough the cost at newegg to upgrade from 2.0 to 3.0 is about the identical as at the mac store.

justswitched

Go to the mac store. The dual core mac pro configurations use Woodcrest not Clovertown as you suggest (thst's the quad core processor type), and I quote..

"performance with two Dual-Core Intel Xeon "Woodcrest" processors available in speeds up to 3.0GHz"

From Newegg.com
5150 Woodcrest 2.6GHz $749.99
5160 Woodcrest 3.0GHz $882.99
A difference of $130/processor or $260 for a set of two vs. $1098 that apple charges.

justswitched

@justswitched: just to clarify, apple sells the FIRST 500gb hdd for $129, which is a decent price, but the second, third and fourth 500GB hdd cost $329/ea, which is hugely inflated. That is WAY overpriced compared not the hdd out there.

Not to mention the price of their ram: $700 for 4gb of ram is outrageous. You can buy 4gb of high performance ram for less than half that. Unfortunately, Apple was not stupid, there is no option to purchase a Mac Pro without a hdd or ram. You must get at least 1gb of ram and a 250gb hdd.

atomicrabbit

@atomicrabbit:

Ram and HDDs are both user replaceable parts, meaning Apple will not void your warranty for replacing them yourself. Apple customers are in no way forced to pay the high prices for extra RAM or hard drives. There is a reason Apple offers the MacPro with only 1 GiB of RAM, which is obviously not enough to exploit the power of four or eight Xeon cores. The same goes with the tiny 250 GB hard drive. They don't expect the customer to keep such specs, but they aren't going to sell a system that doesn't work out of the box, so they have to put some RAM and a hard drive in it.

Of course, those who don't do their homework will end up paying August 2006 market prices for RAM and hard drive space, but the people who don't research their purchases tend to overpay for lots of things from lots of manufacturers.

@badsandwich:

Standard configuration:
NVIDIA GeForce 7300 GT with 256MiB of GDDR2 SDRAM (single-link DVI / dual-link DVI)

Upgradeable to:
• ATI Radeon X1900 XT with 512MiB GDDR3 SDRAM (two dual-link DVI)
• NVIDIA Quadro FX 4500 with 512MiB GDDR3 SDRAM (Stereo 3D, two dual-link DVI)
• Up to four NVIDIA GeForce 7300 GT with 256MiB

UCF_Chris

After all the yelling and shouting, I still haven't seen a realistic comparison. Surely someone has a MacMini they could test and compare? Well, actually, I do. My MacMini is a stock 2.0GHz Core2Duo with 1GB RAM and the 120GB Hard Drive. It's price? Direct from Apple, $799. Of course, It's not as expandable as the Hackintosh that was put together, but it's a comparable price and processor. My XBench 1.3 test results are below:

Results 114.33
System Info
Xbench Version 1.3
System Version 10.5 (9A581)
Physical RAM 1024 MB
Model Macmini2,1
Drive Type Hitachi HTS541612J9SA00
CPU Test 115.58
GCD Loop 237.10 12.50 Mops/sec
Floating Point Basic 113.72 2.70 Gflop/sec
vecLib FFT 92.91 3.07 Gflop/sec
Floating Point Library 92.32 16.08 Mops/sec
Thread Test 162.78
Computation 174.36 3.53 Mops/sec, 4 threads
Lock Contention 152.64 6.57 Mlocks/sec, 4 threads
Memory Test 132.07
System 143.10
Allocate 238.50 875.87 Kalloc/sec
Fill 115.63 5622.02 MB/sec
Copy 123.11 2542.78 MB/sec
Stream 122.61
Copy 113.06 2335.24 MB/sec
Scale 116.71 2411.26 MB/sec
Add 132.51 2822.70 MB/sec
Triad 130.49 2791.51 MB/sec
Quartz Graphics Test 179.00
Line 142.69 9.50 Klines/sec [50% alpha]
Rectangle 187.18 55.88 Krects/sec [50% alpha]
Circle 153.04 12.47 Kcircles/sec [50% alpha]
Bezier 150.03 3.78 Kbeziers/sec [50% alpha]
Text 419.73 26.26 Kchars/sec
OpenGL Graphics Test 299.63
Spinning Squares 299.63 380.10 frames/sec
User Interface Test 261.49
Elements 261.49 1.20 Krefresh/sec
Disk Test 38.30
Sequential 69.22
Uncached Write 67.16 41.24 MB/sec [4K blocks]
Uncached Write 66.92 37.86 MB/sec [256K blocks]
Uncached Read 64.92 19.00 MB/sec [4K blocks]
Uncached Read 79.69 40.05 MB/sec [256K blocks]
Random 26.47
Uncached Write 9.47 1.00 MB/sec [4K blocks]
Uncached Write 52.49 16.80 MB/sec [256K blocks]
Uncached Read 64.46 0.46 MB/sec [4K blocks]
Uncached Read 91.55 16.99 MB/sec [256K blocks]

As you can see, it was running Leopard and had only 1GB RAM. I've since upgraded to 2GB, and for XBench, it really makes no difference. The Core2Duo has 4MB Cache, and it seems like most of the code runs in the cache.

Comparing to the hackintosh results, the Mini's score is about 16% lower overall.
The CPU test is about 10% lower, which is in keeping with the 160MHZ speed difference.
The thread test is about 6% lower.
The memory test is about 13% lower, which isn't bad considering the bus speed on the mini is 667MHZ and the bus on the hackintosh is 1GHz (+33%).
The graphics performance is odd to say the least. The Quartz test results for the Mini are 15% lower than the hackintosh, but the OpenGL performance is actually about 66% HIGHER on the Mini. The GMA 950 likes OpenGL, I guess. The UI Test is back the other way with the hackintosh in the lead by about 20%.
Finally, the Disk Test favors the hackintosh significantly, coming in about 32% higher. This is also not a surprise, as the Mini uses a 5400RPM 2.5in laptop drive vs the 7200RPM 3.5in drive in the hackintosh.

So, what did all that prove? Well, as the original article said, you can build a computer user commodity hardware that is (nominally) cheaper and (nominally) faster than an Apple-built computer, with better expandability, that will run OSX.

At the $800 price point, though, I don't see the hackintosh being a MUCH betther deal than the Mac Mini. With the Mini at least, you don't need to go through the hassle of finding and/or hacking a copy of OSX just to get it to work with your hardware. That said, building PCs can be, and is, a lot of fun. I've built my share running everything from DOS 3.1 (ah the good old days) to Vista Ultimate. It's just that, like an aging hot rodder with money, I prefer to let other do the work for me now. That's why I have a Mac Pro, 2 Minis, and and iMac. Flame on!

Mojoxer

I see a lot of comments from the Mac side all mentioning support. Support for what? I'm at a loss here cause the other thing I always here is how rock solid the Mac hardware is and how easy and intuitive the software is compare to a PC. why all the need for support.

I built my own PC 3 1/2 years ago...my first build...not one bit of hardware or OS problems. No need for support of any kind. Obviously the typical Mac user must be someone who really needs the support.

And really...do you seriously think that Asus is building those Macbook Pros any better than their own brand laptops or their PC mother boards? Extra heavy solder? Special hard drives. Super heavy duty interconnects? Right.

JohnnyL

Something that would be a fair fight would be how much something costed.

Dooga