More Evidence Apple is Shifting Towards Intel-Only Software Releases?

[Dokter_Mac]

The difference here is that Apple isn't writing high level Objective-C code for the core operating system. So it is not as easy as checking a box, they would have plenty of assembly code to port over from x86 to PPC especially with the performance enhancements they are doing with OpenCL. By eliminating PPC support they save many man-hours porting machine code back to PPC.

Okay, it's not always that easy as checking a box in Xcode. There are some nuances like little endian, big endian, %d, etc.
But claiming that eliminating PPC support will save man-hours in porting is just not right. I work for a company which ports x86 Windows apps into Universal apps for OS X. Some of those apps are huge with millions of lines of code. I can say for sure that porting/building Universal apps is not at the cost of man-hours!
The main reason OS X ports will become Intel only is performance. Future programs will need more power/performance. The only way for getting more power/performance is new/future hardware. Apple has chosen for Intel hardware so it's therefore logical that the solution will be Intel only programs. Not because building/porting a Universal program (or even a system) costs more in man-hours.

About OpenCL: OpenCL is a technology that will run mainly on future hardware. At this time only a few Intel Macs are capable to make "full use" of the "current" technology/specs.
OpenCL is "OPEN" (not CLOSED). Open is not only Intel or nVidia. Open will be Intel, PPC, ATI, nVidia & many more.
A quad core G5 (even a dual proc) with a high end GPU is in a way equally OpenCL ready then a current Intel Mac. It depends on the hardware and how many of the "current" optimized OpenCL code will be utilized.
Sure the latest/newest hardware/mac will have the highest benchmarks with OpenCL fully utilized. But that does not mean that the quad G5 will not benifit from OpenCL (optimized for PPC G5).

 

[Dokter_Mac]

Let me solve that problem for you: OpenCL isn't supported on PPC.

OpenCL 1.0
OpenCL 1.0 at a glance
OpenCL (Open Computing Language) is the first open, royalty-free standard for general-purpose parallel programming of heterogeneous systems. OpenCL provides a uniform programming environment for software developers to write efficient, portable code for high-performance compute servers, desktop computer systems and handheld devices using a diverse mix of multi-core CPUs, GPUs, Cell-type architectures and other parallel processors such as DSPs.

"Cell is a heterogeneous chip multiprocessor that consists of an IBM 64-bit Power Architecture™ core"

 

[Eric S.]

Your G4 PM would be handily outperformed by a high end Mac Mini.

Performance isn't everything. I want multiple Firewire ports, PCI slots, and the ability to easily add more memory and internal disks (that aren't laptop size). I want a midrange tower. I prefer my G4 over the mini.

 

[Amdahl]

OpenCL 1.0 ... Cell-type architectures and other parallel processors such as DSPs.

"Cell is a heterogeneous chip multiprocessor that consists of an IBM 64-bit Power Architecture™ core"

I didn't mean that PPC was incapable of doing OpenCL, just that if supporting it required new development effort for Apple, they could simply not support it on their PPC. Clearly, PPC is one of the top chip architectures in the world, thanks to IBMs design & fab capabilities.

 

[BenRoethig]

Performance isn't everything. I want multiple Firewire ports, PCI slots, and the ability to easily add more memory and internal disks (that aren't laptop size). I want a midrange tower. I prefer my G4 over the mini.

Not only that but taking a page from the iMac's strategy page, you need basically two power cords for all of that. Start adding external devices to the Mini or iMac and you have a mess on you desk and a full surge protector.

 

[Joe The Dragon]

Performance isn't everything. I want multiple Firewire ports, PCI slots, and the ability to easily add more memory and internal disks (that aren't laptop size). I want a midrange tower. I prefer my G4 over the mini.

Also the g4 can do dual head video and run better video cards then the mini.

 

[twoodcc]

i hate to see support for ppc gone, but i can't blame them for it.

 

[newb16]

Can't see why this would be a surprise to anyone. In fact, Apple have held onto PPC applications longer than I thought they would.

Aug 2006 ( ppc power mac discontinued) - Mar 08 (iphone sdk) is longer than what?

 

[JFreak]

I'd go as far to say that NT4 was probably the best version of Windows ever made.

I agree. Until SP3, that is. After that things began to break.

It was cross platform (ran on RISC), it had SMP, it had the best OpenGL support (last Windows OS where Microsoft used OpenGL as there main graphics API). It's still capable of computing on a modern level and even on an old Pentium 3 it will boot and run twice as fast as Vista or XP on a Core 2 Duo.

Yes; it was designed to be cross-platform, because it was developed by multi-architechtural team. But once the coalition broke and Microsoft began to develop the system, the cross-platform idea was thrown out of the window. Pun intended.

2000 was good for the most part. XP was 2000 with a piss poor GUI but still stable and reliable. Yes OK, Vista did suck.

NT4SP3 was the last good Windows. After that, the solid core was broken bit by bit, because they needed multimedia more than stability and security. Originally there was plans of NT5, but once the original developers went away, the plans were suddenly null and void. W2K was nothing but broken NT4 with added multimedia. If you were lucky, you didn't notice it was broken but many of us suffered. And yes, you're right, XP is nothing more but W2K with uglier skin. Nothing is fixed or improved.

Vista core is a good start in Microsoft realizing what damage they've done in 10 years. It sucks now, but mainly because they have too many versions and too much trouble supporting them. Let them release Windows 7 with fewer versions and I promise I will give it a shot (using PC's of my employer). Shoud the W7 be half as good as NT4SP3, it's gonna rock the Windows people's worlds apart, because not many of them remember how great NT was. And it was even better before the W95 GUI

 

[ZipZilla]

Oh my. You People are really toeing the company line on this one.

The greater demand for cores and clock speed has more to do with hurried, sloppy coding by programmers and forced obsolescence than it does anything else.

"Learn To Play" lessons in Garageband no more require an Intel Core Duo machine than TextEdit does. It's just that unrefined bloatware won't run on anything else.

The idea that a 2.5 ghz quad-core PowerMac G5 sold 2 1/2 years ago can't run iLife '09 or the new OS is ridiculous. Intel Core Solo Minis are out in the cold too.

Apple can do what they want, but people shouldn't sugarcoat the decision as being a technical one, because it isn't.

 

[andyljohnson]

Time to go

My old 12" powerbook was a great machine, but I love MBP. It is time to move on. We can't get revolutionary software if we hang onto the past. That said, it can be cost prohibitive for people to upgrade, and it is sad that they will be left behind.

 

[MacDaniel007]

heard this before!!

... I don't mind them dropping support for the non-intel processors, it should allow them to downsize the operating system, make OS X even more stable and faster than it is now (which is already pretty stable and fast, but there's always room for improvement) and allow for the operating system to take up less disk space. I will still run my Mac Mini until it dies completely, which from what I have seen so far may be never, but I am not going to complain when there is no more support for it, as I understand there has to be that cut off sometime. Windows Vista will only run efficiently on systems that were about 1 year old when it came out, so really not an issue.

One of the promises of Vista was removing code for backwards compatibility- "streamlining" the OS as has been mentioned here. XP did not fill a CD, Vista fills five or so. Ahhh... The real issue with all software and hardware is what do you need it for?? I love the latest and greatest technology, but for real work you avoid it. Ubuntu has the "latest" and a long term support -stable release for real work. Software/ hardware companies want to sell stuff!! not make it perfect, then support it forever. How many people just must have the latest word processor features? I have not seen stats, but I bet the number using new features is VERY low (except for multimedia perhaps )

 

[MacDaniel007]

revolutionary software....

My old 12" powerbook was a great machine, but I love MBP. It is time to move on. We can't get revolutionary software if we hang onto the past. That said, it can be cost prohibitive for people to upgrade, and it is sad that they will be left behind.

What software is revolutionary? Revolutionary would be software that had bugs and user driven features moved as close to perfection as possible before being abandoned for the next "revolutionary features" and the next version. Is Office 2007 for example, revolutionary? Are 64 bit processors really critical for your average application?

 

[MacDaniel007]

The difference here is that Apple isn't writing high level Objective-C code for the core operating system. So it is not as easy as checking a box, they would have plenty of assembly code to port over from x86 to PPC especially with the performance enhancements they are doing with OpenCL. By eliminating PPC support they save many man-hours porting machine code back to PPC.

Actually, operating systems are (mostly) written in higher level languages like C++ these days. Most fussy, assembly level programming is for the beginnings of a compiler, or to speed up a very small, specific bit of code. Modern compilers are very good, and nothing is worse than trying to figure out some clever, poorly documented trick in assembly. Forth can be written in itself!! (see the open firmware standard that some macs use)

 

[iTattoo]

I Can Understand Why Apple is going this Way

Let's start with a couple of assumptions. 1, lets assume that when Snow Leopard was announced last June, that "about a year from now" stretches out, like it did for the release of Leopard, and like it does for most major software releases, and Snow Leopard comes out in late October when Leopard turns 2, and 2 - lets assume that most people running on a G5 machine by that point in October aren't early adopters, and would wait until at least 10.6.2 before upgrading, which would take until this time next year before we see that. At that time the most current G5 machines aside from the PowerMac are 4 years old, or more. At 4 years of age many consumers are starting to consider replacing their machines, and even if they aren't the current OS they are running will continue to work.

The vast majority of the G5 and G4 machines out there are not in the category of the PowerMac, and PowerMac represented only a small portion of the sales at that time as they do now.

Further lets not loose sight of the fact that for the past several years Apples sales have been increasing by leaps and bounds. By the time Snow Leopard, which is supposed to contain very few if any new features aside from improved background performance for future growth of the OS, will work on the vast majority of the machines in circulation.

Lets not also loose sight of the fact that this OS is not supposed to be a major upgrade from the user experience point of view, and Apple will continue to offer security updates for Leopard.

It would likely be more cost effective for Apple to offer a generous trade in offer for G5 PowerMac users on the purchase of a new Mac Pro, than to spend the money to develop software for what is figurative speaking only a handful of the existing customers.

 

[designergrey]

And so on and so on and so on....

I'm glad my world has long stopped revolving around Apples product releases. In time I am sure that I will be all Intel, but for now I am all over the map which doesn't really bother me.

I am currently using a G3 mac with a G4 processor upgrade in the kitchen for recipes and websites and just random daily stuff and it works great, a 500mhz Powerbook upstairs for pretty much the same kinda random stuff, a G5 dual core in my home office connected to a myriad of monitors printers external hard drives and my Intel Macbook Pro that bridges the other G5 and Intel MacPro server that I use at work. I am between Jaguar (I think its Jaguar, I have lost track) to Leopard and rarely find myself caring which OS i'm using. But what Apple decides to support and not support is unimportant. In the mix are several versions of Quark, Creative Suites, Suitcase Fusions and everything else in between.

Nothing is ever on the same page at the same time. They all have to stand in line. They will all get my money soon enough. I am a legacy user, not the target audience. They got me so what do they care. You early adopters and the holdouts are not all that important either...they got you too. New users to the market, that's where they gotta go and they need to wow, dazzle and advance the market. Which means adios PPC. Its not personal its business and Apple could care less that you bought your PPC just three short years ago. They got you too. Where do you think you are gonna go? You're not going anywhere.

Little Sally is heading to college and "Snow Cat" just came out and like a crack addict to a lit pipe she's headed right towards the light and all Apple cares about is that she ends up getting a taste for a little Apple crack before the Vista dealer gets to her first. Once she's hooked....she is just the next PPC waiting to happen. Sally won't care, her new rhinestone covered Mac will run Snow Cat just fine—for now. And the cycle continues.

So I really don't see the point getting upset with the PPC strongholds or the ones that are packing their camping gear as we speak to be the first ones to get the latest gospel from almighty Apple. The world won't end overnight. Those PPC will still be gettin -er done the day after. We need the earlier adopters to piss and moan about the bugs and get the product where it needs to be for the mainstream.

In the meantime, I can still run a business quite well on any of the given version of OS. While Leopard is nice, it has never been a deal killer. But it is a nice upgrade.

I think that if you waited to the end of a production cycle to get the last of the PowerPC systems then you probably are not an earlier adopter anyway and having the fastest, latest, and newest isn't priority #1. I have been on both ends of it and I felt more burned by being first than somewhere mid pack. The blue & white G3 was a killer since there were no USB products to support it. That sucked.

Really, Snow Leopard? no wonder PC users think mac users sound so lame.

The debate is fun to read though. Since this is my first and most likely one of my only, I will get back to just reading and leave y'all to it!

 

[Skyfarmer]

The thread title is :

More Evidence Apple is Shifting Towards Intel-Only Software Releases?

but I have been hearing rumours that they (Apple) are already planning for the next step when they leave Intel behind and move on to something new.. some form of modern upgraded PPC arcitecture perhaps?

I think we are talking 4-5 years from now, just wondering if anyone else had heard this?

 

[BenRoethig]

The thread title is :

More Evidence Apple is Shifting Towards Intel-Only Software Releases?

but I have been hearing rumours that they (Apple) are already planning for the next step when they leave Intel behind and move on to something new, some form of modern upgraded PPC arcitecture.

I think we are talking 4-5 years from now, just wondering if anyone else had heard this?

Only if they want to lose a bunch of money in the process. Trust me, if they were still using PowerPCs, no matter how advanced, they wouldn't be making the kind of profits they are now.

 

[Eric S.]

but I have been hearing rumours that they (Apple) are already planning for the next step when they leave Intel behind and move on to something new, some form of modern upgraded PPC arcitecture.

That's just silly. There are strong arguments against that, and not much that I know that favors it.

I think we are talking 4-5 years from now, just wondering if anyone else had heard this?

Only wild speculation from people who have no real knowledge of Apple's future plans.

 

[Skyfarmer]

Repeated below...double posted

 

[Skyfarmer]

I probably should have said;

some form of modern upgraded PPC arcitecture.?

I was not told what the move was going to, but it was implied that they were definetly moving away from Intel

As to what, I have not idea....probably should have started a new thread for this question really. Thought I would post to see if anyone else had heard similar and could shed any light on it

Only wild speculation from people who have no real knowledge of Apple's future plans

Yes and No.
I personally have no idea of Apples future plans BUT they guy I talk to does contract work for them. He mentioned that he is working on something for them for the future that does not include Intel....again no idea what and I did not press for any more details as I knew I would not get them I did not want to push him in to a situation where he would have had to say NO

Just interested if anyone else had heard similar

 

[themoonisdown09]

That's just silly. There are strong arguments against that, and not much that I know that favors it.

I agree. Why would Apple switch to Intel and then switch back to PPC? One huge reason there have been so many "switchers" is because it's capable of running Windows natively. If they switched back to the PPC architecture, this would not be supported.

 

[Peace]

I probably should have said;

some form of modern upgraded PPC arcitecture.?

I was not told what the move was going to, but it was implied that tehy were definetly moving away from Intel

As to what, I have not idea....probably should have started a new thread for this question really. Thought I would post to see if anyone else had heard similar and could shed any light on it




Yes and No.
I personally have no idea of Apples future plans BUT they guy I talk to does contract work for them. He mentioned that he is working on something for them for the future that does not include Intel....again no idea what and I did not press for any more details as I knew I would not get them I did not want to push him in to a situation where he would have had to say NO

Just interested if anyone else had heard similar

Apple will be using different chips for embedded devices in the future. That's probably what your friend was referring to.

 

[Skyfarmer]

Not sure, I know we also discussed some changes that were going on in relation to the onboard devices, but also he mentioned the processor side of things too..


We were rebuilding a motobike at the time, so it was not a major topic of conversation

 

[Eric S.]

I was not told what the move was going to, but it was implied that tehy were definetly moving away from Intel

As to what, I have not idea....probably should have started a new thread for this question really. Thought I would post to see if anyone else had heard similar and could shed any light on it

Well they did move away from Intel's chipset in the latest Macbooks by using Nvidia instead (although they'll probably move back to Intel at some point), but that's different than the CPU line. AMD would be Intel-compatible but one wouldn't think Apple would switch to them with their recent issues in being late to get their products to market.

I personally have no idea of Apples future plans BUT they guy I talk to does contract work for them. He mentioned that he is working on something for them for the future that does not include Intel....again no idea what and I did not press for any more details as I knew I would not get them I did not want to push him in to a situation where he would have had to say NO

Just interested if anyone else had heard similar

It's not unlikely that Apple tests OS X on other CPU architectures, including the latest PPC. The only claims I've seen that Apple will switch back are from people who miss the good old PPC days.

I agree. Why would Apple switch to Intel and then switch back to PPC? One huge reason there have been so many "switchers" is because it's capable of running Windows natively. If they switched back to the PPC architecture, this would not be supported.

That's one big reason. Others would be the wide availability of industry-standard parts, which results in higher margins on their products, and the lack of PPC mobile CPUs (the inability to produce a Powerbook G5 being one of the main instigators of the switch). On the negative side, going with Intel means losing some sales to hackintoshes and Psystar, but that's probably not much revenue lost.