- The vast majority of studies estimate the marketshare of the macintosh at somewhere around three to five percent. - One study estimates it (installed base) at sixteen percent.
There is a difference between installed base and marketshare.
If a PC user buy a new PC every other year and a Mac user buy a Mac every four years, you would see that the PC has 60% marketshare, but the installed base is still really only 50%.
Given that Mac users have claimed, for a while, about how long they last (a combination of higher price and higher satisfaction, I'm sure, in that they can't afford to buy a new Mac every other year, and that when they bought it in the first place it met their needs to the point that they didn't need to buy or upgrade a couple years later because it was slow or unsatisfying or virus infected), it wouldn't surprise me if Mac users replaced their Macs every 8 years while PC users have traditionally replaced their PCs every 3.
Probably the same thing that is stopping us from installing Windows on an Airport Base Station (which was powered by an AMD 486 incidentally).
Windows only installs on all modern x86 for two reasons: 1) Most hardware strives to be backward compatible to an IBM PC (BIOS, CGA/VGA/VESA, PCI/ISA, IRQs, etc) 2) Windows comes bundled with a lot of drivers
An Apple x86 won't be: 1) Backwards compatible with an IBM PC and it's legacy hardware 2) driver compatible with Windows
It might be hard finding x86 motherboards using OpenFirmware.
It's not exactly trivial right now to find PPC motherboards that can run OS X; they usually cost more than the pre-assembled Mac! I would expect a similar situation in x86 land.
Apple has never released a 1.5GHz part. I suspect, from the power usage of the 1.6GHz and the modern 1.8GHz that taking the newest 1.8GHz part and underclocking it to 1.5GHz will allow it to fit in a Mini form factor.
The problem of course is that will people buy such a low clocked part? For $499 they will. For $1299?
Nothing is impossible as processors become more powerful.
There's been plenty of research on dynamic on the fly compilation. IBM has DAISY, HP has Dynamo, and 10 years ago Dec had FX!32 on the Dec to allow Windows x86 NT binaries to run on Dec Alpha CPUs.
Nothing is stopping something similar (what with all the research Transmeta has done, and what the 'state of the art' is now) for a dynamic runtime compiler to sit as a driver to allow PPC code to run on a suitably powerful PC.
I don't believe it will happen, for business reasons, but I don't think there is anything technical stopping this kind of solution either.
Except it can't happen. The P4 is more or less dead ended. If there is any processor I want to see Apple adopt, it's AMD, due to the fact that those CPUs have integrated memory controllers. There is always the possibility of using the Pentium M, but there's really no advantages because it's a more expensive part.
You're telling me that IBM can't match those by 2006?
They've been making multi core CPUs for years now. They're making three core CPUs for Microsoft and 9 core CPUs for Sony. They've got a low power PowerPC part in GameCubes and in Blue Gene, several world class top 10 supercomputers Finally the G5 transistor count is about 1/3 that of the P4, and pricewise should be similarly cheaper.
We'll see what happens in 2006, but I doubt it will be x86 OS X
Um, what is stopping Apple from using Transitive's software to 'recompile' PPC -> PPC code. Since, for example, they have two architectures (G4 and G5), they can compile to a basic PPC, and then use this software to optimize for the native hardware, depending on how many Altivec units they have, FPUs, ALUs, etc.
There's no reason why they couldn't even use this to autovectorize precompiled unvectorized code.
The point being that you don't have to emulate EVERYTHING to run OS X on Intel hardware; you don't have to emulate the BSD core, the base libraries, or the kernel and it's associated extensions, if they already exist in the Darwin OS. They just need to emulate enough for the software on top of that to think it's running on a PPC. Wine allows Linux to run Windows applications by making those applications think they are accessing native Windows libraries.
Similarly this would fool OS X applications and parts of the OS that the underlying libraries are native when they are not.
It's certainly possible, but as you said, there is no incentive.
They haven't done it yet. Darwin has been out for, what, three years now?
All they have to do is port Darwin to a modern P4 or AMD architecture, and then create enough emulation to fool the OS X installer to 'install' on a P4 or AMD.
They could always create their own motherboard, chipset, and drivers. I mean, right now Darwin doesn't run on any x86 other than the 440BX chipset. so if Apple get's it's own chipset I don't see why OS X would run on any other. Right now every release of new hardware has a corresponding point release of the OS that includes firmware and drivers for the new machine.
I suppose C|Net could be right, there's nothing technically stopping a 'switch' to Intel, but I don't see what Intel has in 2006 that IBM can't match, or AMD, or whoever.
Wow, bookpcs are pretty expensive. The cheapest one is still $200 more expensive than a mini, and it doesn't come with a CD-RW, lacks an OS, software, and it uses integrated video!
I suspect that the introduction of the Mac mini will push down the price of systems like these; how else are they going to compete? On raw CPU speed?
No, I understand there are PC centric genres, such as sims, FPSes, MMOs, and other games that consoles can't/won't/don't touch. I was providing a counterpoint to the OP who thought consoles were dumb PCs with the implication that they were stupid because they weren't as flexible and general purpose as PCs.
I can't disagree. But the same logic says the PC is a similarly stupid gaming purchase. Spend $99 for a PS2, or $999 for a PC? Of course they have different games, and if you want to play UT or HL2, you have to buy a PC.
That's why I buy a Mac for my computer (10% more expensive, 2x more effective), and a console for my games (10% of the cost, 10x more effective). Each person has their own criteria, of course, for what is fun and what is cost effective.
Nothing has ever stopped a console from having a keyboard.
You can game on a $99 PS2, or a $1k PC. You can't reasonably game on a $499 PC without first spending $150 on an average video card, $200 on a decent one, or $300 on an outstanding one; then factor in the ram, the hard drive, etc, and you've got roughly 10x the cost; do you get 10x the pleasure?
I understand there are FPS and RTS gamers, but I think they are dwarfed by the number of casual gamers and non FPS or RTS gamers. For that reason 'investing' in a console is more cost effective for the majority, while gaming PCs will continue to go up in price as they become more niche (relatively, just as Macs have been relatively expensive as PC prices go down).
Slashdot Mirror
Yes. Your first statement should be:
- The vast majority of studies estimate the marketshare of the macintosh at somewhere around three to five percent.
- One study estimates it (installed base) at sixteen percent.
There is a difference between installed base and marketshare.
If a PC user buy a new PC every other year and a Mac user buy a Mac every four years, you would see that the PC has 60% marketshare, but the installed base is still really only 50%.
Given that Mac users have claimed, for a while, about how long they last (a combination of higher price and higher satisfaction, I'm sure, in that they can't afford to buy a new Mac every other year, and that when they bought it in the first place it met their needs to the point that they didn't need to buy or upgrade a couple years later because it was slow or unsatisfying or virus infected), it wouldn't surprise me if Mac users replaced their Macs every 8 years while PC users have traditionally replaced their PCs every 3.
Probably the same thing that is stopping us from installing Windows on an Airport Base Station (which was powered by an AMD 486 incidentally).
Windows only installs on all modern x86 for two reasons:
1) Most hardware strives to be backward compatible to an IBM PC (BIOS, CGA/VGA/VESA, PCI/ISA, IRQs, etc)
2) Windows comes bundled with a lot of drivers
An Apple x86 won't be:
1) Backwards compatible with an IBM PC and it's legacy hardware
2) driver compatible with Windows
Woah! News to me!
It might be hard finding x86 motherboards using OpenFirmware.
It's not exactly trivial right now to find PPC motherboards that can run OS X; they usually cost more than the pre-assembled Mac! I would expect a similar situation in x86 land.
What, you're saying someone is going to code up VMWare to look like an x86 non IBM PC compatible machine to run OS X?
I wonder if VMWare right now can run Darwin?
What part of '1.5GHz' don't you understand?
Apple has never released a 1.5GHz part. I suspect, from the power usage of the 1.6GHz and the modern 1.8GHz that taking the newest 1.8GHz part and underclocking it to 1.5GHz will allow it to fit in a Mini form factor.
The problem of course is that will people buy such a low clocked part? For $499 they will. For $1299?
Nothing is impossible as processors become more powerful.
There's been plenty of research on dynamic on the fly compilation. IBM has DAISY, HP has Dynamo, and 10 years ago Dec had FX!32 on the Dec to allow Windows x86 NT binaries to run on Dec Alpha CPUs.
Nothing is stopping something similar (what with all the research Transmeta has done, and what the 'state of the art' is now) for a dynamic runtime compiler to sit as a driver to allow PPC code to run on a suitably powerful PC.
I don't believe it will happen, for business reasons, but I don't think there is anything technical stopping this kind of solution either.
You don't think they could fit a 1.5GHz G5 by early 2006 in a Mac mini?
Essentially underclock the CPU and lower the voltage, underclock the FSB to 500MHz and lower the memory controller's voltage, and you have a SFF G5.
That, or leverage IBM's multicore technology and create a G5 with an integrated memory controller and an integrated northbridge.
Except it can't happen. The P4 is more or less dead ended. If there is any processor I want to see Apple adopt, it's AMD, due to the fact that those CPUs have integrated memory controllers. There is always the possibility of using the Pentium M, but there's really no advantages because it's a more expensive part.
You're telling me that IBM can't match those by 2006?
They've been making multi core CPUs for years now.
They're making three core CPUs for Microsoft and 9 core CPUs for Sony.
They've got a low power PowerPC part in GameCubes and in Blue Gene, several world class top 10 supercomputers
Finally the G5 transistor count is about 1/3 that of the P4, and pricewise should be similarly cheaper.
We'll see what happens in 2006, but I doubt it will be x86 OS X
So there's only one reason?
I suppose that means you've got no choice now but to buy from, hrm, IBM or Sun, right?
Um, what is stopping Apple from using Transitive's software to 'recompile' PPC -> PPC code. Since, for example, they have two architectures (G4 and G5), they can compile to a basic PPC, and then use this software to optimize for the native hardware, depending on how many Altivec units they have, FPUs, ALUs, etc.
There's no reason why they couldn't even use this to autovectorize precompiled unvectorized code.
I didn't ever say it was easy, did I?
The point being that you don't have to emulate EVERYTHING to run OS X on Intel hardware; you don't have to emulate the BSD core, the base libraries, or the kernel and it's associated extensions, if they already exist in the Darwin OS. They just need to emulate enough for the software on top of that to think it's running on a PPC. Wine allows Linux to run Windows applications by making those applications think they are accessing native Windows libraries.
Similarly this would fool OS X applications and parts of the OS that the underlying libraries are native when they are not.
It's certainly possible, but as you said, there is no incentive.
They haven't done it yet. Darwin has been out for, what, three years now?
All they have to do is port Darwin to a modern P4 or AMD architecture, and then create enough emulation to fool the OS X installer to 'install' on a P4 or AMD.
How does volume help Apple?
Right now the Pentium M is Intel's most expensive CPU, and there's really no alternative to it if Apple wants an Intel CPU in the Mac mini
They could always create their own motherboard, chipset, and drivers. I mean, right now Darwin doesn't run on any x86 other than the 440BX chipset. so if Apple get's it's own chipset I don't see why OS X would run on any other. Right now every release of new hardware has a corresponding point release of the OS that includes firmware and drivers for the new machine.
Wasn't April a couple months ago?
I suppose C|Net could be right, there's nothing technically stopping a 'switch' to Intel, but I don't see what Intel has in 2006 that IBM can't match, or AMD, or whoever.
You're right, the size isn't the selling point, it's the price.
The Cappuccino PC right now costs at least $200 more for the same basic features:
CPU
HD
RAM
OS
So you can either pay $674 for a system without a CD-RW, or $499 for the Mac mini AND get yourself a CD-RW, DVI, and a nice software bundle.
Wow, bookpcs are pretty expensive. The cheapest one is still $200 more expensive than a mini, and it doesn't come with a CD-RW, lacks an OS, software, and it uses integrated video!
I suspect that the introduction of the Mac mini will push down the price of systems like these; how else are they going to compete? On raw CPU speed?
The iMac fits your bill perfectly:
2004 technology
DVD writer
Upgradeable memory
Upgradeable hard drive
I bet your complaint then is that it's attached to a 17" or 20" LCD.
No, I understand there are PC centric genres, such as sims, FPSes, MMOs, and other games that consoles can't/won't/don't touch. I was providing a counterpoint to the OP who thought consoles were dumb PCs with the implication that they were stupid because they weren't as flexible and general purpose as PCs.
It's too bad your software investment ties you down to the PC. Macs are pretty nice.
If you want smaller than a shoebox, have you considered a Shuttle PC?
I can't disagree. But the same logic says the PC is a similarly stupid gaming purchase. Spend $99 for a PS2, or $999 for a PC? Of course they have different games, and if you want to play UT or HL2, you have to buy a PC.
That's why I buy a Mac for my computer (10% more expensive, 2x more effective), and a console for my games (10% of the cost, 10x more effective). Each person has their own criteria, of course, for what is fun and what is cost effective.
Nothing has ever stopped a console from having a keyboard.
You can game on a $99 PS2, or a $1k PC. You can't reasonably game on a $499 PC without first spending $150 on an average video card, $200 on a decent one, or $300 on an outstanding one; then factor in the ram, the hard drive, etc, and you've got roughly 10x the cost; do you get 10x the pleasure?
I understand there are FPS and RTS gamers, but I think they are dwarfed by the number of casual gamers and non FPS or RTS gamers. For that reason 'investing' in a console is more cost effective for the majority, while gaming PCs will continue to go up in price as they become more niche (relatively, just as Macs have been relatively expensive as PC prices go down).
Granted, but likewise we shouldn't underestimate what $500 can do to make a game platform as well.
A decent PC to play FFXI and Half Life 2, vs a PS2 plus 15 games (including FFXI).