OS X on x86?

FusionJunky asks: "There has been some talk surrounding Apple's OS X and its potential to be released for the x86 platform. Sites like OSXonIntel.com have been trying to get the message to Apple that we feel the consumers are ready to see OS X on x86 boxes. I'm wondering what the Slashdot community thinks this would do to Apple, would it adversely affect their hardware sales? Could Apple move away from selling G4s from Motorola and start producing Intel Macs. Do you think Apple should release an x86 version of their next gen OS?" We asked earlier whether you felt if Linux would be threatened by OS X, with the possibility of OS X working on x86 machines, has your answer changed?

If they do it at all, it has to be soon

If Apple plan on releasing OS X on Intel, they'll have to do it soon, or it won't be worth it. When Sun released Solaris for Intel, it was seen as too late -- they were already comitted to SPARC, and the market saw the Intel move as halfhearted. SPARC was always going to be the preferred platform, and Intel was always very much a second rate citizen in the Solaris world. The same situation exists with OS X. Unless an Intel version is released very soon, OS X on PPC will be too entrenched. If it reaches that point, Apple won't be able to convice developers to code for both platforms, and they'll stick to just PPC. Ultimately, the support of third party developers is what makes or breaks an OS.

I'm actually in two minds about whether or not Apple should release an Intel OS X. On one hand, more competition is always a good thing. On the other hand, if it's successful, it could hasten the demise of viable non-x86 alternatives.

Re:It might be interesting-

I'm by no means a programmer, but it would seem to me a lot of work to port existing apps over to OSX on x86

Under NeXTSTEP, most cross-processor porting consisted of the following grueling steps:

• Step 1: Open the app in Project Builder.
  
• Step 2: Check "Intel," "NeXT," "Sparc," and "HP," then hit "Compile."
  
• Step 3: There's no step 3.
  

<chuckle>there's no step three....

Re:Affect hardware sales?

What matters ultimately is how they run, not how well they fulfill YOUR idea of what a server architecture should be

"Try moving your sound card to a different PCI slot."

"Why won't my motherboard enable 2x AGP?"

"How can I keep my video card and my NIC from sharing IRQ 9 under Windows 2000?"

"My system hangs if I try to hibernate it."

"Ever since I upgraded my BIOS, all of my games run slowly."

All facts of life, day and day out, in the PC world. Check any hardware/gaming oriented message board and see. Sure, it's great if you are a tweaker, but that's not exactly Apple's core market, and it's the sort of thing that their current customers abhor.

OS X on Intel would be another one of those specilty OSes that you would pretty much have to spec out a whole machine for before buying (kinda like OpenStep 4.2!) Apple could solve that problem by selling nicely designed 'certified' boxes, but that wouldn't be open enough for the people who are campaigning for this (who want to run OSX on the box that's sitting on their desk).
--

Re: NEXTSTEP and porting

Under NeXTSTEP, most cross-processor porting consisted of the following grueling steps:

• Step 1: Open the app in Project Builder.
• Step 2: Check "Intel," "NeXT," "Sparc," and "HP," then hit "Compile."
• Step 3: There's no step 3.

Oh, and you forgot to mention that you end up with one binary which through clever sharing of resources (all interfaces, resources, etc are in the .app folder rather than in the executable) is actually smaller than for example, 4 binaries of the same app for different architectures under *NIX. It also means you could essentially fire up Interface Builder on the .nib and localize apps yourself/move buttons around and so on.

IMHO these are the most important points about OSX people on /. are missing, just look at the amount of posts here saying "it's too hard". In the case of OSX, it's not. Apple has quietly been keeping Darwin running on X86, and NEXTSTEP/OPENSTEP (OSX's ancestor) always had support for X86. So an X86 port of OSX would be a matter of updating NEXTSTEP/OPENSTEP drivers, writing some new ones for hardware that didn't exist at the time (note I'm only talking about drivers for those subsystems that Darwin does not include, i.e. Aqua) and recompiling. Presto, you have OSX/X86.

What I still don't understand is how Apple managed to turn NEXTSTEP (runs fine on my NeXTStation Turbo which is a 33Mhz 68040 with 32MB of RAM) into OSX (apparently only runs reasonably well on a XXXMhz PowerPC G4 with 128MB of RAM)? As far as I can see, they are essentially the same OS (OSX has a couple more APIs and more Eye Candy, but surely doesn't justify such a jump).

Things I don't need

Yeah, that's what I've been clamoring for. Yet another OS to run on my x86, preferably one without a killer app or widespread support, and even better if it's one that won't support my existing apps. I want something pretty. It doesn't need to do anything that Linux/Windows/Beos does. I've always thought that the world needs more OS's. I'd love to go back to the early 80's, when every time I sat down at a new desk, I had to learn a new set of commands just to get my work done. Woohoo! My heart beats with anticipation.

Seriously, folks, what is it about Mac OS X that would make you shell out money for it when Linux is free? Do you think it can compete with Windows where Linux can't? Do you think it's genuinely better? I don't get it.

Mac OS X won't penetrate large organizations, either. Network admins have their hands full with 95/98 desktops, 2000 desktops, 2000 servers, Linux servers, and Netware servers. Mac OS X will be looked at as just another unproven alternative with no real history, an answer looking for a question.

I also can't believe posters haven't mentioned the failed 3rd-party hardware problems generated when Apple let other manufacturers build PowerPC boxes. Remember that? Apple couldn't stand letting other people build boxes and run Apple software. Now we're hearing rumbles that not only will it run on other boxes, but it'll be boxes that Apple hasn't put their golden blessing on. That's ridiculous - Apple dropped that hot potato years ago, and they're not likely to pick it up again.

BeOS model

A quick skim over some of the higher rated comments shows a common thread -- a lot of people are concerned that moving to x86 would be to abandon the superior Apple hardware and to embrace a lot of crappy legacy stuff. Could Apple not do what BeOS does, by saying "yeah, you can run it on a regular x86 box, but you have to have at least x for a graphics card, at least y for audio, and at least z for the $whatever.

That way, you can support the platform without supporting all the cruft that has built up over the years. I can almost see Apple defending their hardware (and the higher prices for it) by saying something to the effect of "yeah, you can run OSX on PC hardware, but you need at least a PIII / 1.2 gHz or equivalent processor, 196 mb of ram, a DVD burner...", etc -- just make the requirements so high that it would be a relatively expensive kit anyway, so it wouldn't hurt G3 based sales that badly, and it would avoid a lot of the legacy issue.

Re:Affect hardware sales?

I'm one of the Mac-using media people you talk about. You're right that price is king, but Macs are cheaper than x86 PC's. Don't just do a CompUSA comparison of the hardware sticker-price, although Apple's notebooks, at least, will win that easily, and the desktops would surprise you if you added the comparable features to a PC. You have to evaluate what the machine will cost you over its lifetime, and what it will give you over its lifetime.

Support. There are almost zero IT costs related to Macs. Apple has an article on their site right now about a 150-user consulting firm that is a Mac-only shop and has only two IT guys, and they take care of the phone system, too. In many places, the "business people" use Windows PC's and the "creatives" use Macs, and the IT people only work on the PC's and the "creatives" take care of their own Macs (Microsoft gives this as the reason why their Mac apps are installed with a drag and drop of the folder from the CD to the hard disk, and their Windows apps have a complicated installer). If IT does support the Macs as well, they typically have a 10:1 Windows:Mac ratio on their support calls.

Productivity. The interface is better and makes users more productive, especially in media apps, which were all originally Mac-only and it shows. You only have to learn one set of key shortcuts and they work everywhere. You don't have to know or care about pathnames. You can move or rename apps and they keep working. You don't have to use filename extensions. You can mount drives formatted with HFS, FAT, UDF, and other formats. Mac OS 9 is more stable than any DOS-Windows version, and more stable than NT 4. It probably won't quite go head to head with Windows 2000, but Mac OS X does, easily (been using it for over a year and no crashes, and these are alphas and betas). These things really add up.

Hardware. Apple's hardware is high-quality, and components are optimized for each other and work well together. You can add RAM to any Mac in under a minute, because they all have easy-access doors, even the notebooks (the keyboard is a door). Their displays are all color-calibrated, and integrate with ColorSync in the OS. Towers have built-in gigabit Ethernet, and all other Macs have 10/100 built-in (for years now). All Macs have two built-in 802.11 wireless antennaes, and an internal slot for a $99 optional wireless networking card. They all have FireWire built-in (except for the low-end iMac, I believe). They all come with optical mouses and quality keyboards. The machines come with a bunch of stuff that you'd have to add yourself, otherwise.

Saving a few hundred bucks off the initial sticker price just doesn't help you in the long run. This is why Mac users are willing to pay a little extra. We don't pay much more, though. The $3499 G4 tower has a $995 CD/DVD-RW combo drive in it, as well as iMovie, iDVD, iTunes, iTools, DVD player, CD-RW burning built into the OS, internal antennaes, gigabit Ethernet, FireWire, an audio amplifier for external speakers, power and USB for the display through the video card, optical mouse, room for 1.5 GB RAM, room for two removable drives and four hard drives, four empty PCI slots, easy open case with the mobo on the door, nVidia graphics. That's an awful lot of features and capabilities right out of the box.

There are so many misconceptions about Apple, because all the other PC manufacturers are in bed with each other and not with Apple. It's worth actually checking into this stuff ... it saves you money and makes your daily work easier. And they look cool and run Mac OS X. You can't beat that.

it may be iX86, but it'll be Apple's iX86

If apple were to run OS-X on iX86 based machines that they provided, I think you'd find that that there wouldn't be many similarities between apple's iX86 boxes and off the shelf windows iX86 boxes.

Look at a G4, and look at a standard iX86 box. They basically share the same memory bus, the same PCI bus, the same AGP port, the same USB ports; the Apple lacke a normal UART serial (except for the modem card that I don't think is an option any more) and has a FireWire (highspeed) serial instead. What else is different... well the processor, its cache, but most importantly the Firmware/ROM.

Apple, in some deal of wisdom, has used Sun's OpenBoot ROM (named openfirmware) to keep their system bootable, iX86s use the uber-patched BIOS. If Apple were to make intel based machines, it would make its own custom motherboard, and stick with the clearly supirior OpenBoot styled Firmware. It would be nice if you could get this off the shelf, but you can't since MS never bothered to prod developers into supporting it.

now consider that the only cost differences are the price of the Mot's or IBM's latest PPC (who-knows) vs the price of Intel's latest chip (the who-cares), using current history, you'll see that Apple is actually saving money by buying the cheeper chip. And it has better bang for buck no-doubt.

So, considering we know Apple won't use off the shelf part, whyever would Apple increase its production costs? Is the world such a sad place that it's going to sell a 1.1Ghz chip that does 0.5 instructions a cycle peak with more success than a 500Mhz chip that does 2 instructions per cycle average?

-Daniel

Meaningful performance comparisons

One interesting thing about this would be that it would remove the OS as a major consideration when doing performance comparisons. I know, I know, compilers and all that. But it would indeed bring the two hardware platforms a greater degree of parity on the OS front.