Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Mac OS X 10.6.3 Build 10D573 Seeded to Developers
- Thread starter MacRumors
- Start date
- Sort by reaction score
Oh boy this sounds fun.
I wonder what sweet nothings nVidia spoke into Apple's ears to bet the farm that they'd be able to supply integrated IGP plus I/O hubs for Intel processors from here to eternity.
As both a shareholder, an Apple customer and someone who is looking forward to the next Mac mini, I too would like to see some new Mac hardware releases.
Apple runs the Mac division as a separate business unit. The only reason I would see this unit getting less attention would be if certain things happened: decreasing sales, decreasing margins, decreasing market share, etc. None of those things are happening right now. The only continuing trend is that desktop computer sales shrinks as a total percentage of overall computer sales. That's not really a flaw in the business plan, it's really what the customers are buying.
If Apple were serious about changing the computer division, the first plausible change we would see might be them trimming the number of desktop models (iMacs, Mac Pros, and Mac minis) by reducing the number of configurations.
Again, we have not seen that and with their increased sales, my guess is that the desktops are staying flat whereas the majority of the growty is happening in the notebooks.
There aren't a lot of retail investors of AAPL like myself. As far as I can tell, even most of the participants here, at AppleInsider, and at other Apple-related forums aren't shareholders (based on their apparent ignorance of Apple's financials and overall business strategy).
Most of the outstanding float (about 72%) is held by institutional investors, like Janus, T. Rowe Price, FMR, State Street, etc. These guys don't really care about shiny new toys from Steve-O. They care about increased shareholder value, the numbers.
Frankly, I think the average fanboy's understanding of the value of Apple Inc. would deepen if they owned stock (or at least read the financial statements). Plus, it might result in some of them whining a bit less.
In the overall scheme of things, it really doesn't matter if 10.6.3 ships next week or next month.
So you're implying that Apple needs to wait until April 12 for the LLVM 2.7 release, recompile a bunch of critical core libraries like OpenGL with an unreleased, unproven brand-new compiler, requal everything that had been working adequately compiled on LLVM 2.6, then release 10.6.3?
And that this is a showstopper?!?
Why not release 10.6.3 with components compiled with LLVM 2.6 and then work on releasing LLVM 2.7 compiled libraries with 10.6.4?
I can understand the purported benefits of LLVM 2.7, but to delay the release of a minor maintenance release for such an item, one that is completely unproven as of now? Assuming that 10.6.3 is in an internal release candidate phase (since Apple doesn't publicly announce release candidates), it seems highly highly unlikely that they would just say, "oh yeah, let's recompile OpenGL with a new compiler in two weeks and see how that works".
Again, I reiterate my suspicion why they wouldn't simply freeze 10.6.3 and stick this in 10.6.4? Is this typical Apple behavior? Do they regularly change compilers in the late stages of a minor maintenance release?
This is a very fascinating scenario however I'm really having a hard time seeing the plausibility of it.
Please convince me why.
I'm with you. When I've worked on production code, the toolset was the first to get a hard freeze.
When you were counting down to a release, you didn't change anything in the tool chain unless you found a horrible bug that made the whole release questionable. Then you fixed the tool bug, reset the release countdown, and started testing again with the newly compiled bits.
Too late for new compilers....
Doesn't matter. There are still going to be bugs in 10.6.3. Thats just the way of software.
The point which you missed is this… Release 10.6.3 now to address the bugs you have found and fixed. There will always be another update with more fixes. If Apple is waiting for the perfect bug free update it will never be released.
Are you advocating that people who experience problems just drink the kool-aid and shut up?
Let's see if they can get this one out first.
Not only that, this seems to be the absolute antithesis of how Apple Engineering proceeds.
From what I can tell, they are very conservative in their approach, using older tools (some would argue that they're "well proven") and older libraries.
As far as I can tell GNU's version of the current release of gcc is 4.4.3. The version of gcc on my MacBook (10.6.2) is 4.2.1, apparently released in July 2007. I'm actually surprised the OpenSSL on my Mac is a fairly recent version (0.9.8l from Nov 2009) as usually this security library trails by versions from many months ago.
There is nothing that currently convinces me that Apple Engineering would be willing to drop in a new compiler for key libraries right before a minor maintenance release. This would be completely opposite of their normal behavior.
It makes no sense.
No, I am not. People who have legitimate, properly documented evidence of problems should submit their inquiries directly to Apple Inc.
If people want to submit new instances of bugs here, that's fine by me, although clearly the MacRumors.com News Discussion bboard is not a technical forum and such discussions are somewhat outside the scope of this venue. I have read some complaints here that were incorrectly attributed or worse, so poorly written that the complainant's message was getting hampered by the delivery of a really poor explanation. One would think that the tech support bboards at Apple would be a more suitable venue for such matters anyhow; a well-crafted bug report should receive better reception there, and possible some sort of acknowledgement in the form of a trouble ticket or case ID number associated with the person's Apple ID for proper and timely follow-up.
I am a strong advocate of the development of a down-ranking comment scoring system with filtering, one that would bury certain comments. Ignore lists aren't effective and don't scale. I'm continue to shock myself when I say this, but the comment scoring system at Slashdot has returned that site to usefulness after years of what I considered to be one of the worst online communities on this planet. The garbage posts now get buried.
A down-ranking comment scoring system here would essentially take care of those whiners who aren't legit. It's a community-based judgment per comment, and not specific to the author, but to the quality of comment's contents itself. Thus, no one is banned, but anyone who isn't making meaningful discussion (i.e., content that has long-term pedagogical value or is particularly amusing) is not seen at the default-level filters.
I do not disagree with your sentiment, but I have greater faith/belief in the engineering skill of Apple, one of the largest capitalized corporations on the planet, to have 2-3 backup plans.
On the other hand my faith could be misplaced in a zealots quandry . . . .
Let's watch.
I did specifically say Apple is trailing in releases relative to the industry segment you cited.
Yes, it is likely that Apple Inc. -- a Fortune 500 company and fourth-largest capitalized U.S. corporation -- built several prototypes with various combinations of components (CPUs, GPUs, chipsets, screens, etc.) in their investigations of potential avenues for development, rather than bet the farm on one number on the roulette table.
The current Nvidia-Intel legal standoff should not be a surprise to anyone in Apple Engineering. My guess is that alternate plans have been discussed on a regular basis in biz team meetings for well over a year, maybe even two.
Every time there's a post about about a new OS X build, there's always someone who asks this question.
Sorry, we don't know. This is not the best venue to be asking such questions. Find an OS X technical bboard. This is just a rumor site. Very few people here are technical and even fewer are developers. As far as I can tell, most are just fanboys and have never compiled anything on their systems.
I've seen a tremendous amount of problem solving on MR, and lots of good advice given. I'm absolutely in favor of that, and if people want to vent some steam, personally that's fine with me too.
Sounds absolutely draconian.
10.6.3 will be the last update for Snow Leopard lol. So they want to get it perfect.
10.7 is next
And they'll need five or six months to get it perfect
!(No significant software work is ever "perfect". QA is deciding which bugs to ship - not eliminating all bugs. The key is deciding "which bugs" are serious - Apple seems to be slipping on determining that lately.)
It sounds like you never read Slashdot at its worst (let's say circa 2005) nor have you looked at it in years.
The down-ranking comment scoring system basically took one of contemptible sites on the Internet and filtered out the noise.
If you have a legitimate argument/point/complaint and voice it clearly and back it up with facts, etc., your comment will be upranked by the community. If you mindlessly blather about off-topic nonsensical B.S., flame-bait, etc., your comment gets quietly buried.
The down-ranking comment scoring system lets the community moderate itself by deciding what it wants to be.
Forum moderation by individuals scales poorly and actually ends up with more draconian behavior. A classic example of this was Boing Boing's hiring of Teresa Nielsen Hayden who ended up being the wrong personality type for forum moderation. For a brief time, her use of "disemvoweling" got mentioned in the news, but ultimately, it's a bad practice since the detritus of the editing is still visible and scars the neighborhood. It's like shooting people and using permanent paint to outline their bodies, instead of chalk. The short-term vindictive satisfaction from embarrassing a miscreant is ultimately blights the community with long-term and permanent evidence of the moderation act.
A borderline commenter who visits a site using a down-ranking comment scoring system has several choices: A.) continue to blather and have most of one's comments ignored, B.) grow up and actually contribute in a more thoughtful and meaningful way, or C.) go elsewhere. The ideal result is B, but it's really up to the individual. There's no specific individual to blame in a community-based scoring system. The comment content is being judged, not the individual's account/login ID. There is no specific moderation team that can be blamed.
It's a peer vote. It's basically a truly democratic form of forum moderation. Your vote counts as much as anyone else's (unlike the forum moderator, who is basically a god and/or megalomaniac with a banhammer).
It also reminds new users that "First, suckers!" and "Yo Mama!" aren't considered valuable comments concerning long-term importance.
In many cases, forum moderation by individuals is highly variable and in the long run, simply not sustainable. Most forum moderators burn out before they quit and basically do a poor job for a period of time before their departure (some actually don't moderate at all, and basically let the forum run rampant).
MacRumors and AppleInsider (as well as many other tech forum sites) should take this threat seriously. After all, almost all of the big sites have gone through it (Slashdot survived). I refuse to read Boing Boing due to past moderation issues, and Digg's community is trash.
I think this is pretty insightful. Here is my guess as to what is going on: Because Intel forces the HD graphic processors on everyone with the mobile i3,i5 and i7 processors and the Intel HD is essentially identical (albeit faster) than the 4500mhd, and the 4500mhd was used ubiquitously last year in pretty much every consumer grade laptop sold, Apple has been forced to support this chipset with one big downside: The moment they release the drivers for these graphic chips, the hackintosh community will enable every laptop with an 4500mhd to become a macbook. That is what is bothering upper management about 10.6.3; It will need support for these graphic chips in order to support the mobile intel processors that MUST go into the new macbooks, and that will enable thousands of new hackintoshes. This pisses Steve off, I'm sure of it.
Anyone know if the problem with the desktop background changing after connecting to an external monitor is fixed?
Is the problem where after coming back from sleep or coming back from putting display to sleep and trackpad is non responsive fixed?
Is Safari snappier?
These are the only major problems that annoy me.
Anybody know?
Well, I'm on 10.6.2 here and don't have those issues - have you reported them to Apple? Do remember that issues suffered by some are not common problems - there are many variables in software/hardware/firmware which have nothing to do with the OS.
Do you have a link to a technical bboard on the TRIM/SSD/OSX issue? I've been ready to buy an SSD drive for a while, and the best time would be with a 10.6.3/10.6.4 release as im still on 10.5.8.
Its getting really tiring having apple fall way behind on this.