Clarification: Steve Jobs and OS X on Intel

[cgmpowers]

Re: BTW

Originally posted by sjs
Did you ever ask yourself why the question of using other companies chip rests on finishing the OSX transition? Logically it seems to make no sense. Upon further thought, perhaps they have several more features they are working on as updates for Jaguar in the latter part of this year (incl dot-mac stuff). So the software engineers are too busy for anything else.

Then in the early part of next year, let Jaguar rest for a while and they could begin re-compiling OSX to run on Intel/AMD chips in Mac hardware.

Thoughts???????

Thanks, I wasn't sure if the chipsets/processors made it any easier for games to be translated into Mac OS X or not..

Christopher

 

[wdodd]

Originally posted by kenohki
SGI hasn't been too successful with the Intel platform, Sun dropped Intel support for Solaris since it was too much of a pain, and let's not even get started on the whole NeXT fiasco.

Linux, on the other hand, has been enormously successful because it runs on standard x86 hardware.

Apple would be successful with a switch to x86 or Itanium if they used custom-designed motherboards, chipsets and ASIC's with Intel or AMD processors. That way they can get the right economies on one of the most expensive components on the MB. They can also match the specs of PC's better in their advertising. It also leaves an option open for "native" Windows emulation.

I don't think Apple should create a retail box version of OS X that installs on standard PC hardware. That puts them square between MSFT and Linux and they will fail in that space.

There's also an opportunity to work with nVidia on their new x86 chipsets as well as the GPU's.

Looking forward to seeing MWSF04! 🙂

 

[pdeli]

Originally posted by TechKid
I don't think Apple will ever go x86. Simply because the pc world is moving away from it too with 64bit processors and OS's. The Itanium is not x86 and won't run on your retail version of XP or Win2K. It needs XP for 64bit. Someone brought up a great point in another thread. They said Apple can't afford to switch platforms again as they've just done with OS X. Software companies don't want to code their apps all over agian. However, the software companies are going to have to do that for the pc world because they'll have to make their apps 64bit compatible (by the year 2005 they say). The next version of Windows (codename Longhorn) will be a 64 bit only OS and won't be compatible with older hardware (due out in early 2005). This is where Apple could jump into the pc hardware market. When the companies are converting their apps to 64bit, they may be more apt to code apps for OS X on pc hardware.

Hello,

While I understand the argument in the first part of your post which is complemented by the quote of another person, my question is: Why should the software companies have to code their apps all over again knowing that, precisely, it's a processor for which a big majority of softwares have been written? Allow me to explain myself: If Apple should ever switch to x86 type of processor I deduce that the "recoding" should be minimal as all the applications (all the Windows applications) do already exist for that processor...

There should, however, be an adaptation to the FBSD-type system.

This said, I like the second part of your message where you talk about the opportunity, for Apple, to switch to a 64bit architecture MP because this would not hamper all the "speed none-importance" campaign made by the company.

I think overall the game is fair. Both Apple and Microsoft are making good and bad steps. Now if one takes the opportunity of jumping in the footsteps of the other while the latter is committing the aforementioned error, that's too bad for me/you/them (whomever they are). I have the impression that we're just observing another darwininan (not Apples' core OS) effect of life: The fittest (not necessarily the best) survives...

What do you think?

Best regards,
Pit 🙂

++++

 

[kenohki]

Originally posted by pdeli


Why should the software companies have to code their apps all over again knowing that, precisely, it's a processor for which a big majority of softwares have been written? Allow me to explain myself: If Apple should ever switch to x86 type of processor I deduce that the "recoding" should be minimal as all the applications (all the Windows applications) do already exist for that processor...

There should, however, be an adaptation to the FBSD-type system.

This said, I like the second part of your message where you talk about the opportunity, for Apple, to switch to a 64bit architecture MP because this would not hamper all the "speed none-importance" campaign made by the company.
++++

In regard to the first part of your message, just because Windows apps will run on an x86 processor doesn't mean they'll run on an x86 Macintosh. Windows and Mac OS X use vastly different APIs which would totally negate the ability to run Win32 software.

Secondly, the POWER and PowerPC specs are already 64-bit (POWER4 is a 64-bit chip). Moving to a 64-bit PowerPC implementation would make more sense than moving to Itanic as you would get binary compatibility with your 32-bit applications a la what Sun has with the UltraSPARC (being able to run SPARC32 binaries natively). Nevermind the fact that the prices for Itanium and Itanium 2 would be prohibitive, especially for a chip that hasn't proven itself much in the performance arena.

Then again, I don't see Macs needing 64-bit chips until Sybase and Oracle RAC are ported. 😀

 

[wdodd]

Originally posted by kenohki
Then again, I don't see Macs needing 64-bit chips until Sybase and Oracle RAC are ported. 😀

This is my question as well. I understand the benefit of 64-bit to very large memory, etc. but what exactly is the benefit (if any) to multimedia (or other target markets) on the Macintosh?

 

[gjohns01]

Re: BTW

Originally posted by sjs
Did you ever ask yourself why the question of using other companies chip rests on finishing the OSX transition? Logically it seems to make no sense. Upon further thought, perhaps they have several more features they are working on as updates for Jaguar in the latter part of this year (incl dot-mac stuff). So the software engineers are too busy for anything else.

Then in the early part of next year, let Jaguar rest for a while and they could begin re-compiling OSX to run on Intel/AMD chips in Mac hardware.

Thoughts???????

The main reason Apple must wait to finish the OS X transition is that not everyone is onboard with OS X right now. THe conversion of the current userbase to OS X is only 10%. One of the reasons the conversion to OS X is slowed is a lack of software. (Yeah I know 1000s have been converted already) The OS X is still in tweak/fix mode and the power user apps are just making their way to the market. While some are never coming to OS X. Why throw another monkey wrench in the works by moving to an X86 architecture (prematurely if ever)? They might as well close up shop.

 

[Catfish_Man]

Very little...

Originally posted by wdodd

This is my question as well. I understand the benefit of 64-bit to very large memory, etc. but what exactly is the benefit (if any) to multimedia (or other target markets) on the Macintosh?

...the reason the Itanium2/Itanium/MAJC/etc... are so fast is not because they are 64 bit. Being 64 bit is sort of a side effect of the architecture that makes them so fast (VLIW chips do things in 128 bit chunks [the whole chip works like Altivec], so it's a piece of cake to have them do two 64 bit instructions instead of 4 32 bit ones. Altivec already does that.). The reason why Power4 and the other non-VLIW 64 bit chips are so fast is because they're huge and expensive. Power4 has 128MBs of L3 cache, and something like 10GB/sec of memory bandwidth. 64 bit is in there because they need the added memory. A 64 bit chip in a mac would do very little to help performance (they might be able to rig it so it could do a pair of 32 bit SIMD instructions instead of a 64 bit one [like a mini-Altivec], but that's about all). What will help performance is RapidIO/HyperTransport and on-chip DDR controllers.

 

[pdeli]

Originally posted by kenohki


In regard to the first part of your message, just because Windows apps will run on an x86 processor doesn't mean they'll run on an x86 Macintosh. Windows and Mac OS X use vastly different APIs which would totally negate the ability to run Win32 software.

Secondly, the POWER and PowerPC specs are already 64-bit (POWER4 is a 64-bit chip). Moving to a 64-bit PowerPC implementation would make more sense than moving to Itanic as you would get binary compatibility with your 32-bit applications a la what Sun has with the UltraSPARC (being able to run SPARC32 binaries natively). Nevermind the fact that the prices for Itanium and Itanium 2 would be prohibitive, especially for a chip that hasn't proven itself much in the performance arena.

Then again, I don't see Macs needing 64-bit chips until Sybase and Oracle RAC are ported. 😀

Thank you for your response Kenohki.

I see. Thanks for the clarification regarding the difficulty of running the same code, made for the same chip, but on different plateform. However, I would see one reason why a switch would be made: the simple cessation of the existence of the Power4/PC chips by their creators. In the light of economical difficulties of Motorola (which might lead to the abandoning from this company of the PPC developpment scheme) and a probable recentralization of IBM to it's core business (as Apple did when S. Jobs came back onboard), I don't really see Apple making a buyout from either or both of these companies in order to perpetuate the life of the chip. Let's not forget that all this is governed by the law of economics. If economically, the developpment of the PPC chip is not viable for Motorola and/or IBM, Apple or not, they will abandon it. I belive that if I thought about this, Steve Jobs et al. did it too. I think that Reuters article quoting the CEO of Apple is quite revealing (read the last 3 paragraphes):

http://www.reuters.com/news_article.jhtml?type=technologynews&StoryID=1214469

Of course, Reuters and what the latter wants to make us understand might only be a misinterpretation of my part, but in any ways what I wanted to say about this is that a change of the chip on which Mac OS X is now running might be inevitable. In the case that this turns out to be true, which would be, today (it might of course change in the future: Crusoe or stuff), the most obvious move?

Best regards,
Pit 🙂

++++

 

[mcrain]

Re: Clarification: Steve Jobs and OS X on Intel

Originally posted by arn
Now, as you point out, once our transition to Mac OS 10 is complete, which I expect will be around the end of this year or sometime early next year and we get the top 20% of our installed base running 10, and I think the next 20 will come very rapidly after that.

Ok, the above quoted section seems to me to be the most interesting part of this whole story. Steve is talking to the media about OSX adoption and about the processors (roadmap too).

He says that in all this time, they've got 20%, but for some reason the next 20% will adopt very quickly after the end of this year or sometime early next year.

Why on earth would that next 20% come very rapidly after the end of this year or early next year?

That's what I want to know.

 

[TechKid]

Re: Re: Clarification: Steve Jobs and OS X on Intel

Originally posted by mcrain


Why on earth would that next 20% come very rapidly after the end of this year or early next year?

That's what I want to know.

Here's what I get out of that:

A) A lot of people had a 'wait and see' attitude about OS X. Now that they've had decent exposure they'll convert

B) Jaguar. It offers a lot of improvments... sort of like the second service pack for OS X, if you will.

C) The new iApps that have been introduced and the latest software that's been ported (i.e. Maya, Warcraft, Real Player)

D) Macworld. I think a lot of people wait for the bi-annual event to get remotivated about Apple

E) Economists are saying that the economy has been showing signs of rebounding. Maybe people will now go out and upgrade to OS X

F) New hardware. Everyone knows new hardware is due out soon. Once it comes people will buy.

G) Back to school upgrades

H) X-mas

I) One last reason why I think he said that.... he's beeing really optimistic and needs to deflect Microsoft's criticism about the slow adoption of OS X

I happen to agree with Mr. Jobs. OS X will pick up steam in the next coming months.

 

[mccoma]

64-bit then power

Originally posted by wdodd

This is my question as well. I understand the benefit of 64-bit to very large memory, etc. but what exactly is the benefit (if any) to multimedia (or other target markets) on the Macintosh?

The very large memory is the big thing. For the same reasons databases are helped greatly by 64-bit addressing, video applications will benefit. I think raw Digital Video is 220MBytes a minute. This adds up for a two hour program. Having a very large memory address space and the corresponding ability to build a machine with a large main memory (> 4 gigs) will help video people. Databases are just the most common example sited.

 

[brdvance]

Real Reason waiting for adaption of OS-X

The real reason that APPLE/Jobs needs to wait for the adoption of OS-X before they could even consider changing chips is because OS-X is what would make changing chips incredibly simple, and even almost invisible to the end-users.l
Nobody will have to re-write OS-X code if APPLE decides to switch chips. All they have to do is re-compile the code they have and release an 'update'. In fact, the way that I understand the way that applications are created ('*.app' directory with a bunch of files in it), the application could have BOTH BINARIES in the same application (different BINARY files in the '*.app' directory), so it would be invisible to the user (kind of like FAT binaries in the transition to PowerPC).
In fact, think about this... Once OS-X is fully adapted, it would be EASY for APPLE to use *ANY* hardware. They could work on SUN/SPARC machines, Itanium machines, or any new processor that comes up... All application developers would need to do to get it running on ALL those machines would be to re-compile the source-code they ALREADY HAVE.
This is what would make it MUCH EASIER for APPLE to become a SOFTWARE company, and get away from being a HARDWARE company...

Originally posted by kenohki


In regard to the first part of your message, just because Windows apps will run on an x86 processor doesn't mean they'll run on an x86 Macintosh. Windows and Mac OS X use vastly different APIs which would totally negate the ability to run Win32 software.

Secondly, the POWER and PowerPC specs are already 64-bit (POWER4 is a 64-bit chip). Moving to a 64-bit PowerPC implementation would make more sense than moving to Itanic as you would get binary compatibility with your 32-bit applications a la what Sun has with the UltraSPARC (being able to run SPARC32 binaries natively). Nevermind the fact that the prices for Itanium and Itanium 2 would be prohibitive, especially for a chip that hasn't proven itself much in the performance arena.

Then again, I don't see Macs needing 64-bit chips until Sybase and Oracle RAC are ported. 😀

 

[alex_ant]

Re: Real Reason waiting for adaption of OS-X

Originally posted by brdvance
This is what would make it MUCH EASIER for APPLE to become a SOFTWARE company, and get away from being a HARDWARE company...

Why would they want to do this (be a software company), though?

 

[mccoma]

Intel, Apple, and needs

First I hope Apple never becomes a software only company. It would be the death of them.

Having an Intel / AMD chip would not make the Macintosh a PClone. Sun had a machine (not very successful) that had an i386 and didn't run MS Windows. If I was Apple, I would make sure that Windows wouldn't work on these machines. If you need Windows you could by Virtual PC (which should run a whole lot faster).

Also, Apple could skip to the 64-bit versions. Developers are going to have to recompile anway. On a side note, developers are also going to have to watch for endian issues that might come up in code.

on the other hand......

My problem with Intel / AMD is the heat. I would imagine that LCD iMacs would need to use the mobile versions of these chips. This would put Apple at a disadvantage anyway.

If Apple is going to switch from the PowerPC I hope they pick a chip low in power consumption with the ability to have multiple processors.

 

[pdeli]

Re: Real Reason waiting for adaption of OS-X

Originally posted by brdvance
The real reason that APPLE/Jobs needs to wait for the adoption of OS-X before they could even consider changing chips is because OS-X is what would make changing chips incredibly simple, and even almost invisible to the end-users.l
Nobody will have to re-write OS-X code if APPLE decides to switch chips. All they have to do is re-compile the code they have and release an 'update'. In fact, the way that I understand the way that applications are created ('*.app' directory with a bunch of files in it), the application could have BOTH BINARIES in the same application (different BINARY files in the '*.app' directory), so it would be invisible to the user (kind of like FAT binaries in the transition to PowerPC).
In fact, think about this... Once OS-X is fully adapted, it would be EASY for APPLE to use *ANY* hardware. They could work on SUN/SPARC machines, Itanium machines, or any new processor that comes up... All application developers would need to do to get it running on ALL those machines would be to re-compile the source-code they ALREADY HAVE.
This is what would make it MUCH EASIER for APPLE to become a SOFTWARE company, and get away from being a HARDWARE company...

That's good to know. Unfortunately, the knowledge I have in programmation and computers doesn't go that far, but if you're right, we have less to fear bad fortune for the Mac OS. After all, I don't really care on which hardware it runs, as long as the look of the computer is good from the outside and the operating system continues on the path it has taken now.

Now let's hope that Apple will not make too many bad strategical moves. Mr. Jobs did a great work taking out Apple from the dark ages in the late ninties. Maybe did he now attain the limit to what he can do for the company. Maybe should he now pass the stick to someone else before he reaches his limit of incompetence (Peter's Principle)...

There is one thing I would like to know, though: What do you (whomever wrote that) mean by "adaptation of OS X"? Adaptation to what? I see a continuous evolution of it, but an adaptation?

Cheers all,
Pit 🙂

++++

 

[brdvance]

Re: Re: Real Reason waiting for adaption of OS-X

The reason APPLE __MAY__ want to transition to a software company is because there are (currently) higher profit margins for software than hardware.

This DOESN'T mean that APPLE __DOES__ want to become a SOFTWARE-ONLY company, but it gives them options to go that route if it makes sense. I think they will be a hardware AND software company for at least 5 years, while they build market share. They use the hardware and software to build on eachother. However, once they reach a certain saturation level in the market, it would make it much easier to allow other manufacturers to make macs (if a larger chunk of PROFIT comes from SOFTWARE sales than HARDWARE), which could also help increase market share in business (business like multiple vendors where possible).

In short, it gives them more options. Like Steve said - we like to have options...

Originally posted by alex_ant

Why would they want to do this (be a software company), though?

 

[Brent]

What better way to get a significant percentage of the user base to upgrade to OS X by charging $129? I know I can't think of anything more logical.

 

[brdvance]

Re: Re: Real Reason waiting for adaption of OS-X

By 'adaption' I meant that their customers are adapting (or USING) OS X. If I remember the keynote, Steve said that only about 15-20% of Apples customers are using OS-X, and only about 60% leave OS-X as their default operating system on their new machines (that come with BOTH 9 and X).

The reason it's critical to get more (>90%) users to OS-X in order have the option of seamlessly changing hardware is because CLASSIC wouldn't work on other hardware (it would be WAY too costly to try and get that to work also).

I agree with you - OS-X will continuously evolve. But the way that OS-X is structured makes it so that the evolution doesn't have to be hardware specific. To look at it another way, for any UNIX-BASED system, virtually all the hardware-specific stuff is taken care of by the compiler. All the protocols/objects/structures etc are identical accross different hardware-bases, making the code MUCH more portable from one UNIX-platform to another. I didn't word it well, but it's the same reason why you can take just about any LINUX-program and compile it and get it to work on any LINUX system (powerPC or intel)...

Originally posted by pdeli


That's good to know. Unfortunately, the knowledge I have in programmation and computers doesn't go that far, but if you're right, we have less to fear bad fortune for the Mac OS. After all, I don't really care on which hardware it runs, as long as the look of the computer is good from the outside and the operating system continues on the path it has taken now.

Now let's hope that Apple will not make too many bad strategical moves. Mr. Jobs did a great work taking out Apple from the dark ages in the late ninties. Maybe did he now attain the limit to what he can do for the company. Maybe should he now pass the stick to someone else before he reaches his limit of incompetence (Peter's Principle)...

There is one thing I would like to know, though: What do you (whomever wrote that) mean by "adaptation of OS X"? Adaptation to what? I see a continuous evolution of it, but an adaptation?

Cheers all,
Pit 🙂

++++

 

[pdeli]

Re: Re: Re: Real Reason waiting for adaption of OS-X

Originally posted by brdvance
By 'adaption' I meant that their customers are adapting (or USING) OS X.

Thanks for the explanation brdvance. Upon what you wrote, one can only rejoyce to the possibilities that such a move might create. Hardware independence... isn't that great?

Cheers,
Pit 🙂

++++

 

[KingRocky]

A WinTel user perspective. . .

I have been using Intel/AMD processors and MSWindows for over 15 years now, so I have a bit of experience to draw on.

From my point of view, OS X is Linux/Unix done right--you install it and it WORKS. Period. The old addage "you get what you pay for" is dead on when it comes to Linux.

One thing we PC users DO enjoy, however, is the fierce competition between Intel and AMD, as it gives us incredibly fast processors at very low prices. This competition is partly due to the ravenous appetites of hardcore PC gamers who demand the absolute fastest equipment they can get. . . . which brings me to my major beef with Apple:

Let's do a simple price comparison, shall we?

AMD Athlon XP 2200: $211 (source: pricewatch.com)
Encore/ST G4 1GHz Upgrade: $699 (source: sonnettech.com)

Does anyone notice a slight price disparity here? Even if we throw out the processor speed differences, upgrading to Apple's fastest processor costs me 3 TIMES AS MUCH as upgrading to AMD's fastest processor?? Okay, so the processor card DOES come with a 2MB DDR cache, but 3 TIMES AS MUCH? I can build myself an entire PC for that price.

This teensy bit of information is what keeps me from running out and buying myself a Mac. It becomes cheaper to buy a new Mac instead of upgrading the one you already have.

So, I'm ALL FOR Apple switching to a 64-bit X86 processor if it means that I'll be able to upgrade my Mac for $211 instead of $699. Yes, I know that the current generation of X86 processors is not 64-bit, but I can GUARANTEE you that they will NEVER cost $699. PC users won't pay that price, and Mac users shouldn't have to either.

 

[arn]

Re: Arn, what misrepresentation?

Originally posted by sjs
I have been wondering what was misrepresented between the Reuters article and the actual conference. I see no difference at all. In both places they quoted SJ as saying that after the OSX transition is done, about the end of this year, they will have options regarding the question of which chips are used. There is no contradiction.

Not to be a conspiracy theorist, but are you trying to divert attention from this issue? Did Apple send you a cease and desist?

🙂

The difference is the emphasis... as well as the ommision of his comment about Motorola and IBM's roadmap looks pretty decent.

It's subtle... but the Mac press certainly took this as an admission from Jobs that he's looking into a x86 transition... while it was more of an innocent comment if you actually hear the transcript.

arn

 

[Cappy]

As I was sitting here contemplating what things might be like if Apple began using AMD or Intel chips, I started thinking about how the one advantage that this would bring could hurt Apple as well. We're all aware how doing so would allow Apple to attain the same speeds as the Dells, Compaqs, and Gateways but the question I started thinking of was "Can they keep up?"

By this I mean look at how often the cpu's get refreshed. There would definitely be no more grand showing at Macworld for the latest never before seen cpu for laptops and desktops. They would then be in a footrace with Dell to see who is has systems out with the latest, greatest cpu. I'm sorry to say but Dell's going to beat them everytime. They're that efficient.

Also consider the fact that Dell and others can just roll cpu and motherboard inventory down into lower end models over time because they carry so many lines and models. Apple doesn't unless they change. Not impossible but unlikely as this would now confuse Mac users. So if they leave the number of models limited now they have to watch inventory that much closer since they're competing with the likes of Dell in that footrace I mentioned.

No matter how you stack it this will be the riskiest move ever for Apple if *they* switch. Essentially MS and Intel have forced Apple's hand to play on their turf. Things could get very interesting.

 

[alex_ant]

I agree, Cappy. I posted this on OSNews, and since this thread has the same topic, what the hell, why don't I spam it here as well:

Well, what if Apple does nothing but continue to bump the G4 in the coming months. Dual 1.2GHz + DDR in a month, dual 1.4GHz by Christmas perhaps, and then dual 1.6GHz by next spring.

The G5 has been elusive, but the fact is that the PPC 8540 (one non-Mac variant of the G5) is already released. What it lacks is AltiVec and other features necessary for use in Macs (it's an embedded chip). The G5+, which is expected to run at 2.4GHz, will, as a conservative estimate, according to Motorola's roadmap as of Dec. 2001, arrive sometime between Q3 of this year and Q2 of 2003.

We know the G5 in the Mac is not going to happen in August, but what if Apple sticks this CPU drought and this crappy economy out for another 9 months? The preliminary G5 SPEC_CPU2000 scores are very impressive - in excess of 1300 in both int and fp, and that's running at only 1.6GHz. Which puts it well ahead of the P4 at 3GHz. Add AltiVec II to this, which is a beefier double-precision version of current AltiVec.

Yeah, I know, roadmaps schmoadmaps. But the fact is that, besides the crappy G4 rollout, Motorola has been quite good about sticking to its roadmaps. Look at the 680x0 series, and the early years of the PPC. Sure times are different, but it's really not a question of Motorola not caring about the desktop market, it's a question of when this crap economy will pick up and when Motorola can get their **** back together. Isn't it? They are a semiconductor giant with lots of potential. As is IBM. It would be FOOLISH of Apple to hop the fence and abandon these two great allies just because the fruit on their tree is not as ripe as the fruit on the other side.

Have some faith, people. Yes, Macs are slow and getting slower, but Apple is hardly dying. Do you file for divorce from your spouse the moment he/she gains 30 pounds too? I can see Apple going to another architecture - not necessarily x86 - as a last-resort bail-out move, but just because the PPC is slower than the Pentium/Athlon at this moment in time and is forecast to be that way for the next year or less is no reason to undergo such a HUGE transition.

Can Apple hold out for the G5 without going out of business? I sure as hell believe they can. And I believe that when the G5 DOES arrive in the Mac, and it will, damnit, the Mac zealots will look back on this thread and laugh, while the PC zealots will get all flustered with their challenges of, "Well, but can you build a complete ____ system for $137 + shipping?!?" and so on. I'm sure you can take it from there.

Alex

 

[TechLarry]

In other words, he doesn't see the problem.

My goodness, we ARE truly in trouble 🙁

Mr. Jobs needs to understand, quickly, that his RDF is shrinking rapidly, and some of us are not getting sucked in by it any more.

x86 is kicking the PPC's Ass, and as Apple customers we'll never get relief until the Man himself 'gets it'.

TL

 

[alex_ant]

Re: A WinTel user perspective. . .

Originally posted by KingRocky
Let's do a simple price comparison, shall we?

AMD Athlon XP 2200: $211 (source: pricewatch.com)
Encore/ST G4 1GHz Upgrade: $699 (source: sonnettech.com)

Does anyone notice a slight price disparity here? Even if we throw out the processor speed differences, upgrading to Apple's fastest processor costs me 3 TIMES AS MUCH as upgrading to AMD's fastest processor?? Okay, so the processor card DOES come with a 2MB DDR cache, but 3 TIMES AS MUCH? I can build myself an entire PC for that price.

This teensy bit of information is what keeps me from running out and buying myself a Mac. It becomes cheaper to buy a new Mac instead of upgrading the one you already have.

So, I'm ALL FOR Apple switching to a 64-bit X86 processor if it means that I'll be able to upgrade my Mac for $211 instead of $699. Yes, I know that the current generation of X86 processors is not 64-bit, but I can GUARANTEE you that they will NEVER cost $699. PC users won't pay that price, and Mac users shouldn't have to either.

Just to explain why that processor upgrade costs $700: It's not coming directly from Motorola, it's coming from a third party, Sonnet or whatever, that buys them in small quantities and attaches them to proprietary Mac-compatible CPU cards. As I understand it, it's the whole processor card you're buying, not just the processor itself. Yes, it still sucks that the upgrade costs so much, but, that's the Mac world for ya.

I think the mentalities in regard to what you mentioned are a bit different in Mac Land than in PC Land. In PC Land, there is a continual, endless stream of upgrades. You buy the components you need, and as the individual components get outdated, you upgrade them, and over time, your machine morphs into something completely different than what it used to be, to the point where it no longer has any parts in common with what it used to have. Mac Land is a bit different; although the desktop Macs ARE upgradeable, they are not realistically as upgradeable as a PC, so instead of a continual series of small upgrades, a Mac user will buy the Mac, use it for X period of time (2-3 years or whatever), then sell it and buy a new one. The Mac user can do this because the Mac's resale value is MUCH higher than the PC's. If you work out what you're spending in total over the years, and you're not shopping for PC parts at the absolute cheapest places you can find them, I think the ultimate cost will eventually turn out similar for both platforms, depending on PC upgrade frequency & Mac replacement frequency of course.

So, with that said, I think the hardcore gamer crowd will always be better off on the PC. More games and cheaper hardware that can be continually upgraded so that Machine X will always be at the top of the speed pack. With a Mac, and spending a similar amount of money, the hardcore gamer will have to put up with the "dips" in speed in between system replacements.

Alex