I know that MS plans something similar with Longhorn due in 2004 I believe. Right now Windows buffers windows horribly. I see why Apple has went to great extremes to develop Quartz. I hate watching the GUI just melt when the system becomes unresponsive. It looks very unprofessional.
These cheap Apple x86 boxes don't have to cost as much as a Dell. First, Dell uses Intel. AMD's chips are cheaper. The nForce motherboard is also pretty cheap and is made by nVidia, an Apple partner. Next, Apple can sell these boxes with dual or triple-boot capability but they don't have to ship them with Windows. They can just leave a partition open on the drive for somebody else to install windows or linux if they want to. That way Apple can avoid the MS licensing fees that everybody else has to pay. Thus there are two ways in which Apple's costs are lower and we already know that Apple will shrink their margins to near zero if it will seal an educational deal. Apple's other hardware and software products are already eating the cost of OS X development so the OS could basically be included for free on these cheap educational boxes.
Next, OS X on x86 would mean we lose classic and probably carbon apps, but Cocoa Apps should be able to be recompiled for x86 with the updated version of Project Builder that Apple would be releasing to go along with all this. The GCC compiler, which project builder uses, is already on x86, and the way cocoa is designed, it should just require things to be recompiled, (not counting assembly code which isn't used in most applications.)
In an education setting, the Macs wouldn't necessarily be hurt by the lesser amount of apps that would be available. The basic uses are already provided by Apple, with the exception of a web browser, but maybe OmniWeb 5 could be bundled with OS X by then. As for office software, there was talk of the open source office software coming to OS X and it is compatible with MS Office.
As an educational tool, people could learn e-mail, web browsing, word processing, spreadsheets, web site development, video editing, programming in a plethora of different languages, digital photography, and more just using the software that would already be included. Throw in third party software and you've got a lot of learning that can be done with a $500 box and a monitor they probably already have lying around.
Maybe an OS X port of Lindows could be thrown into the mix? A single .Mac account would have enough room to host websites for an entire classroom of students. There are lots of possibilities here.
I don't think it's too far fetched, and it would be nice to see apple selling some dirt cheap headless boxes to gain market share.
Maybe Apple could even market the "Switchboard", a variant of the nForce motherboard that PC users can buy and put into their existing cases to run Windows and Mac OS X in a dual-boot configuration. That way Apple would not require people to buy a Mac to run OS X, but the market share of OS X would increase and Apple would still be making hardware sales. Once all developers are writing Cocoa applications (which will be the only way to write apps on OS X for x86), making project builder create an x86 and PPC executable wouldn't be very difficult. Heck, all of the existing Cocoa tutorials on Apple's website would still work, because things would look exactly the same if you're running OS X, regardless of the processor in the machine.
A side effect is that people will realize that the grass isn't really greener. They will find that the perceived performance of the x86 world is just that... a perception. Mac OS 10.3 will probably be running faster on the fastest PPC hardware than on the fastest x86 hardware, and sadly, people will accuse Apple of crippling the x86 performance even though the reality will be that the x86 chips aren't as fast as people think they are and that OS X requires more processing power because it does more.
You have to keep in mind that when this time comes, OS 9 booting will no longer be supported and any OS 9 legacy support built into the motherboards will be gone. The result will be leaner meaner machines coupled with the hardware advances we'll be seeing on the PPC platform in the next 6 months and a more optimized GCC 3.3 compiler, and we'll start to see how much power the PPC really has.
The number of Carbon and Classic applications will continue to decline since they primarily exist for OS 9 support which doesn't exist on X86 anyway. Apple can eventually drop Carbon support from Project Builder and focus on developing the Cocoa APIs which will make things easier for developers.
[quote] Next, Apple can sell these boxes with dual or triple-boot capability but they don't have to ship them with Windows. They can just leave a partition open on the drive for somebody else to install windows or linux if they want to. That way Apple can avoid the MS licensing fees that everybody else has to pay. <hr></blockquote>
Yes Apple has neatly avoided the licensing but now YOU have to pay $180 for Windows. Remember the $89 WinXP you see are upgrades only. The full OS that can be installed cleanly is twice that price.
[quote] Next, OS X on x86 would mean we lose classic and probably carbon apps, but Cocoa Apps should be able to be recompiled for x86 with the updated version of Project Builder that Apple would be releasing to go along with all this. <hr></blockquote>
But the very lifeblood of Apple Developers are all on Classic or Carbon. Name one Cocoa App with the market of Quark or Photoshop. They don't exists.
[quote] I don't think it's too far fetched, and it would be nice to see apple selling some dirt cheap headless boxes to gain market share. <hr></blockquote>
I think you are being far fetched. Your whole arguement is based on low cost. However you're missing one important variable. You don't know the cost of Apple hardware. What you're doing is assuming that the actual BOG Bill of Goods on Apple products are higher than that of the typical PC White boxer and this is dangerous. Look at the Powermac motherboards and tell me what make you think they cost Apple anything significantly higher than the Taiwanese PC Mobo's. Price and Apple is determinded by what they feel is an acceptable profit margin. PC manufacturers are forced to "accept" much lower margins because they are in a much more competitive market.
I'm am still waiting for any benefits of X86. Everything i've heard is based on conjecture and that and .50 will buy you a cup of coffee.
What's this? An admission that Apple profit margins are in fact insane? In the words of Ali-G, "Respect."
I think what Apple has to realize more and more is that it is in the same price competition with Wintelon. Being a 'Mac' doesn't justify their prices to 97% of the market. Maybe most of those wouldn't buy a mac anyway, but a sizeable portion would if prices and (flexibilty of configurations) were better.
Buying edu (and I mean institutional, not individual) I don't get any great discount from Apple, about the same as the Apple-Store edu discount. They just cost too much for the overwhelming majority of schools. I can't recommend then, and, even if I were inclined to, I can't get them approved by higher-ups. That's a problem because regardless of what Apple and it's fans say, this is a situation where Apple is in direct competition with Windows. It's buyers don't care that it's a Mac and that's a wintel, they only care that they can buy the configuration they want at a cheap price with good support thrown in.
<strong>Yes Apple has neatly avoided the licensing but now YOU have to pay $180 for Windows. Remember the $89 WinXP you see are upgrades only. The full OS that can be installed cleanly is twice that price.</strong><hr></blockquote>
True, but we've already seen over the years that low cost of entry sells, despite how many hidden costs you are stuck with down the road. No matter how many TCO reports are published showing Apple as being cheaper, people still buy PCs "because they cost less." Not everybody would need Windows anyway, and many of the people that do could transfer their license from the computers the Apple boxes are replacing. Apple could also make "Windows XP" a BTO option using the license price for those who want it.
People think Microsoft would slaughter Apple for pulling such a stunt, but they'd still want the extra sales. All this would really do is force Windows XP and Mac OS X to compete on a level playing field. Any negative retaliations from Microsoft would not look good to the Antitrust lawyers.
[quote]<strong>But the very lifeblood of Apple Developers are all on Classic or Carbon. Name one Cocoa App with the market of Quark or Photoshop. They don't exist.</strong><hr></blockquote>
Maya. The next version of MS Office is supposed to be Cocoa as well. Carbon was used as a stop-gap. The changeover won't happen instantly, but with the lack of OS 9 booting on future Macs, do you really think Classic development is the way to go? And regardless of the advantages and disadvantages, Carbon apps are more work to write than Cocoa apps. Cocoa gives you a lot of things for free. There are lots of new developers coming onboard who are using Cocoa and writing apps that nobody has the patience to write in Carbon or for OS 9.
[quote]<strong>I think you are being far fetched.</strong><hr></blockquote>
I've always respected your opinions and posts on these boards and instead of acknowledging any good points I may have made you just say that. Does it really help you make your point?
[quote]<strong>Your whole arguement is based on low cost.</strong><hr></blockquote>
Educational and business purchases are usually based on low cost.
[quote]<strong>However you're missing one important variable. You don't know the cost of Apple hardware. What you're doing is assuming that the actual BOG Bill of Goods on Apple products are higher than that of the typical PC White boxer and this is dangerous. Look at the Powermac motherboards and tell me what make you think they cost Apple anything significantly higher than the Taiwanese PC Mobo's. Price and Apple is determinded by what they feel is an acceptable profit margin. PC manufacturers are forced to "accept" much lower margins because they are in a much more competitive market.</strong><hr></blockquote>
Quantity. Less sales means there must be a higher margin to offset development costs. Do you believe it is not more expensive for Apple to develop a motherboard and sell 500,000 machines with it than for nVidia to develop a motherboard and sell 5,000,000?
[quote]<strong>I am still waiting for any benefits of X86. Everything i've heard is based on conjecture and that and .50 will buy you a cup of coffee.</strong><hr></blockquote>
The benefit is market share. For consumers we'll still see PPC machines, but for these bulk educational purchases where cost is a primary factor, Apple needs something to compete with or they sell nothing to these buyers. The limited software choices caused by a lack of classic (and maybe Carbon) support is irrelevant if the computers will be sold to a specialized market with specialized needs that are met by the software that is available. Sure, Baldur's Gate II won't run on one of these boxes, but most teachers would consider that to be a good thing.
We're already paying for the development of OS X on x86, Apple may as well at least use it where they can.
Personally, I do not want to own an x86 box that runs OS X, but it could be useful to some markets and Apple would be wise to get OS X into as many markets as possible. More markets = more developers = more software = more feedback = more improvements.
You gotta be kidding, no developer will bother recode applications
Windows
Steve will not be seduced from the dark side.
What is likely happen?
1. Power4 lite version as G5
2. Motorola might release a 10 pipeline G5 with Rapid-IO
3. Open standard CPU design from Intel competitors namely IBM, Motorola and AMD to form AAIM alliance. AAIM will leverage each company's expertise to provide a suitable CPU line for Apple, which will also help Apple to gain market share
You gotta be kidding, no developer will bother recode applications.</strong><hr></blockquote>
You don't have to recode, just recompile. That involves clicking a checkbox and nothing more. Every developer would do that if it would make their software available to more users.
[quote]<strong>What is likely happen?
1. Power4 lite version as G5
2. Motorola might release a 10 pipeline G5 with Rapid-IO
3. Open standard CPU design from Intel competitors namely IBM, Motorola and AMD to form AAIM alliance. AAIM will leverage each company's expertise to provide a suitable CPU line for Apple, which will also help Apple to gain market share</strong><hr></blockquote>
That's all good, but none of that addresses the low end in the near future. Realistically, I think IBM and their partners are the future of PPC on the desktop, but those Power 4 derivatives aren't going to go into a sub $500 machine. Besides, it would be nice to see x86 in low-end machines... where it belongs.
You don't have to recode, just recompile. That involves clicking a checkbox and nothing more. Every developer would do that if it would make their software available to more users.
</strong><hr></blockquote>
At the risk of upsetting the Apple cart I can testify to the fact that going from PPC to x86 is a bit more than clicking a check box (endian issues). Please no comments about lousy programmers...
Also, if you are an Adobe or the like you would have to spend money on testing both versions. For the average Joe in a garage writing Cocoa apps however clicking the check box would constitute testing...
Making OS X available on x86 is a question of money. Can Apple make any without pissing off a bunch of their developers?
I'm a student at the University of Houston. Most of the computers here are DELLs. I haven't seen a single Mac in the Library or Student Center, just rows of black PCs. The University recommends students looking to purchase a computer for college buy a DELL with a special university discount. The University is overwhelmingly PC. My roommate didn't even know what a Mac was until he saw my iBook. That's not to say Apple doesn't have a presence here. The Campus Computer Store is one of the few remaining Apple strongholds. There are two PCs on display (Compaq), and everything else is Apple. I've seen a few Macs scattered around Campus, but I'm guessing this accounts for less than 5% of the total computers here on campus. DELL really is kicking Apple's butt in education. The only reason why there are 100 DELLs running as IE terminals in the library instead of 100 Apples is because the DELLs were probably cheaper. Price is king in the EDU market. DELL realizes this, which is why they are now the undisputed #1 in education.
But here?s what?s really starling:
Microsoft is beating Apple down as well. They want to increase their installed base of Windows and Office XP, so you know what they're doing? Next month every student at UH will be able to go to the Campus Computer Store and purchase a copy of WinXP or OfficeXP for $7. Students looking to upgrade their Mac to Jaguar will have to pay $70 for their OS. Want Office for your Mac? $200 please. Ouch! I can see why Apple has lost their foothold in Education.
Marklar may be necessary to regain lost ground, but Apple afford to give it away for $7? And it still doesn't solve the Office problem. Do you want to spend $7 or $200?
[quote] I've always respected your opinions and posts on these boards and instead of acknowledging any good points I may have made you just say that. Does it really help you make your point? <hr></blockquote>
Sorry I should have put a in my post because my intended tone was never meant to be combative. I love talking about the possiblilites of X86 OSX. I think it's a good possibility however I fear the condition that Apple would most likely be in to do so.
At the risk of upsetting the Apple cart I can testify to the fact that going from PPC to x86 is a bit more than clicking a check box (endian issues). Please no comments about lousy programmers...
Also, if you are an Adobe or the like you would have to spend money on testing both versions. For the average Joe in a garage writing Cocoa apps however clicking the check box would constitute testing...
Making OS X available on x86 is a question of money. Can Apple make any without pissing off a bunch of their developers?</strong><hr></blockquote>
Ok, there can be endian issues, depending on what sorts of things your code does with the numbers, and that is a potential problem. However, there is also a lot of code that can be compiled on several platforms with no tweaking at all. I've written programs that will compile on x86 linux and PPC OS X with no changes at all. Lots of code that would have endian issues has a check to see what processor it's on and uses the proper routines where endian issues would appear. Of course, this doesn't do anything for existing software that doesn't have such checks. (which is most non-unix/linux software)
However, this x86 box is not designed for mainstream use. Everything doesn't need to run on it. Unlike the "Education-only" eMac which many non-edu people were begging for, this Education-only machine would stay education-only. If a library needs a dozen web browsing terminals, it doesn't matter if it's x86 or PPC as long as the software runs, and if it's using OS X it not only means sales for Apple but it gets people using Macs, even if they don't realize it at first.
I don't believe it's feasible for consumer use, at least not for a long time, (3+ years - and in 3 years, x86 will be garbage anyway) but it certainly seems feasible for cheap specialized-use computers.
With regards to higher testing costs as somebody mentioned... developers are currently writing apps for OS 9 and OS X. That is two completely different operating systems. That is at least as much work as using the same set of APIs to develop software for the same OS on two different types of processors. Most of the hardware differences are invisible to the user and the developer except in situations where there are endian issues or assembly code. The foundation and GUI code will be the same, and looking at things like map editors for computer games like Heroes IV or Neverwinter Nights, the GUI code is where most of the time is spent.
Quake II was ported to OS X in less than a week after the source code was released. In that port, there were not only endian issues, but also operating system differences. It's obviously doable, but I do agree that expecting developers to write all their apps for x86 and PPC is asking for too much right now. That's why this cheap box won't be for everybody.
"Apple's Mac sales and marketshare have dropped substantially from what they were two years ago. At that time Apple was selling over 1 million Macs a fiscal quarter and since then, Mac sales have not gone over 850 thousand in any three months period of time.
Apple has kept sales dollar volume steady in the last two years by mainly adding more products. A new OS last year and iPod pumped up sales dollars last year. If it wasn't for adding the flat panel iMac, overall Mac sales would have fallen substantially in the first six months of this year.
Both iBook and PowerBook sales fell 11% year-to-year, in Apple's last reported fiscal quarter results. Yet, worldwide PC notebook sales rose 9% in the same period.
Power Mac sales were down 26% in Apple's last reported fiscal quarter, compared to the same three months from the previous year and they have been falling in seven of the last nine fiscal quarters.
The addition of Xserve and the eMac will probably keep overall Mac sales fairly steady for the July-September quarter and high gross margins from Jaguar sales are certain to boost Apple's profits substantially.
With the average PC desktop price at $800 now, Apple will probably have to cut prices considerably to gain marketshare. But, there has been very little opportunity for Apple to lower prices recently due to low sales volume. Several fiscal quarters in the last two years Apple lost money on product sales, but they showed a profit from interest off of savings. The sales spark by way of a new PowerPC design from Motorola or IBM in Macs next year would give Apple a prime opportunity to substantially lower prices for a boost in marketshare. Since at a given price, Apple's gross margins rise with sales volume, Apple would be able to maintain current gross margins while at the same time cutting prices. The risk in lowering prices is what will the sales be like when the initial demand for the new processors is satisfied? Will the sales volume be sufficient to obtain a profit after that initial sales rush has passed?
One way for Apple to lower the risk of insufficient prices to cover expenses is to use only one processor in Power Macs when a substantially faster processor is included next year. The main reason Apple uses two processors for Power Macs now is from trying to keep up with the performance of the Pentium 4. Motorola gave a list price of $295 for the 1 GHz G4 in January. So, eliminating a processor and its L3 cache, Apple could potentially reduce Power Mac prices by $300 and yet still maintain profit margins, even at the same sales volume."
I caught that from Macweek's reader 'feedback' to article on 'Apple doomed again' analysts predict...shock, horror etc.
I thought it may compliment Pismo's comments (which I agree with...)
My understanding was: OSX only, no dual boot; for the eduction market only; and designed to compete with cheap Wintel boxes.
Some of this makes sense. The issue of software compatibility is significant but that may be the rub. x86 boxes could be education only because they would be sold with a complete suite of (limited) software that could be tailored for individual school needs by Apple. Appleworks, iApps, maybe a few others which would minimize the need for third party software. Maybe one .Mac account per classroom...
Perhaps someone in the education field could chime in with some input on how this would fit with software needs in the primary and secondary education markets.
The stumper remains why does Apple need to go to x86 to accomplish this. Are their costs truly so prohibitive that the couldn't produce a $350-$400 box with a 1 GHz G3? Is their another advantage to using x86?
It's interesting that people assume that X86 would make for a lower production cost. I don't really see it. PPC is cheaper than X86. 295 for an 1Ghz was the price when they first appeared, but the prices on slower chips drop dramatically. At that time an 800 was 125. Apple put two in their machines because they can afford it. The PPC does not make Macs expensive, at least not directly. In fact, there's little on a Mac board to make it expensive as it is.
I think more than anything the PPC has hurt Apple because it has not provided a good spread of speeds with which to stratify their offerings. The price issue is really independent, it's Apple's own stubborn refusal to compete on price (even a little), or offer reasonable/affordable-yet-expandable machines. With Apple, you can have one or the other, but not both. Too bad really, whenever a machine even approaches a decent price performance and upgradability/expansion potential, it sells well -- look at the DP867.
As you can see, Apple's market share keeps shrinking, and it does so even in growing markets (like laptops). Should there be any surprise? The iBook now has a CPU better suited for a graphing calculator, and the price of the TiBook is simply ridiculous. We might acknowledge that yes, the actual performance is a little better than the spec suggests, but at any given price the products are behind, many of them considerably.
It's not all dark, you don't get speed, and you don't get price (well, you get over-priced, but you know what I mean.) You do generally get superior quality, reliability, integration, true plug and play ease of use, and better apps. You also get unique features sooner, ie. combo and superdrives, iDVD, etc...
A switch to X86 could concievably adress only two things. (1.) Hardware disparities. For better or worse, the hardware question would disappear: you'll have what they have, no more HW envy, in either direction. And, (2.), really just a subset of (1.), a nice range of chips for all your machines.
But CPU architecture doesn't speak to their main problem: price. Configuration speaks to that. In many ways, Apple has their configurations figured right, and the public to a large degree uses a poor formula for their buying decisions. Apple would rather sell you a slightly slower machine, with slightly slower tech (relative to what they can do) but include more features and tighter integration. The most popular X86 boxes tend to supply the fastest CPU/Mobo with only standard features. So you can get a VERY fast PC that won't burn discs, but it has a big display and comes with a free printer. You can add things later. You might never do it, but it'll make you feel better about buying it -- Expansion is the other MAJOR problem, although mainly psychological in most cases.
There's nothing but philosophy stopping Apple from making cheaper machines. Even a low end Apple (which still falls in the mid price range of computers in general) can make movies and manage photos and tunes out of the box. But how many budget buyers are out to make movies, or own DV cams or Digital cameras and MP3 players? Not many. These people want big screens, fast games, pirated music and porn, and fast web cruising at a low price. Apple refuses to give it to them: Macs comprise 3% of the market, PC's >95%.
You can be right about what makes an interesting computer, but most people aren't, and never will be, interesting. You can make all the art house films you want, but you'll go broke without a "get your rocks off here" summer action block-buster. Just doesn't happen. That's the reality. The current miseries of Apple price performance aside, Apple thinks it's smarter than the common slob. They're right, they are, and that's the problem. Regardless of what you know, you can only sell people what they want to buy. No amount of edumaction will change the luddites. The differences between Mac and PC are rooted in an area that has nothing at all to do with computers. It does not help that for a vast majority of common tasks PC's are actually very good, and that for many professional tasks, they have surpassed the mac. Nor does it help, that the overwhelming majority of people will NEVER do much more than common tasks on their computers, or will not do them untill they become common.
Fast, relatively stupid machines will win the market. Price counts most for most people, period. If Apple wants a bigger market thsy will also need fast, relatively stupid machines at a decent price. They will have to rely on expansion (theirs and 3rd party) and consumer exploration to make dumb macs more interesting.
None of this means that consumers want their machines to be strictly boring, but they don't need all those interesting things so much to justify Apple's premiums, they'll wait for that kind of functionality to reach common-place penetration and then take it when it becomes more or less the minimum spec.
This is the real issue facing Apple. It is why they lost market-share even when they were the undisputed performance leaders. Technology marches on relentlessly, in a generation or two al the others will usurp your unique features and the average consumer is content to wait for them.
Micheal Dell says, "I see what you want and here's a way for you to get it a the best price."
Steve Jobs says, "This is what you should want if you had any brains and just in case you do, well you know it's gonna cost, right?"
Retail Athlon XP 2200+ costs $155. I'm sure a bulk order would bring that down significantly, plus AMD would probably sweeten the deal to secure an order from such a high profile computer maker as Apple.
<strong>Why can't apple just take the eMac form factor, shrink it down to a flat 15inch CRT, a 500mhz 7410,128mb ram, CD drive, 20 gig HD, geforce2, $599.
I think this is a competitive price and is plausible config for the price</strong><hr></blockquote>
<strong>Why can't apple just take the eMac form factor, shrink it down to a flat 15inch CRT, a 500mhz 7410,128mb ram, CD drive, 20 gig HD, geforce2, $599.
I think this is a competitive price and is plausible config for the price</strong><hr></blockquote>
Comments
I know that MS plans something similar with Longhorn due in 2004 I believe. Right now Windows buffers windows horribly. I see why Apple has went to great extremes to develop Quartz. I hate watching the GUI just melt when the system becomes unresponsive. It looks very unprofessional.
These cheap Apple x86 boxes don't have to cost as much as a Dell. First, Dell uses Intel. AMD's chips are cheaper. The nForce motherboard is also pretty cheap and is made by nVidia, an Apple partner. Next, Apple can sell these boxes with dual or triple-boot capability but they don't have to ship them with Windows. They can just leave a partition open on the drive for somebody else to install windows or linux if they want to. That way Apple can avoid the MS licensing fees that everybody else has to pay. Thus there are two ways in which Apple's costs are lower and we already know that Apple will shrink their margins to near zero if it will seal an educational deal. Apple's other hardware and software products are already eating the cost of OS X development so the OS could basically be included for free on these cheap educational boxes.
Next, OS X on x86 would mean we lose classic and probably carbon apps, but Cocoa Apps should be able to be recompiled for x86 with the updated version of Project Builder that Apple would be releasing to go along with all this. The GCC compiler, which project builder uses, is already on x86, and the way cocoa is designed, it should just require things to be recompiled, (not counting assembly code which isn't used in most applications.)
In an education setting, the Macs wouldn't necessarily be hurt by the lesser amount of apps that would be available. The basic uses are already provided by Apple, with the exception of a web browser, but maybe OmniWeb 5 could be bundled with OS X by then. As for office software, there was talk of the open source office software coming to OS X and it is compatible with MS Office.
As an educational tool, people could learn e-mail, web browsing, word processing, spreadsheets, web site development, video editing, programming in a plethora of different languages, digital photography, and more just using the software that would already be included. Throw in third party software and you've got a lot of learning that can be done with a $500 box and a monitor they probably already have lying around.
Maybe an OS X port of Lindows could be thrown into the mix? A single .Mac account would have enough room to host websites for an entire classroom of students. There are lots of possibilities here.
I don't think it's too far fetched, and it would be nice to see apple selling some dirt cheap headless boxes to gain market share.
Maybe Apple could even market the "Switchboard", a variant of the nForce motherboard that PC users can buy and put into their existing cases to run Windows and Mac OS X in a dual-boot configuration. That way Apple would not require people to buy a Mac to run OS X, but the market share of OS X would increase and Apple would still be making hardware sales. Once all developers are writing Cocoa applications (which will be the only way to write apps on OS X for x86), making project builder create an x86 and PPC executable wouldn't be very difficult. Heck, all of the existing Cocoa tutorials on Apple's website would still work, because things would look exactly the same if you're running OS X, regardless of the processor in the machine.
A side effect is that people will realize that the grass isn't really greener. They will find that the perceived performance of the x86 world is just that... a perception. Mac OS 10.3 will probably be running faster on the fastest PPC hardware than on the fastest x86 hardware, and sadly, people will accuse Apple of crippling the x86 performance even though the reality will be that the x86 chips aren't as fast as people think they are and that OS X requires more processing power because it does more.
You have to keep in mind that when this time comes, OS 9 booting will no longer be supported and any OS 9 legacy support built into the motherboards will be gone. The result will be leaner meaner machines coupled with the hardware advances we'll be seeing on the PPC platform in the next 6 months and a more optimized GCC 3.3 compiler, and we'll start to see how much power the PPC really has.
The number of Carbon and Classic applications will continue to decline since they primarily exist for OS 9 support which doesn't exist on X86 anyway. Apple can eventually drop Carbon support from Project Builder and focus on developing the Cocoa APIs which will make things easier for developers.
Of course, this is all just speculation.
[ 09-17-2002: Message edited by: rogue27 ]</p>
Yes Apple has neatly avoided the licensing but now YOU have to pay $180 for Windows. Remember the $89 WinXP you see are upgrades only. The full OS that can be installed cleanly is twice that price.
[quote] Next, OS X on x86 would mean we lose classic and probably carbon apps, but Cocoa Apps should be able to be recompiled for x86 with the updated version of Project Builder that Apple would be releasing to go along with all this. <hr></blockquote>
But the very lifeblood of Apple Developers are all on Classic or Carbon. Name one Cocoa App with the market of Quark or Photoshop. They don't exists.
[quote] I don't think it's too far fetched, and it would be nice to see apple selling some dirt cheap headless boxes to gain market share. <hr></blockquote>
I think you are being far fetched. Your whole arguement is based on low cost. However you're missing one important variable. You don't know the cost of Apple hardware. What you're doing is assuming that the actual BOG Bill of Goods on Apple products are higher than that of the typical PC White boxer and this is dangerous. Look at the Powermac motherboards and tell me what make you think they cost Apple anything significantly higher than the Taiwanese PC Mobo's. Price and Apple is determinded by what they feel is an acceptable profit margin. PC manufacturers are forced to "accept" much lower margins because they are in a much more competitive market.
I'm am still waiting for any benefits of X86. Everything i've heard is based on conjecture and that and .50 will buy you a cup of coffee.
I think what Apple has to realize more and more is that it is in the same price competition with Wintelon. Being a 'Mac' doesn't justify their prices to 97% of the market. Maybe most of those wouldn't buy a mac anyway, but a sizeable portion would if prices and (flexibilty of configurations) were better.
Buying edu (and I mean institutional, not individual) I don't get any great discount from Apple, about the same as the Apple-Store edu discount. They just cost too much for the overwhelming majority of schools. I can't recommend then, and, even if I were inclined to, I can't get them approved by higher-ups. That's a problem because regardless of what Apple and it's fans say, this is a situation where Apple is in direct competition with Windows. It's buyers don't care that it's a Mac and that's a wintel, they only care that they can buy the configuration they want at a cheap price with good support thrown in.
<strong>Yes Apple has neatly avoided the licensing but now YOU have to pay $180 for Windows. Remember the $89 WinXP you see are upgrades only. The full OS that can be installed cleanly is twice that price.</strong><hr></blockquote>
True, but we've already seen over the years that low cost of entry sells, despite how many hidden costs you are stuck with down the road. No matter how many TCO reports are published showing Apple as being cheaper, people still buy PCs "because they cost less." Not everybody would need Windows anyway, and many of the people that do could transfer their license from the computers the Apple boxes are replacing. Apple could also make "Windows XP" a BTO option using the license price for those who want it.
People think Microsoft would slaughter Apple for pulling such a stunt, but they'd still want the extra sales. All this would really do is force Windows XP and Mac OS X to compete on a level playing field. Any negative retaliations from Microsoft would not look good to the Antitrust lawyers.
[quote]<strong>But the very lifeblood of Apple Developers are all on Classic or Carbon. Name one Cocoa App with the market of Quark or Photoshop. They don't exist.</strong><hr></blockquote>
Maya. The next version of MS Office is supposed to be Cocoa as well. Carbon was used as a stop-gap. The changeover won't happen instantly, but with the lack of OS 9 booting on future Macs, do you really think Classic development is the way to go? And regardless of the advantages and disadvantages, Carbon apps are more work to write than Cocoa apps. Cocoa gives you a lot of things for free. There are lots of new developers coming onboard who are using Cocoa and writing apps that nobody has the patience to write in Carbon or for OS 9.
[quote]<strong>I think you are being far fetched.</strong><hr></blockquote>
I've always respected your opinions and posts on these boards and instead of acknowledging any good points I may have made you just say that. Does it really help you make your point?
[quote]<strong>Your whole arguement is based on low cost.</strong><hr></blockquote>
Educational and business purchases are usually based on low cost.
[quote]<strong>However you're missing one important variable. You don't know the cost of Apple hardware. What you're doing is assuming that the actual BOG Bill of Goods on Apple products are higher than that of the typical PC White boxer and this is dangerous. Look at the Powermac motherboards and tell me what make you think they cost Apple anything significantly higher than the Taiwanese PC Mobo's. Price and Apple is determinded by what they feel is an acceptable profit margin. PC manufacturers are forced to "accept" much lower margins because they are in a much more competitive market.</strong><hr></blockquote>
Quantity. Less sales means there must be a higher margin to offset development costs. Do you believe it is not more expensive for Apple to develop a motherboard and sell 500,000 machines with it than for nVidia to develop a motherboard and sell 5,000,000?
[quote]<strong>I am still waiting for any benefits of X86. Everything i've heard is based on conjecture and that and .50 will buy you a cup of coffee.</strong><hr></blockquote>
The benefit is market share. For consumers we'll still see PPC machines, but for these bulk educational purchases where cost is a primary factor, Apple needs something to compete with or they sell nothing to these buyers. The limited software choices caused by a lack of classic (and maybe Carbon) support is irrelevant if the computers will be sold to a specialized market with specialized needs that are met by the software that is available. Sure, Baldur's Gate II won't run on one of these boxes, but most teachers would consider that to be a good thing.
We're already paying for the development of OS X on x86, Apple may as well at least use it where they can.
Personally, I do not want to own an x86 box that runs OS X, but it could be useful to some markets and Apple would be wise to get OS X into as many markets as possible. More markets = more developers = more software = more feedback = more improvements.
[ 09-17-2002: Message edited by: rogue27 ]</p>
You gotta be kidding, no developer will bother recode applications
Windows
Steve will not be seduced from the dark side.
What is likely happen?
1. Power4 lite version as G5
2. Motorola might release a 10 pipeline G5 with Rapid-IO
3. Open standard CPU design from Intel competitors namely IBM, Motorola and AMD to form AAIM alliance. AAIM will leverage each company's expertise to provide a suitable CPU line for Apple, which will also help Apple to gain market share
<strong>Swtitching to X86
You gotta be kidding, no developer will bother recode applications.</strong><hr></blockquote>
You don't have to recode, just recompile. That involves clicking a checkbox and nothing more. Every developer would do that if it would make their software available to more users.
[quote]<strong>What is likely happen?
1. Power4 lite version as G5
2. Motorola might release a 10 pipeline G5 with Rapid-IO
3. Open standard CPU design from Intel competitors namely IBM, Motorola and AMD to form AAIM alliance. AAIM will leverage each company's expertise to provide a suitable CPU line for Apple, which will also help Apple to gain market share</strong><hr></blockquote>
That's all good, but none of that addresses the low end in the near future. Realistically, I think IBM and their partners are the future of PPC on the desktop, but those Power 4 derivatives aren't going to go into a sub $500 machine. Besides, it would be nice to see x86 in low-end machines... where it belongs.
<strong>
You don't have to recode, just recompile. That involves clicking a checkbox and nothing more. Every developer would do that if it would make their software available to more users.
</strong><hr></blockquote>
At the risk of upsetting the Apple cart I can testify to the fact that going from PPC to x86 is a bit more than clicking a check box (endian issues). Please no comments about lousy programmers...
Also, if you are an Adobe or the like you would have to spend money on testing both versions. For the average Joe in a garage writing Cocoa apps however clicking the check box would constitute testing...
Making OS X available on x86 is a question of money. Can Apple make any without pissing off a bunch of their developers?
[ 09-17-2002: Message edited by: Pismo ]</p>
But here?s what?s really starling:
Microsoft is beating Apple down as well. They want to increase their installed base of Windows and Office XP, so you know what they're doing? Next month every student at UH will be able to go to the Campus Computer Store and purchase a copy of WinXP or OfficeXP for $7. Students looking to upgrade their Mac to Jaguar will have to pay $70 for their OS. Want Office for your Mac? $200 please. Ouch! I can see why Apple has lost their foothold in Education.
Marklar may be necessary to regain lost ground, but Apple afford to give it away for $7? And it still doesn't solve the Office problem. Do you want to spend $7 or $200?
[ 09-17-2002: Message edited by: Pismo ]</p>
[quote] I've always respected your opinions and posts on these boards and instead of acknowledging any good points I may have made you just say that. Does it really help you make your point? <hr></blockquote>
Sorry I should have put a
<strong>
At the risk of upsetting the Apple cart I can testify to the fact that going from PPC to x86 is a bit more than clicking a check box (endian issues). Please no comments about lousy programmers...
Also, if you are an Adobe or the like you would have to spend money on testing both versions. For the average Joe in a garage writing Cocoa apps however clicking the check box would constitute testing...
Making OS X available on x86 is a question of money. Can Apple make any without pissing off a bunch of their developers?</strong><hr></blockquote>
Ok, there can be endian issues, depending on what sorts of things your code does with the numbers, and that is a potential problem. However, there is also a lot of code that can be compiled on several platforms with no tweaking at all. I've written programs that will compile on x86 linux and PPC OS X with no changes at all. Lots of code that would have endian issues has a check to see what processor it's on and uses the proper routines where endian issues would appear. Of course, this doesn't do anything for existing software that doesn't have such checks. (which is most non-unix/linux software)
However, this x86 box is not designed for mainstream use. Everything doesn't need to run on it. Unlike the "Education-only" eMac which many non-edu people were begging for, this Education-only machine would stay education-only. If a library needs a dozen web browsing terminals, it doesn't matter if it's x86 or PPC as long as the software runs, and if it's using OS X it not only means sales for Apple but it gets people using Macs, even if they don't realize it at first.
I don't believe it's feasible for consumer use, at least not for a long time, (3+ years - and in 3 years, x86 will be garbage anyway) but it certainly seems feasible for cheap specialized-use computers.
With regards to higher testing costs as somebody mentioned... developers are currently writing apps for OS 9 and OS X. That is two completely different operating systems. That is at least as much work as using the same set of APIs to develop software for the same OS on two different types of processors. Most of the hardware differences are invisible to the user and the developer except in situations where there are endian issues or assembly code. The foundation and GUI code will be the same, and looking at things like map editors for computer games like Heroes IV or Neverwinter Nights, the GUI code is where most of the time is spent.
Quake II was ported to OS X in less than a week after the source code was released. In that port, there were not only endian issues, but also operating system differences. It's obviously doable, but I do agree that expecting developers to write all their apps for x86 and PPC is asking for too much right now. That's why this cheap box won't be for everybody.
Apple has kept sales dollar volume steady in the last two years by mainly adding more products. A new OS last year and iPod pumped up sales dollars last year. If it wasn't for adding the flat panel iMac, overall Mac sales would have fallen substantially in the first six months of this year.
Both iBook and PowerBook sales fell 11% year-to-year, in Apple's last reported fiscal quarter results. Yet, worldwide PC notebook sales rose 9% in the same period.
Power Mac sales were down 26% in Apple's last reported fiscal quarter, compared to the same three months from the previous year and they have been falling in seven of the last nine fiscal quarters.
The addition of Xserve and the eMac will probably keep overall Mac sales fairly steady for the July-September quarter and high gross margins from Jaguar sales are certain to boost Apple's profits substantially.
With the average PC desktop price at $800 now, Apple will probably have to cut prices considerably to gain marketshare. But, there has been very little opportunity for Apple to lower prices recently due to low sales volume. Several fiscal quarters in the last two years Apple lost money on product sales, but they showed a profit from interest off of savings. The sales spark by way of a new PowerPC design from Motorola or IBM in Macs next year would give Apple a prime opportunity to substantially lower prices for a boost in marketshare. Since at a given price, Apple's gross margins rise with sales volume, Apple would be able to maintain current gross margins while at the same time cutting prices. The risk in lowering prices is what will the sales be like when the initial demand for the new processors is satisfied? Will the sales volume be sufficient to obtain a profit after that initial sales rush has passed?
One way for Apple to lower the risk of insufficient prices to cover expenses is to use only one processor in Power Macs when a substantially faster processor is included next year. The main reason Apple uses two processors for Power Macs now is from trying to keep up with the performance of the Pentium 4. Motorola gave a list price of $295 for the 1 GHz G4 in January. So, eliminating a processor and its L3 cache, Apple could potentially reduce Power Mac prices by $300 and yet still maintain profit margins, even at the same sales volume."
I caught that from Macweek's reader 'feedback' to article on 'Apple doomed again' analysts predict...shock, horror etc.
I thought it may compliment Pismo's comments (which I agree with...)
Fuel for the fire etc.
Lemon Bon Bon
Some of this makes sense. The issue of software compatibility is significant but that may be the rub. x86 boxes could be education only because they would be sold with a complete suite of (limited) software that could be tailored for individual school needs by Apple. Appleworks, iApps, maybe a few others which would minimize the need for third party software. Maybe one .Mac account per classroom...
Perhaps someone in the education field could chime in with some input on how this would fit with software needs in the primary and secondary education markets.
The stumper remains why does Apple need to go to x86 to accomplish this. Are their costs truly so prohibitive that the couldn't produce a $350-$400 box with a 1 GHz G3? Is their another advantage to using x86?
I think more than anything the PPC has hurt Apple because it has not provided a good spread of speeds with which to stratify their offerings. The price issue is really independent, it's Apple's own stubborn refusal to compete on price (even a little), or offer reasonable/affordable-yet-expandable machines. With Apple, you can have one or the other, but not both. Too bad really, whenever a machine even approaches a decent price performance and upgradability/expansion potential, it sells well -- look at the DP867.
As you can see, Apple's market share keeps shrinking, and it does so even in growing markets (like laptops). Should there be any surprise? The iBook now has a CPU better suited for a graphing calculator, and the price of the TiBook is simply ridiculous. We might acknowledge that yes, the actual performance is a little better than the spec suggests, but at any given price the products are behind, many of them considerably.
It's not all dark, you don't get speed, and you don't get price (well, you get over-priced, but you know what I mean.) You do generally get superior quality, reliability, integration, true plug and play ease of use, and better apps. You also get unique features sooner, ie. combo and superdrives, iDVD, etc...
A switch to X86 could concievably adress only two things. (1.) Hardware disparities. For better or worse, the hardware question would disappear: you'll have what they have, no more HW envy, in either direction. And, (2.), really just a subset of (1.), a nice range of chips for all your machines.
But CPU architecture doesn't speak to their main problem: price. Configuration speaks to that. In many ways, Apple has their configurations figured right, and the public to a large degree uses a poor formula for their buying decisions. Apple would rather sell you a slightly slower machine, with slightly slower tech (relative to what they can do) but include more features and tighter integration. The most popular X86 boxes tend to supply the fastest CPU/Mobo with only standard features. So you can get a VERY fast PC that won't burn discs, but it has a big display and comes with a free printer. You can add things later. You might never do it, but it'll make you feel better about buying it -- Expansion is the other MAJOR problem, although mainly psychological in most cases.
There's nothing but philosophy stopping Apple from making cheaper machines. Even a low end Apple (which still falls in the mid price range of computers in general) can make movies and manage photos and tunes out of the box. But how many budget buyers are out to make movies, or own DV cams or Digital cameras and MP3 players? Not many. These people want big screens, fast games, pirated music and porn, and fast web cruising at a low price. Apple refuses to give it to them: Macs comprise 3% of the market, PC's >95%.
You can be right about what makes an interesting computer, but most people aren't, and never will be, interesting. You can make all the art house films you want, but you'll go broke without a "get your rocks off here" summer action block-buster. Just doesn't happen. That's the reality. The current miseries of Apple price performance aside, Apple thinks it's smarter than the common slob. They're right, they are, and that's the problem. Regardless of what you know, you can only sell people what they want to buy. No amount of edumaction will change the luddites. The differences between Mac and PC are rooted in an area that has nothing at all to do with computers. It does not help that for a vast majority of common tasks PC's are actually very good, and that for many professional tasks, they have surpassed the mac. Nor does it help, that the overwhelming majority of people will NEVER do much more than common tasks on their computers, or will not do them untill they become common.
Fast, relatively stupid machines will win the market. Price counts most for most people, period. If Apple wants a bigger market thsy will also need fast, relatively stupid machines at a decent price. They will have to rely on expansion (theirs and 3rd party) and consumer exploration to make dumb macs more interesting.
None of this means that consumers want their machines to be strictly boring, but they don't need all those interesting things so much to justify Apple's premiums, they'll wait for that kind of functionality to reach common-place penetration and then take it when it becomes more or less the minimum spec.
This is the real issue facing Apple. It is why they lost market-share even when they were the undisputed performance leaders. Technology marches on relentlessly, in a generation or two al the others will usurp your unique features and the average consumer is content to wait for them.
Micheal Dell says, "I see what you want and here's a way for you to get it a the best price."
Steve Jobs says, "This is what you should want if you had any brains and just in case you do, well you know it's gonna cost, right?"
X86 won't change any of this.
I think this is a competitive price and is plausible config for the price
<strong>Why can't apple just take the eMac form factor, shrink it down to a flat 15inch CRT, a 500mhz 7410,128mb ram, CD drive, 20 gig HD, geforce2, $599.
I think this is a competitive price and is plausible config for the price</strong><hr></blockquote>
That's too much.
<strong>Why can't apple just take the eMac form factor, shrink it down to a flat 15inch CRT, a 500mhz 7410,128mb ram, CD drive, 20 gig HD, geforce2, $599.
I think this is a competitive price and is plausible config for the price</strong><hr></blockquote>
needs to be 1GHz to be truely competitive.