Cringely: OS X on Intel

sti writes: "Cringely's column this week argues that Apple should port OS X to the Intel platform. He makes an interesting case for it. I would definitely favour this. I've always had this warm spot in my heart for Apple but rarely had the money to pay for their overpriced hardware."

Comments

Why Apple has, and why Apple won't

[gwernol]

Apple has already ported Mac OS X to Intel. And I don't just mean the Darwin open source foundation. The entire operating system including Cocoa, Carbon, Quartz and Aqua runs and runs well on Intel CPUs. At one point there was also an Alpha port but that was discontinued well before Mac OS X went beta.

Apple won't release a general Intel port of OS X. It makes no sense for them to do so. Apple makes the vast majority of its revenue through hardware sales, somewhere around 90-95%. If they released Mac OS X for Intel their hardware sales would fall dramatically. Because the unit cost of an operating system is much less than the cost of a hardware box (say $100 compared with $2000) Apple's revenues would fall precipitously.

No company can gp to Wall Street and say: I'm going to chop my annual revenues down from $8 billion to $500 million. Can you imagine what would happen to the Apple stock price if they announced this? It simply can't be done.

So why do Apple keep the Intel port of OS X alive? After all it costs real money to keep all that software running cross-platform.

There are two reasons. First as a hedge against Motorola or IBM screwing Apple on the PowerPC processor. In the last few years the clock rate (and other key performance measures) of the PowerPC line has fallen a long way behind Intel. If IBM/Moto can't get competitive again, then Apple wants the option of putting Intel CPUs into Macs. This would not mean you could buy an off-the-shelf Gateway/Dell/whatever and run OS X on it. You can bet Apple would make sure it only ran on a "real" Mac to preserve their hardware revenues.

The second reason they keep the port up is because it helps them produce better code. Having to write code that runs on more than one CPU family is a good engineering discipline. The different architectures stress different parts of the code and you will often see bugs on one platform that are hidden on the other.

So Apple already have OS X on Intel, but don't expect to see it in the marketplace anytime soon.

Re:Why Apple has, and why Apple won't

[gwernol]

Where'd you learn this?

Four years as a senior software engineer on Apple's OS teams.

The last non-PPC port of OS X was Rhapsody DR2, to my knowledge, which lacked Aqua, Quartz, and Carbon. (It was, at that point, still essentially OPENSTEP 5 with a Platinum interface and QuickTime Media Layer injected, which at the time included QuickDraw GX and QuickDraw 3D.) It ran only on Intel and PowerPC.

What you say is true but incomplete. Mac OS X on Intel has been kept up until at least beta. After that I don't have first-hand knowledge, but I'd guess they still build it, as most of the work was done then. Aqua, Quartz and Carbon were included. Classic was not.

After that release, Jobs announced that Rhapsody was DBA (Dead Before Arrival) and announced his new Mac OS X scheme, which included the fact that the new operating system would not run on Intel. Mac OS X DP1 and later did not run on Intel hardware. And at no point did I hear anything about Alpha, and find it highly unlikely if for no other reason than due to the Darwin sources that were initially released.

Well I've seen Mac OS X beta running on Intel and I've seen the source code that supports it too, so it is real. The Alpha port went away much earlier, back in the Rhapsody days.

As you may remember, when Darwin was first released, many people wanted it to run on Intel, and this ended up being a massive job that still isn't finished. It wasn't that anything had been removed; it's that it simply hadn't been maintained at all since the old Mach 2.5 version, so the foundation, while there, was simply horrendously out of date. Had Apple continued Intel ports, and especially if they had done an Alpha port, it seems as though that code would have been included as well.

Imagine the situation where Apple did not want the outside world to know that they were continuing to maintain an Intel port. They would have released a version of the Darwin source that had the Intel parts switched out. Internally Apple has a different Darwin source tree than the one that has been released to the community.

Recently, in fact, as Darwin's been gotten to limp along on a few varieties of Intel motherboards (and "limp" is definitely the right word here), Apple's been helping a bit with the Intel port, but, again, they're having as much trouble as anyone. No "Here's a secret250,000-line patch to make it work." Just problem solving line by line, conflict by conflict. Given all that, I've always regarded the "OS X is secretely running on Intel" rumor as just that. A rumor.

Well it isn't. I've seen it, used it, worked on it. It doesn't really matter if you think I'm wrong, I have been in a privileged position that you haven't, sorry.

Re:Why Apple has, and why Apple won't

[gwernol]

Sorry for doubting, but you didn't mention you used to work at Apple on the OS X Intel port. :)

Just to clarify, I didn't work directly on the Intel port but the software I was working on was ported and so I had to keep it maintained and tested on Intel.

Obviously that would change things a bit, to say the least. My one question, though, if you happen to know the answer, is why Apple hasn't bothered to give the Darwin community the source now, since they would still like it tremendously and it would be very helpful. I'm asking as a question of logistics, not truth.

Mainly because people would draw exactly the right conclusion if Apple did release it: that Apple is preparing to move away from PowerPC to Intel. That would cause a lot of problems for Apple with its investors, with Motorola/IBM (which isn't exactly a stable relationship at the best of times), with its current customers and with Microsoft. As has been noted elsewhere, if Apple did go head-to-head with Microsoft in the Intel-based OS market it would put Office for OS X at serious risk.

So I don't think its a matter of logistics. In fact because Apple has to maintain two Darwin code bases (internal and external) to logistics of not releasing the Intel version are somewhat costly.

Counterpundit

[The Famous Brett Wat]

Cringely's done his punditing, and I'll counterpundit in return. I'll take the safe bet that Apple will do no such thing any time soon. No big deal; most people here seem to think that. I haven't seen anyone mention the reason why I think it's so, though.

Technical hurdles and business considerations aside, cast your memories back to 1997 when Jobs shocked the world by teaming up with Gates. Remember that $150 million in non-voting Apple stock purchased by Microsoft, and patent cross-licensing deal? Anyone? Here's the Apple Press Release in case you forgot. Apple was in bad shape, and Microsoft was up for monopolistic practices. Jobs agreed to make IE the default browser for the Mac, and Gates agreed to give Office better treatment on the Mac platform.

According to my vivid imagination, Jobs had a word in Gates' ear, saying words to the effect that Gates could crush Apple like a bug if he cared to, but then he'd have no real competitor to point at in defense of monopoly charges. Why not just let Apple have its little niche, whispers Jobs to Gates, and we'll agree not to get cocky and muscle in on your turf? The IE and Office deals merely consummated the marriage, as it were. Jobs is happy because Apple gets to survive, and Gates is happy because he has a harmless competitor that he can act all panikcy about.

This is pure speculation on my part, of course, but if there's much truth in it, you can expect Apple to be totally uninterested in the OSX for PC idea. I'm thinking that both Jobs and Gates would still prefer a no-compete situation.

Re:Actually, the problem is still the apps.

[osgeek]

Speaking as a long-time Macintosh software developer, I literally drool at the possibility of selling my apps to an intel-sized audience with a simple recompile. Apple uses gcc, so setting a compile switch to generate the right binary will work without any hassles. BeOS had a similar PowePC to Intel transition, and building either binary couldn't have been easier. Well, okay, you had to install some extra libraries to build, but Apple would sort that out. Oh, and endian issues on the BeOS were rarely a problem (htonl() and its friends work quite nicely).

Trust me. Standard application developers won't be worried about shipping two binaries if it means doubling (tripling, quadrupling?) the market for their products.

Three things wrong

[jpellino]

1. Supporting MacOS on god-knows-what hardware configs is a nightmare that would cripple it's reputation. When WIN doesn't work, users don't call the box maker, they curse the OS maker. Something about WIN made all of you stop using it - some of that was lack of HW toleration - did you go buy a new box? Nope - you switched OSs.

2. Overpriced hardware is a myth bordering now on The Big Lie - go to Dell, Gateway, Compaq, HP and match any level of the new G4 iMac - then count yer change.

3. Bob, it WOULD cannibalize hardware sales - Apple's largest edge is the OS/box integration, the Mac faithful would still buy the mac boxes, but your average new user would - and does - buy the rattiest box they can find - blind to the reality of the $599 specials. And good luck getting it to run reliably on some box that, as is typical, doesn't even know the names of the cards slapped in it.

Sticking to HW/SW is not so bad - Apple knows that typical system turnover is about three years - would they rather rachet up to making box money or start tomorrow with a herculean effort at supporting all the hardware in the world to make license money? Think you can open a storefront and sell licenses? Or would you rather have a store that can sell someone a solution and make box money?

Anyone know what portion of their business MS makes on licensing the OS alone? Remember, MS makes a lot of software - odds are Apple would not - this number needs to be known before convincing anyone that ramping up the software biz would be their saviour.

I have an iBook2 with OSX because since day one, I open it up, it does everything I ask of it as a plain old person, teacher, writer, webmaster, admin, tourist, scientist, etc. I have yet to crash OSX after 11 months, anything I plug into it fits and works. It is an order of magnitude above any previous HW/SW I've seen or owned. I could run windows on it tomorrow.

But I won't, and not because of religion. because of integration.

Not Overpriced Hardware, it's STILL Microsoft's!

[BadlandZ]

I dissagree. The hardware isn't HALF the magic, it's ALL the magic. OS X is nice and all, but it's not going to make me buy ANY system, ever. I'm not worried as long as NetBSD, FreeBSD or Linux runs on the hardware, I'll take it if it's QUALITY.

Look at the iBook. Small, light, preforms decent. Try to find a brand name x86 for the same money with similar equiptment. Same for the iMac.

Yes, you can say that you can _build your own_ for less with x86. x86 to Apple is already comparing apples to oranges, so to further try to compare a home built to off the shelf brand name is not a fair comparison.

SO, what's the REAL problem with APPLE?

When you can get an iMac for $799, an iBook for $1199, and then have to pay $550 for MS Office X who wants to buy it? When you can get at least the basic MS Office bundled with almost all x86 brand name hardware for almost nothing!

Don't bother arguing the Open Source office suites to me, I know. That doesn't change the fact that public perception is in the believe that you NEED MS Office to make a computer useful.