[analogika]

Originally Posted by Simon 

So do I understand correctly, you think Apple should wait till August 2011 before they start adding Intel-only code to the OS?

No, I don't think so at all.

[Big Mac]

Originally Posted by Chuckit 

I even remember that Big Mac was being pessimistic that this would happen when Apple switched to Intel and everybody was like, "Oh, don't be silly. PowerPC will go the way of the dinosaur eventually, but Apple will give it a normal, decent lifespan."

Thank you, Chuckit. It's nice to receive recognition for being proven right, although I don't relish in it too much.

[0157988944]

Seriously... PowerPC users are NOT missing out. This is Leopard, enhanced for their new machines. The current leopard on PowerPC is the same thing comparatively.

[Big Mac]

"a new generation of core software technologies that will streamline Mac OS X, enhance its performance, and set new standards for quality" implies more than that, adam.

[0157988944]

and its all based on it's latest computers and their contained technologies. No?

[Big Mac]

No.

[0157988944]

OpenCL is the only thing that would remotely work with PowerPC Macs, and of those only G5s would really be worth it. I'm a bit surprised they don't include G5 support, but this is a future-proofing OS, and the future doesn't contain PowerPC. Its not like them making Leopard Intel only and leaving 2 year old PowerMacs stuck on Tiger. Most likely they will continue to encourage Universal Binaries that will work on Snow Leopard and Leopard PPC and Intel.

[mduell]

Even using 4 year support period, ~90% of the Mac installed base could run Snow Leopard if it were released this year; by next year that will be >95% (using a 3 year standard next year would include ~99%).



Graphic from RoughlyDrafted

[Chuckit]

Originally Posted by Person Man 

And to be fair, Tiger has not "just" gotten security updates and bug fixes only. What do you call QuickTime updates? There have been security only and bug fix releases, but also ones that add functionality, like iTunes movie rentals. Tiger has been getting new versions of iTunes. Tiger also got Safari 3.0 and early indications are that Tiger will likely get Safari 4.0 as well if the developer previews available on ADC are to be believed.

Those aren't bug fixes or feature additions to the OS, they're updates to standalone applications that are also available on Windows. Tiger did not get the new version of, say, Mail.

Originally Posted by Person Man 

Honestly, what would you do in Apple's place? Let's say that it takes x time to develop Grand Central for the Intel Core architecture, and it takes 2x time to develop it for PPC because there are enough differences that "just clicking the checkbox in XCode" won't cut it (and if you think that it's "just a matter of clicking the checkbox" for something like this then I've got a bridge I'd like to sell you). Knowing that PPCs would only have a "useful" life of 1.5 to 2 years vs. Intels, would you seriously say, "Well, we have to do it because we promised 5 years of support."

If it were really impossible to port Grand Central to the PowerPC, I would announce that Grand Central requires an Intel processor, not the whole operating system. They've done similar things with Altivec support, Quartz Extreme, Core Image, etc.

Originally Posted by adamfishercox 

Its not like them making Leopard Intel only and leaving 2 year old PowerMacs stuck on Tiger. Most likely they will continue to encourage Universal Binaries that will work on Snow Leopard and Leopard PPC and Intel.

Unless their marketing material is quite deceptive, we already know there are going to be new features and APIs. Whether it's "like them" or not, that's what they're doing.

Originally Posted by mduell 

Even using 4 year support period, ~90% of the Mac installed base could run Snow Leopard if it were released this year; by next year that will be ~95% (using a 3 year standard next year would include ~99%).

You know that doesn't make a difference to me, right? Those numbers don't make my computer any better supported.

[lpkmckenna]

Yeah, that graph pretty much kills any reason for Apple to release 10.6 for PPC.

[Chuckit]

Originally Posted by lpkmckenna 

Yeah, that graph pretty much kills any reason for Apple to release 10.6 for PPC.

Not really. It's missing a year that would push the PPC's numbers up (if we're going for a five-year horizon) — and also, Apple of all companies should realize that 10% of the market is not insignificant.

[Person Man]

Originally Posted by Chuckit 

You know that doesn't make a difference to me, right? Those numbers don't make my computer any better supported.

It comes down to you, huh? (Not meant to be an insult).

Every individual computer user thinks their needs are the most important. It's human nature. I'm not any happier than you about the loss of PPC support either, but I understand the reasons for doing it.

EDIT: You know, if you actually read that series of articles from which that graph came, the author makes a pretty good case for the parallel release of Snow Leopard. Instead of a linear progression to 10.6 like every release before it, he's examining all the evidence and concluding that 10.6 really will be intended for newer Intel computers but also that 10.5 may continue to see point releases for a while after 10.6 comes out... at least until 10.7. I'm going to make a prediction of my own: When 10.6 is publically demonstrated it's going to calm many of the loudest complainers' fears rather than confirm them.

Suppose Grand Central and OpenCL don't make it into Leopard, but Leopard continues getting 10.5.x releases along side 10.6 until 10.7 comes out. Would that qualify as continuing support in your eyes (Chuckit, I'm talking to you) for PPC.

[Chuckit]

Originally Posted by Person Man 

It comes down to you, huh?

Pretty much, yeah. When somebody does me wrong, I don't particularly care that there are lots of other people he didn't screw over. Sure, I'm happy that lots of other people dodged the bullet, but I've still got a problem.

Originally Posted by Person Man 

You know, if you actually read that series of articles from which that graph came, the author makes a pretty good case for the parallel release of Snow Leopard. Instead of a linear progression to 10.6 like every release before it, he's examining all the evidence and concluding that 10.6 really will be intended for newer Intel computers but also that 10.5 may continue to see point releases for a while after 10.6 comes out... at least until 10.7.

I believe that might be true, but if so, I think the point releases will all be of the kind where the patch notes just read "Improves overall security and stability" rather than the "Adds the features and APIs from 10.6" kind.

Originally Posted by Person Man 

Suppose Grand Central and OpenCL don't make it into Leopard, but Leopard continues getting 10.5.x releases along side 10.6 until 10.7 comes out. Would that qualify as continuing support in your eyes (Chuckit, I'm talking to you) for PPC.

Like I said, if the point releases mean I'm getting everything the Apple folks feel they can deliver for PPC and programs written for 10.6 will run on my Mac, sure. But if that were the case, why wouldn't they just call those point releases "10.6" since they'd include basically the same stuff? That's why I don't think it's going to happen.

[Simon]

Originally Posted by Chuckit 

I believe that might be true, but if so, I think the point releases will all be of the kind where the patch notes just read "Improves overall security and stability" rather than the "Adds the features and APIs from 10.6" kind.

Actually I'm pretty sure that will be the case.

I think it's quite clear what's going on. On one hand we have Apple that doesn't want to devote a lot of resources to <5% of their user base on the old platform. On the other hand we have those users (and there will be even less of them by the time SL ships in 2009) that want to maximize their PPC Mac's life time. Those are conflicting goals. And what Apple seems to be laying out is that those <5% of their user base will just have to switch their Mac if they absolutely need SL. That may be harsh for those forced to make that call, but on the whole it hardly sunds outrageous.

So in short, yes I can understand it's frustrating for those PPC users who want SL. But in terms of allotting R&D resources, I think Apple clearly made the right call.

[OwlBoy]

Damn, if only Macs did not get so popular during the intel transition

[Cold Warrior]

I suspect it's part conservation, part strategy. Develop for the future - which is around Intel chips - and simultaneously leave PPC users behind to entice them to upgrade. Apple has probably done extensive market and user base research and is rolling the dice (with favorable odds) that many PPC users have or will find the means to upgrade over the next few years.

[Person Man]

Originally Posted by Chuckit 

Like I said, if the point releases mean I'm getting everything the Apple folks feel they can deliver for PPC and programs written for 10.6 will run on my Mac, sure. But if that were the case, why wouldn't they just call those point releases "10.6" since they'd include basically the same stuff? That's why I don't think it's going to happen.

Not necessarily. Let's look at the features that Apple has publicly stated will be in Snow Leopard.

1. Microsoft Exchange Support.

"Snow Leopard includes out-of-the-box support for Microsoft Exchange 2007 built into Mail, Address Book, and iCal."

Mail, Address Book, and iCal with Exchange support are all standalone applications that could easily be added to Leopard. For $29.99, just like iChat AV for Panther.

2. Grand Central.

Sounds more like a more efficient implementation of multithreading and task scheduling. Meaning not something developers will have to explicitly support. The OS will do it for them, just as it always has, only Snow Leopard will be optimized to take more advantage of the Intel Core Series implementation of multicore. Since PPC is already a mature platform, there may not be too much more optimization they can do. Universal apps will run on both, but on Snow Leopard they will run more efficiently on Intel hardware.

3. 64 Bit.

64 Bit support is already in Leopard. Apple says "Snow Leopard extends the 64-bit technology in Mac OS X to support breakthrough amounts of RAM — up to a theoretical 16TB..." Well, I don't know about you, but my G5 certainly won't support 16 TB. Point releases of Leopard could easily include the parts of 64 bit support that can be placed on G5 systems and still fall under "improvements to overall stability," since Leopard is already touted as "fully 64 bit."

4. QuickTime X.

"...a streamlined, next-generation platform that advances modern media and Internet standards. QuickTime X features optimized support for modern codecs and more efficient media playback..."

Ok, so they're going to "streamline" QuickTime. Let's get rid of support for older, not much used codecs (if you need them on Snow Leopard, undoubtedly a third party will come up with plugins). QuickTime X will have optimized support for modern codecs. Optimized for Intel Processors. QuickTime 7.x is already mature on PPC systems, and any new codecs in QuickTime X could be added to Leopard with a new point release of 7.x.

5. Safari 4.

Tiger got Safari 3. 'nuff said.

6. OpenCL

Let's turn the GPU on your graphics card into a co-processor. Again, I suspect that much of that will be handled by the operating system "for free." I think developers will be able to access it themselves, but that won't make it into Leopard. I can see high end computation-intense programs requiring it if they explicitly call OpenCL fetaures rather than letting the OS do it for them, but then the requirements for said apps will probably exclude anything using Intel Integrated Graphics anyway so lower end Intel users will be in the same boat as PPC users.

EDIT: Can't forget what people have said about bug fixes. It goes without saying that if Apple continues releasing point releases of Leopard alongside Snow Leopard that bugs in Leopard (even longstanding bugs introduced way back in 10.2) will still be fixed.

Snow Leopard is about bringing Mac OS X up to the same level of maturity for Intel chips as 10.5 is for PPC chips. But wait, what about the secret Intel builds Apple maintained all along? Do you really think Apple made the Intel builds as optimized as the PPC builds? I don't think so. Hence the reason for Snow Leopard. Laying the groundwork for 10.7, which will be the next "full-featured" release.

[turtle777]

A Quantum Leap.
.... Taking a break from adding new features, Snow Leopard ...

Pfff, if that's all, I don't see many people switching.

Apple needs to give users great new features to make them pay $ 129 for a new release.

-t

[- - e r i k - -]

Originally Posted by Person Man 

5. Safari 4.

Tiger got Safari 3. 'nuff said.

Tiger is still getting Safari 4:

[- - e r i k - -]

Originally Posted by analogika 

It sounds like you're confusing them with the last dual-1.25 G4 towers, which were sold concurrently with the G5 towers, and *were* sold only for legacy support, for BOOTING into OS 9.

I stand corrected. Thanks.

[Brien]

Originally Posted by turtle777 

Pfff, if that's all, I don't see many people switching.

Apple needs to give users great new features to make them pay $ 129 for a new release.

-t

I think SL may have a smaller pricetag - not free like 10.1, but maybe $59 or $79.

[- - e r i k - -]

Here's a RD article on this subject:

Originally Posted by Roughly Drafted

Myths of Snow Leopard: 1 PowerPC Support
Apple’s limited comments on Snow Leopard, the next version of Mac OS X due in about a year, have opened the playing field for rampant speculation. Here’s a look at a series of myths that have developed around the upcoming release. The first myth of Snow Leopard:

Apple is dropping support for PowerPC Universal Binaries, so software will dry up for users of PowerPC Macs.

Snow Leopard is going to be the first version of Mac OS X that only runs on Intel Macs. However, this is not going to be a real problem for PowerPC Mac users. For starters, the existing Mac OS X 10.5 Leopard will continue to be updated. It’s not even a year old today. Tiger still receives regular updates and it’s now over three years old.

The new features in Snow Leopard will also have little impact on PowerPC users. Snow Leopard’s main emphasis is on multiprocessor support. Users with needs for high end applications taking advantage of multicore, multiprocessor machines have known about the Intel transition since WWDC 2005, so the idea of a Intel-only version of the operating system being delivered four years later is not a surprise. Technologies like OpenCL, Grand Central, and the new 64-bit kernel would all have negligible benefit for most PowerPC users. Anyone trying to stay on the bleeding edge of processing power shouldn’t be trying to use Macs that will be four or more years old at the release of Snow Leopard.

The main “missing feature” for PowerPC users stuck at Leopard might be Exchange integration. However, Apple could certainly sell Universal Binaries of the new Mail, Address Book, and iCal separately at minimal cost, if there were a significant enough market of users with PowerPC machines who wanted that. Since the biggest audience for Exchange support is in education and corporate markets, there will likely be pressure upon Apple to sell a Universal Binary version of those apps for their existing PowerPC users.


Universal Binaries Outside Apple.
The real question for most PowerPC users is whether third party application developers will continue offering PowerPC support. While Apple’s cost-benefit ratio for delivering Snow Leopard as a Universal Binary is high in cost and low in benefit, for most developers the opposite is true. Application developers already have their PowerPC code and can continue to deliver Universal Binary software with minimal extra effort.

Apple is developing a number of low level technologies that require a lot of work to complete and test in parallel for both PowerPC and Intel architectures; app developers are writing primarily to Apple’s higher level APIs, which abstract the differences between the two platforms. There may be some applications taking advantage of new features in Snow Leopard that would render them Intel only, but PowerPC users don’t need to be excessively concerned that the market for PowerPC software will vanish anytime soon.

Apple reported that roughly 37% of the 27.5 million Mac OS X installed user base is running Leopard, 44% is running Tiger and 19% is using an earlier version. That means that despite the huge influx of new Intel Macs sold in the last couple years, there is still a large active installed base of PowerPC Mac users to sell Universal Binaries to.


The Universal Binary Solution.
Because Apple has made it quite easy to deploy Universal Binaries, developers are not stuck with the dilemma of supporting past users or scuttling legacy efforts to only support new users, as was the case when Apple migrated from the original Motorola 68000 family to PowerPC.

In those days, while 68k code could run on PowerPC in emulation, the reverse was not true. That meant developers had to choose between migrating to PowerPC and killing their 68k development, splitting their efforts across two entirely different code bases, or staying in the past to service the installed base of 68k users.

With today’s Universal Binary architecture, developers can target both Intel and PowerPC Macs with minimal extra effort. One exception is Adobe, which has the unique circumstances of not using Apple’s tools. Adobe has built its own cross platform Windows and Mac development workflow around the Intel architecture, so its latest Mac titles are Intel only. That does not reflect on the future of apps from developers using Apple’s tools however.

Basically what I have said all along more eloquently

[besson3c]

It is rather disingenuous to say that "Tiger is still receiving updates". These are not updates but backports of security fixes. Tiger is off of Apple's radar, and Leopard may be too once Snow Leopard is out.

[Simon]

Originally Posted by - - e r i k - - 

Basically what I have said all along more eloquently

The exact same thought crossed my mind.

[Chuckit]

That Roughly Drafted article is just plain weird. A smaller company performing a significant rewrite of an application apparently has a large PowerPC-compatible code base that won't need any development or testing, but gosh darn it, making OS X run on the platform it was originally written for is just too much work. What?

If PowerPC users aren't worth Apple's time — the company that writes software that is already on every one of these people's computers — how are they worth anyone else's time? One minute the PowerPC user base is too small to be worth supporting, but now in this article they're telling us that there are tons of PowerPC folks and it's totally worth your time to write software for PowerPC (unless you're Apple, for some reason)? That's like three different stories all from the same source.

[analogika]

From what I gather, Roughly Drafted tends to post semi-stream-of-consciousness ramblings that are then actually thought through and picked apart in discussions on other sites - after generating a couple ten-thousand hits.

I really wouldn't take them seriously.

[CharlesS]

My favorite part was this:

The main “missing feature” for PowerPC users stuck at Leopard might be Exchange integration. However, Apple could certainly sell Universal Binaries of the new Mail, Address Book, and iCal separately at minimal cost, if there were a significant enough market of users with PowerPC machines who wanted that. Since the biggest audience for Exchange support is in education and corporate markets, there will likely be pressure upon Apple to sell a Universal Binary version of those apps for their existing PowerPC users.

[Catfish_Man]

Originally Posted by Person Man 

Not necessarily. Let's look at the features that Apple has publicly stated will be in Snow Leopard.
2. Grand Central.

Sounds more like a more efficient implementation of multithreading and task scheduling. Meaning not something developers will have to explicitly support. The OS will do it for them, just as it always has, only Snow Leopard will be optimized to take more advantage of the Intel Core Series implementation of multicore. Since PPC is already a mature platform, there may not be too much more optimization they can do. Universal apps will run on both, but on Snow Leopard they will run more efficiently on Intel hardware.

Wrong on all counts. There may be some optimization in the scheduler for hyperthreading and NUMA (both of which only apply to Nehalem) but that would be unrelated to Grand Central.

3. 64 Bit.

64 Bit support is already in Leopard. Apple says "Snow Leopard extends the 64-bit technology in Mac OS X to support breakthrough amounts of RAM — up to a theoretical 16TB..." Well, I don't know about you, but my G5 certainly won't support 16 TB. Point releases of Leopard could easily include the parts of 64 bit support that can be placed on G5 systems and still fall under "improvements to overall stability," since Leopard is already touted as "fully 64 bit."

No they couldn't; the kernel in Leopard is not 64 bit, and neither are any of the included apps except Chess and Xcode. Replacing those amounts to replacing almost the entire system.

4. QuickTime X.

"...a streamlined, next-generation platform that advances modern media and Internet standards. QuickTime X features optimized support for modern codecs and more efficient media playback..."

Ok, so they're going to "streamline" QuickTime. Let's get rid of support for older, not much used codecs (if you need them on Snow Leopard, undoubtedly a third party will come up with plugins). QuickTime X will have optimized support for modern codecs. Optimized for Intel Processors. QuickTime 7.x is already mature on PPC systems, and any new codecs in QuickTime X could be added to Leopard with a new point release of 7.x.

Optimized for ARM processors actually...

Snow Leopard is about bringing Mac OS X up to the same level of maturity for Intel chips as 10.5 is for PPC chips. But wait, what about the secret Intel builds Apple maintained all along? Do you really think Apple made the Intel builds as optimized as the PPC builds? I don't think so. Hence the reason for Snow Leopard. Laying the groundwork for 10.7, which will be the next "full-featured" release.

You overestimate the effects of CPU-specific optimizations greatly. The *vast* majority of possible speedups in software are algorithmic, not twiddling instruction scheduling or choice for a particular chip.

[besson3c]

There are pieces of OS X that may never be written to be 64 bit, and perhaps should never be. 64 bit does not equal faster.

[Catfish_Man]

Originally Posted by besson3c 

There are pieces of OS X that may never be written to be 64 bit, and perhaps should never be. 64 bit does not equal faster.

Anything using Cocoa definitely should be, since that allows it to use the new runtime (and that, in turn, allows them to eventually be free of their objc++ compatibility headaches and the fragile base class problem, as well as some nice perf boosts). That implies that every library Cocoa uses needs a 64 bit version. Whether the kernel should be or not isn't something I know enough about to say, but apparently it will be. So really, which parts of OSX are you thinking of?

[besson3c]

Originally Posted by Catfish_Man 

Anything using Cocoa definitely should be, since that allows it to use the new runtime (and that, in turn, allows them to eventually be free of their objc++ compatibility headaches and the fragile base class problem, as well as some nice perf boosts). That implies that every library Cocoa uses needs a 64 bit version. Whether the kernel should be or not isn't something I know enough about to say, but apparently it will be. So really, which parts of OSX are you thinking of?

I was thinking of the kernel primarily, as well as various pieces that may not be written in Cocoa (e.g. BSD, Quicktime, Aqua, etc.) but rather possibly Assembly or straight C. I understand that Assembly is always going to be faster than a rich language like Cocoa that includes a lot of garbage collection and whatnot?

[Catfish_Man]

Originally Posted by besson3c 

I was thinking of the kernel primarily, as well as various pieces that may not be written in Cocoa (e.g. BSD, Quicktime, Aqua, etc.) but rather possibly Assembly or straight C. I understand that Assembly is always going to be faster than a rich language like Cocoa that includes a lot of garbage collection and whatnot?

Remember, any library Cocoa links against (libsystem, quicktime, etc...) has to be 64 bit for Cocoa to be 64 bit. Also, most people these days don't know how to write fast ASM

[besson3c]

Originally Posted by Catfish_Man 

Remember, any library Cocoa links against (libsystem, quicktime, etc...) has to be 64 bit for Cocoa to be 64 bit. Also, most people these days don't know how to write fast ASM

So, what's left is possibly the BSD subsystem, kernel, and non-Cocoa languages... There are 64 bit versions of FreeBSD available, but I don't know exactly what components are 64 bit. I can't really think of any purpose in making the kernel 64 bit though, can you?

Edit: I guess you need a 64 bit kernel to deal with more than 4 gig of RAM, even if the kernel itself is not going to be utilizing this RAM.

[Catfish_Man]

Cocoa certainly calls into the BSD subsystem. open(), pthread*, etc...

It also uses Carbon, CoreFoundation, CoreGraphics, etc...

[besson3c]

Catfish: so it's not necessarily a case of justifying moving things to 64 bit for a performance boost, but to consolidate frameworks (for lack of a better term)?

[lpkmckenna]

Originally Posted by analogika 

From what I gather, Roughly Drafted tends to post semi-stream-of-consciousness ramblings that are then actually thought through and picked apart in discussions on other sites - after generating a couple ten-thousand hits.

I really wouldn't take them seriously.

Yeah, here he is again, with no idea about the real world:
sproutcore-mailroom-managing-obamas-presidential-campaign/

[Catfish_Man]

Originally Posted by besson3c 

Catfish: so it's not necessarily a case of justifying moving things to 64 bit for a performance boost, but to consolidate frameworks (for lack of a better term)?

It's more that to get the benefits in the high level frameworks, the low level ones have to be converted as well. A lot of this work is done already, actually. Almost all* the libraries are 64 bit ready in Leopard, it's just that the apps and the kernel aren't. Once all the apps are 64 bit, they can drop 32 bit support, and reap some of the internal rewards (cutting installed size, being able to change instance variable layouts in Cocoa, cutting their compile times by a lot, etc...).

*given the fixes going in semi-recently, I'm pretty sure WebKit isn't, for example, but AppKit, Foundation, libsystem, etc... are.

[Chuckit]

Originally Posted by besson3c 

I was thinking of the kernel primarily, as well as various pieces that may not be written in Cocoa (e.g. BSD, Quicktime, Aqua, etc.) but rather possibly Assembly or straight C. I understand that Assembly is always going to be faster than a rich language like Cocoa that includes a lot of garbage collection and whatnot?

Incidentally, most of the C system libraries were already 64-bit in Tiger. Leopard's contribution was bringing support to the higher-level frameworks. The old QuickTime API is not 64-bit, but QTKit is.

[- - e r i k - -]

Originally Posted by Chuckit 

That Roughly Drafted article is just plain weird. A smaller company performing a significant rewrite of an application apparently has a large PowerPC-compatible code base that won't need any development or testing, but gosh darn it, making OS X run on the platform it was originally written for is just too much work. What?

If PowerPC users aren't worth Apple's time — the company that writes software that is already on every one of these people's computers — how are they worth anyone else's time? One minute the PowerPC user base is too small to be worth supporting, but now in this article they're telling us that there are tons of PowerPC folks and it's totally worth your time to write software for PowerPC (unless you're Apple, for some reason)? That's like three different stories all from the same source.

It is whining like this that makes me want to think "good riddance" to PPC users instead of having an actual discussion.

[- - e r i k - -]

Originally Posted by analogika 

From what I gather, Roughly Drafted tends to post semi-stream-of-consciousness ramblings that are then actually thought through and picked apart in discussions on other sites - after generating a couple ten-thousand hits.

I really wouldn't take them seriously.

While some of his writing can tend to be like that, he does tend to take a common sense approach to all his writing and projection, which is more than can be said of most armchair analysts here.

After initially getting the same "vibe", I place him second only to John Siracusa in terms of real world thinking about the world of technology.

[Chuckit]

Originally Posted by - - e r i k - - 

It is whining like this that makes me want to think "good riddance" to PPC users instead of having an actual discussion.

That wasn't whining. It was pointing out that the site is taking incompatible positions.

1. It makes sense for Apple to cut off PowerPC support because there are so few PowerPC users these days.
2. It doesn't make sense for developers to cut off PowerPC support because there are still so many PowerPC users these days.
2a. It still makes sense for Apple to cut off PowerPC even though there are lots of customers there. Somehow.

[voth]

Is there a particular reason why all the nonsense over Apple dropping PPC support in their next release?

I mean it was bound to happen sooner or later, for those with older models (myself included) you should know by now how fast the computer industry is, how often there are new 'gadgets' (including new versions of software) brought into the market. So before you continue to say 'WHY!!!!' just understand this is how thing are, there is nothing you can do to change it.

Get on the bandwagon and pick a new machine already (really I would suggest waiting awhile), but be glad Apple is more forthcoming on what their operating system requires to run (unlike a certain Redmond company).

Just think by that time replacement hardware (video cards specifically) for the next big mac (no pun intended) will be down around ground level and not in the stratosphere like the current hardware is.

I'm I already know that.

[besson3c]

voth: your comments are very pragmatic, but the problem is that if we all accepted that and were not upset enough to complain to Apple in one way or another, it will make it all the easier for the computing industry to consider this a new established precedent. A 3 year life cycle for an OS is unprecedented.

Similarly, this is a dumb and inaccurate comparison but I'll throw it out there anyway, if a politician is involved in some shady deals, it is easy to excuse that by overlooking it, saying that that's generally what politicians do, etc. However, if everybody did that, what would be stopping another shady deal from happening?

No matter what you are putting your foot down about, it is important that we have people to do this. This should not be discouraged.

[Brien]

So with all this 64 bit hoo-ha is SL going to support Core 1 machines or is that still in the air?

[CharlesS]

Originally Posted by - - e r i k - - 

It is whining like this that makes me want to think "good riddance" to PPC users instead of having an actual discussion.

It is personal attacks like this, completely devoid of any content whatsoever, that often make me want to say "good riddance" to MacNN. Seriously.

[TheoCryst]

Originally Posted by Brien 

So with all this 64 bit hoo-ha is SL going to support Core 1 machines or is that still in the air?

As of right now, it's still supported. No word on whether the final release will be or not (remember: early Leopard builds offered 64-bit Carbon support as well).

[- - e r i k - -]

Originally Posted by CharlesS 

It is personal attacks like this, completely devoid of any content whatsoever, that often make me want to say "good riddance" to MacNN. Seriously.

Don't let the door hit you on the way out.

[Cold Warrior]

Just let it go guys. Let's stay on track.

[TheoCryst]

Originally Posted by Cold Warrior 

Just let it go guys. Let's stay on track.

QFT. Anyone had their hands on a WWDC build that feels like violating their NDA?

[besson3c]

CharlesS: please don't go. I like learning stuff from you.