|
|
|
#11
2005-08-08, 09:27 PM
|
|||
|
|||
Apple Joins Intel, OS X and Windows Face Off
With all the buzz about “the deal with the devil” that Steve Jobs made with Intel, there has been no shortage of speculation and analysis of the future of Apple, the Macintosh line, and its other products. That’s to be expected. Nobody completely understands the reasons why the divorce between Apple and IBM actually happened, but there are some strong indicators. However, with OS X, things are a bit harder to predict. OS X never really competed directly with Windows or Linux, especially since you absolutely had to buy an Apple computer in order to run OS X. Sure, there’s PearPC or its alleged GPL-violating cousin CherryOS, but in theory, you had to have already purchased a Mac to get OS X or you had to have obtained it by, well, “other” means. Now that OS X is moving to x86, we have a whole new can of worms opening up, as well as some direct competition. The key to Windows, Linux, the BSDs, and to a certain extent Solaris/x86 is that they’re designed to run on any white box computer. Anyone could go to newegg.com or NCIX.com, price out a great new computer, and have a slew of operating systems to work with it. They’re also widely available for anybody to obtain and use right away. Windows has the distinct advantage of being produced by Microsoft, and having more than 90% of the operating system market in the x86 world. Windows also has an extensive driver database, hundreds of thousands of applications, a massive development community, is preloaded on almost every computer sold by a major OEM in the world, and a marketing department behind it that has its eyes trained on eating Linux for breakfast. Linux, on the other hand, has the advantage of being free, but doesn’t have all the features and comforts of Windows, let alone the drivers and software. Some see it as being not user friendly, and it hasn’t been successful in making a sizeable dent in Windows market share. The BSDs may even have it worse in this respect. Applications, drivers, user friendliness and cost are mentioned because those are the four biggest components in choosing an operating system for the average desktop. Joe O.S. Purchaser is not going to buy a copy of an operating system if it cannot run their software or if it cannot support their hardware. User friendliness has long been a selling point of Windows, especially since it comes preloaded on most x86 PCs to date. Finally, the price for performance compared to similar systems is a heavy part of any decision. If OS X is going to have any serious inroads in the x86 world, it’s going to have to compete on those four issues. With servers, stability is crucial, but for the most part, a server’s uptime is a result of its administrator. Then there’s security, but last I checked, Dell doesn’t sell computers preconfigured for secure applications, and again, Jane O.S. Buyer is not buying an operating system for security; she leaves that for the 14 year old down the street to handle. With respect to applications, Apple says that developers who have been using XCode is in the clear. Taking programs on OS X currently and porting them to the new hardware setup will be easy enough. Essentially, all that is needed to update a program is some time with a compiler. It’s doubtful that it would take long to get the GCC compiler tool chain ported as well. That way, smaller and open source developers can get back in the game quickly. However, developers that haven’t been using XCode are in for a bit of work, given that the endian issues alone can cause problems for poorly written software. Since Apple releases a lot of the popular programs that run on OS X itself, that body of software will likely be present when OS X/x86 starts shipping on new Macs. This means that some Mac based favorites, like Keynote will be ready. That leaves the commercial developers like Microsoft and Adobe, who likely are already working on porting software like the Office suite and Photoshop. One thing to note is that commercial developers may not produce two versions of the same software, even though there will still be buyers looking for software for their old PPC based Mac. Expect open source developers like the Mozilla Foundation to produce two versions, but depending on how well commercial developers think the PPC version will sell, they may skip producing it. According to Steve Jobs in his WWDC keynote it is basically a matter of small tweaks and a recompile if you are currently using XCode to produce your application. If that ends up being true, along with the “fat binaries” (one program that contains code for both architectures) working the way they are supposed to, there should be no problem with having programs on one CD/DVD that can live a “double life”. Needless to say, applications won’t stop OS X from gaining traction, and it’s possible that it may even gain some ground. As a real positive note for the x86 OS X, CodeWeavers (codeweavers.com) is planning to support the new Intel based Macs. CodeWeavers designs products that will soon allow Windows-only programs, such as games and office programs, to run on the new Macs without having any version of Windows installed. The previous PowerPC architecture prevented them from providing this functionality before, but using Intel architecture opens this possibility. Using their CrossOver products will boost people's ability to switch since they won't have to buy all new software applications for their new system. Also, it will boost OS X's ease of use considerably making it that much less for people to learn. Drivers Drivers are the next factor in determining how well an x86 Mac will do. Drivers are a subset of a larger problem, hardware compatibility and support. While Windows clearly wins in this category due to the seemingly infinite amount of drivers available for most computer hardware, there is the sticking point of people trying to plug in their peripherals into their new computers. Now, while Steve Jobs has made it clear that the x86 version of OS X will not run on any white box computer, that’s not going to stop someone from trying to plug in an MP3 player or scanner into a Mac. Also, because Apple controls what goes into a Mac, the driver pool is going to be very limited, unlike Windows XP which contains drivers for aging hardware, such as my SoundBlaster 16. The limited amount of hardware that actually works on a Mac as it is now is a boon for Apple. Going through computer forums, especially the DevHardware forums, will find posts which state that drivers are notorious for causing all kinds of problems for Windows. Even Windows XP’s Service Pack 2 furthered the problem for some users, causing Blue Screens of Death on badly written drivers when the NX protections were enabled on Athlon 64 processors. Linux users constantly complain that there aren’t drivers for their hardware, but the open source drivers tend to be good after maturing for a while. But when was the last time you heard a Mac user gripe about drivers? Apple has an advantage here. Sure, you may not be able to stick a new PCI Express video card in an x86 based Mac, but a device that supports OS X is essentially guaranteed to work. Not only that, its going to be easy to install and get working. The only problem, as mentioned before, is the bottom line; hardware manufacturers may or may not gamble on providing drivers for a Mac. I have no doubt that the makers of peripherals will add more Apple specific versions of their devices to match the white and aluminum landscape considering the market for iPod accessories. As far as internal devices that operate through a PCI/PCIe bus, that might be a different story. If both nVidia and ATI are able to produce “universal” drivers similar to their Windows efforts (and Linux for nVidia, ATI has been much less successful/determined in that department) then you should be able to make at least that part work if Apple included a Geforce 6200 and you wanted to move up to a 6800GT. Sound cards, wireless LAN, or RAID cards on the other hand aren’t going to be nearly as simple to upgrade, due to a specific driver being needed for each product as opposed to a “universal” one. For those you would be at the mercy of the manufacturer and Apple. User Friendliness User friendliness is not something that can be measured in monetary terms easily. It essentially boils down to a personal preference and the level of comfort the user has with the computer. On the other hand, if companies started doing massive switches to Macs, there will be retraining costs which include lost productivity that are hard to estimate. Considering that almost all Fortune 500 companies rely on Windows for the average desktop, what impetus is there for them to incur such a large cost, especially in a market where profit margins are everything? Windows opponents will argue that there will be money saved in terms of not having to deal with the spyware, adware, virus threats, and other maintenance; however, those savings come over a long period of time. It also assumes that OS X will not be the target of large amounts of spyware, even if OS X does take off. Unfortunately, Apple cannot win this one easily. Finally, there’s the flat monetary cost of OS X. This has been touched on a few times before, but not fully explained yet. OS X is not to be sold standalone. This is not surprising, especially since Jobs said that OS X won’t run on any x86 based computer. That means that OS X is a package deal: buy the hardware, and you get the shiny new operating system with it. This automatically makes it impossibly to precisely compare the cost of OS X to Windows; taken as an O.S. upgrade only, Windows will be incredibly cheaper since you don’t need to ditch your old PC hardware. Of course, the true comparison will be between the prices of fully-loaded PC desktops and brand new Macs. This may prevent OS X from taking off when people want to just upgrade their OS. OEM copies of Windows don’t cost too much if you know where to look. But as has been the case already, people upgrading their entire system can judge OS X and Windows as equals. The difference is that if a Mac-buyer doesn’t like OS X, there is a possibility that it may be easy to migrate that machine to Windows. If, however, Jobs did change his mind, he would have to price OS X somewhere in between Linux (free) and Windows (roughly $200). This also assumes that no one is going to come up with a hack that allows OS X to run on other x86 systems. Piracy and hacking may help OS X, at which point the cost drops to zero and all of the major players in the OS market compete on a level field. However, hacked copies could face a lack of drivers. If Apple itself was to release OS X for a wider install base, it would likely suffer some of the same driver conflicts and issues as windows has. Conclusion So, after all of this, will Apple start raking in large amounts of money because of the switch to OS X? Probably not. Mac users who were happy with a Mac before will only need to wait a bit before getting updated versions of their software. PC users will have to worry about drivers before switching. The look and feel of OS X, let alone its underlying BSD core, are not enough to cause a switch between big companies. Finally, to get a PC user to switch O.S., the user needs to buy a whole new computer, which is no change from Apple’s current setup. Expect the Mac Nation to grow, but slowly. One thing that I think Apple will take advantage of in their marketing for people looking to upgrade their whole system to gain the Apple Experience is that they CAN go back to Windows and keep a nice Apple desktop or laptop. Windows will be able to dual boot on these new machines, or you could dump OS X entirely should you decide it’s not for you. They probably won’t be planning to compete on price (though using Intel CPUs, it will be easier to compare price for performance), but I’d be much more interested on taking my chances with a 17” Powerbook. With the Powerbook, I can go back to Windows for work or play, or if I just can’t get along with OS X after being a longtime Windows user. Once it reaches that critical mass, all hell will break loose. That’s when Steve Jobs is likely to pull another rabbit out of his hat and cause another three years of speculation in the software industry. http://www.devhardware.com/c/a/Opini...dows-Face-Off/ 
|
