Slashdot Mirror
Dvorak Thinks Apple Will Switch to Intel
SeanTobin was among several users who noted that Dvorak's latest column discusses the possibility of Apple going to Intel for future macs. Yeah, this rumor pops up pretty often, but I wonder how long before we'd get binary compatibility between other x86 unix OSs.
He's an idiot!
Takes something with a shred of truth (the people being at said conferences) and blowing it into something "newsworthy".
For one thing, Dvorak thinks Apple will use Itanium. Not exactly binary compatible with other x86 unices...
There's no mention of Apple's most likely upgrade path in the next 12-18 months, the IBM PPC 970. Uh... hello?
You can tell high-level languages are the standard when someone proposes to switch a whole architecture to the x86 platform.
Remember the times the x86 was pointed at because of its lack of registers ? Recently read an pentium to-the-metal optimization guide, and discovered you had to recode your optimizations backwards to port them from p3 to p4 ?
I can't possibly understand how a switch to intel processors can possibly benefit Apple...
Karma cannot be described by words alone.
Dvorak is suggesting that Apple will switch to Itanium, which according to the roadmaps is nowhere near being ready for the desktop. At present, Intel is jamming larger and larger caches into Itanic until it will float against other processors in the server space, giving it an otherworldly transistor count not ready for the desktop in THIS decade -- the fabrication is simply too complex (read: $$$$), the power requirements are through the roof, and the compiler technology for IA-64 is many years from maturity. The Merced core for Itanic is absolutely useless, and I won't even get into the questions about whether even future generations will be viable.
A better 64 bit choice, particularly for Apple, will be IBM's upcoming PPC 970, which doesn't require massive retooling.
I wonder how long before we'd get binary compatibility between other x86 unix OSs.
Using an Intel CPU doesn't mean they have to repeat all IBMs mistakes from the past. They have the oportunity to design a BIOS from scratch that doesn't have to be backwardcompatible. It can become a lot better from that. I hope, that if they really go Intel, at least they bitch realmode and go in protected mode as fast as possible. While an OS that runs on both platforms does not come for free, it shouldn't be a problem to reuse userspace code on very different hardwareplatforms, as long as the CPU is the same. Of course it requires a reasonable OS design. I bet it won't be a long time after such a Mac has been released before you can run Linux on it with all the binary executables you already have for x86. Even WMWare might work, which would be kind of interesting. I wonder how long time it will be before MS ships a Windows that runs natively on Mac. I also wonder how Apple feels about that possibility.
I however still wonder why anyone would design a new architecture with an obsolete CPU. A much better decission could be to use AMDs new soon to be released x86-64.
Do you care about the security of your wireless mouse?
This wouldn't be the first time that the Mac has changed processors. (680x0 -> PPC) It's unlikely that Apple would keep the crappy PC style architechture though. Take a look at the base 1 MB and the terrible interrupt controller cascade. Apple wouldn't want to inherit this, plus if they stay far enough outside the PC, they can maintain their individuality.
I can picture geeks buying x86Mac hardware to run Linux on as it should be more stable than current x86 hardware. I can also picture x86 virtualization software (VirtualPC) being useful. Apple no longer has to deal with the low clock speed stigma.
This sounds like it would be a good thing.
He makes a point that they could release a dual CPU machine with an Itanium and a PPC chip, but this would be slower than a single CPU model for most things (dual CPU where each CPU is a different architecture is tricky and leads to performance hits). Since all Apple's current top of the line models have 2 PPCs, the new machine would be slower than the old ones.
On the other hand, the PPC 970 is comming into production, a 64-bit PPC with 2GHz+ clock speeds. 64 is twice as big as 32, so marketing can claim it's as fast as a 4GHz Pentium 4 (actually it might be almost that fast, since the P4 is famous for high clock rates and low performance per clock). Being a PPC, this chip is also backwards comaptible. Oh, and it has 2 AltiVec units, so all that AltiVec code Apple has been pushing for the last couple of years should really sing. A 900MHz FSB reduces the old memory bottleneck present in current PPCs. I'm not sure how much the PPC970 will cost, but I doubt it will be much more than Itanium, and it's far more attractive from Apple's point of view. This Dvorak guy seems to have forgotten that the Apple IBM Motorola alliance had 3 members...
I am TheRaven on Soylent News
Apple sells Apple computers. An "Apple computer" combines Apple hardware and an Apple OS. Not only would hell have to freeze over, but it would need to be at absolute zero before Apple starts diminshing their brand presence by selling an OS X that runs on non-Apple hardware.
-- Slashdot: When Public Access TV Says "No"
Apple has clearly stated it believes in a laptop future. There is also no way they can drop current Mac application compatibility.
Current speculation in the Mac community is the use of the IBM PowerPC 970 processor which should debut soon at 1.8Ghz. IBM have clearly stated it will support AltiVec instructions - previously only implemented by Motorola with IBM having no plans to use this technology themselves. Couple this with the rumours that some Apple OEM partners claim to have seen PPC970 based motherboard designs...
And then we have Dvorak who goes out on his own to claim a switch to Intel Itanium with a PowerPC inside for backwards compatibility. Quite how the hardware and OS would cope with two totally different processors is quite beyond me but surely the important question is how this would fit with a laptop.
The Itanium processor is not available in laptop form. It's current form requring around 100W vs the PowerPC 7455 (G4 processor used in PowerBooks) mere 20W range and you'll see that just isn't happening. Put both in the same box as Dvorak suggest? The heat and power consumption alone would make it impossible.
[)amien
I have a book here called "Dvorak predicts" from 1994 which states inter alia 'Apple will die if it merges', 'Apple needs to make a run-time Mac', 'the real Unix operating system is an archaic command line', 'Unix has no advantage except it's easy to program', 'Unix is old fashioned in its design and OS/2 or Windows NT architectures are the wave of the future'.
Correct me if I am wrong, but Apple merged and did not die, there is no rt mac archiecture available (excepting some good hacks that no one would use for business critical processes), unix based servers dominate the internet and MS are scared stiff that an old fashioned unix-like os is going to fillet their business.
Mr Dvorak is as entitled as anyone else to make his predictions, but that doesn't mean he is any good at it.
This is the most ridiculous thing in the whole article. Obviously he has forgotten the 68k emulator after the PowerPC changeover, as well as the Classic environment on OS X, both of which have worked perfectly in my experience.
Furthermore, I think there is a higher proportion of old Apple machines still running than equivalent old PC's. I saw an SE/30 doing a fine job as a mail server not that long ago. How many people are still using 286/386 vintage stuff?
-ccm
Too much Law; not enough Order.
KDE is far from being enough clean to that step. Gnome on the other hand is not finished enough.
KDE already runs on OSX
Under X11 emulation, that's far from being native
apple's using a huge load of kde code already
You're probably reffering to Safari. KHTML is not a huge load of code. Just take a look at Apple Cocoa and Carbon code, differences will
KDE is the best desktop enviroment in terms of architecture, customisability, flexiblity, applications , ease of use, i18n (over 80 languages!).
I'm a linux user, so I should be bashing OSX and Aqua (when it would be fair), but KDE has a long way to go.
Like first, spring cleaning of Menu and KControl, lack of support for handycaped persons, as long with other incosistances
Gnome on the other hand has this things but lacks of applications and being finalized on some points of usability. Maybe 2.4 or 2.8, but 2.2 has a way to go too.
So replace X11 with Quartz and you have one blazing system!
Replacing X11 with Quartz? Problems arise with hardware support. Apple uses Nvidia and Ati only so making some functions hardware accelerated is no problem, but it would become much larger problem with introducing larger ammount of different vendors. (I think XFree5 plans address this problems in the way Quartz does, hardware accelerated with soft emulation just like MesaGL-DRI-GLX are doing now)
As for the best, everybody has it's best. But to stay on topic. Apple's OS is the only reason to buy Apple machine, making it Intel based it would just bring a benchmark confrontation with other OS that run on the same hardware. Making it KDE this topic would become much more viable than CPU-only benchmark topic
Signature Pro version 1.13.2-3 release 83.5 beta3try7 after-breakfast edition