Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Wouldn't be surprised if apple starts developing and producing their own chips at some point...for their entire product line.

Maybe they managed to solve the problem of software compatibility.
 
  • Like
Reactions: Benjamin Frost
I'm guessing this is where we're headed. The writing has been on the wall for some time.
And the same will happen again.. Apple or ARM unable to come out with good CPU upgrades on an annual basis.

The same happened with PPC, and now we are seeing Intel having the same issues.

Moving to ARM or any custom Apple processors solves nothing.
 
ARM needs to step it up and give intel some more competition
Issue is ARM and Intel are hitting very different markets. ARM is looking at the mobile and wearable market with emphasis on lower power. From that, line widths are not as much of a concern.

For Intel it is all about speed and performance. As the desktop market starts to shrink, they are moving into workstation and server markets that are all about CPU performance.

If anything, this will further fuel the rumors of a ARM native OS X.
 
Last edited:
While I agree that Xcode on iOS would be awful, I would still like to see Xcode's Playgrounds for iOS; just something to do lightweight coding and jotting of ideas.
Instead of having to put a complier on iOS, maybe have it sent to a cloud for compiling to help reduce the workload.
 
  • Like
Reactions: jeremiah256
Unfortunately, not being a good move doesn't equate to a bad move. There's no competition. If you execute x86-64, you either wait for Intel to release a new processor or... Wait for Intel to release a new processor.

Or wait for AMD to do something if your performance level is satisfactory and want to save a few bucks for deployment.
 
  • Like
Reactions: Borin and vmistery
The day that we see photo/video/web professionals editing and coding exclusively on iOS with touch gestures, I will give you my paycheck.
There are over 3 billion people with computers (internet accessing devices) in this world. Source

There are less than 20 million people that would be editing code. Source

I think that 99% of the 3 billion don't need to be coding on iOS. Most of them just need email, word editor, spreadsheet, browser, movies, pictures, music, apps. And a large portion of those people only need basic capabilities in each of the categories i just mentioned. We need to stop looking at the world as if everyone is a techno-geek like most on MR. The truth is we are in the significant minority.

Now where is your paycheck :D
 
  • Like
Reactions: usersince86
I'll be the devil's advocate here. This is more of a marketing issue for Apple than a true product issue on many levels. The truth is - there is more computing power for a large part of the consumer population than they would even use or need. I am not discounting those that want/need more for video editing/production and other intensive work. But really - how much computer processing power do you need for Microsoft Office (or Apple's suite), facebook, twitter, youtube, etc. Which - let's face it - a HUGE portion of computers are used almost exclusively for. I think this is definitely an issue. Companies always want to tout more/faster/better. But for the consumer (many) it's not that much of an issue.
 
They simply don't care enough about the Mac that's all..

And you agree with me since it is Apple who doesn't want / care to update it is not like Intel has been standing still.

Internally I don't think Apple has much regard for Intel's road map and works largely at their own pace.
 
I'm guessing this is where we're headed. The writing has been on the wall for some time.

Apple's A9x core is already there. It's as fast as a Skylake chip for tablet CPUs.

Bump up the cache, clock frequency, core counts, power-budget, maybe add some HBM, and we're pretty much set for the next Macbook Pros. The only issue is adding x86-64 compatibility modes for legacy or virtualization apps.
 
This is silly. It's not as much about CPU processing power now as it is GPU, fast storage, modern i/o ports. I don't really care if Apple only updates every other year, maybe with silent speed bumps along. The soonest I update my machines is two years anyway, though it's been longer lately b/c Apple hasn't provided me a reason to buy a new machine. I do wish they would built their machines around the best parts available, at least for the "pro" machines. They don't even do that now, and you can't blame Intel for that.
I agree. Also the bus speed needs to be increased. Yes, I see the speed increases for the bold section above (my markup) being available. Also the RAM needs to be faster. These will augment the current CPU speeds more efficiently.
 
  • Like
Reactions: Mr. Retrofire
... maybe have it sent to a cloud for compiling to help reduce the workload.

That was tried on the new dead Palm OS platform as the Aries compiler. You had to compile your code in their cloud with no local compile option. It had very mixed reviews since source code is some of the most confidential data many tech companies have in their house. I still know of houses that, to this day, build on machines not connected to the Internet for fear or intrusion and IP compromise. Several houses dropped Palm OS development altogether when there was no data local build options as third party code was licensed with a clause of not placing the source code on a network.
 
Realistically speaking, there isn't a lot of day-to-day difference between my 2009 i7 iMac and my wife's 2015 i7 retina iMac. The same can be said among all the retina mac pros that we have around the office; the speed differentials are pretty marginal.

Actually there is a big difference-- the latter has SSDs. Which brings up the real point-- the main new product differential isn't the processor any more, it's the peripherals.
 
Apples update cycles face uncertainty without using Intel as an excuse.....

When does Apple reference Intel when speaking on product release cycles?

It's then professional Internet forum experts that use Intel as an excuse.
 
Instead of having to put a complier on iOS, maybe have it sent to a cloud for compiling to help reduce the workload.
And where would the 'cloud' be? Apple's servers?

I don't think many companies would like very much sending off source code to some remote server... oh, and then there's the reliability issues:
"Sorry, we can't compile your code, please try again later".

Or,
"You request is number 50,000 in the queue.. estimated build time, 5 hours..".

Developers like to Compile / Build frequently...

I don't see it working too well.
 
Last edited:
  • Like
Reactions: rbrian
That was tried on the new dead Palm OS platform as the Aries compiler. You had to compile your code in their cloud with no local compile option. It had very mixed reviews since source code is some of the most confidential data many tech companies have in their house. I still know of houses that, to this day, build on machines not connected to the Internet for fear or intrusion and IP compromise. Several houses dropped Palm OS development altogether when there was no data local build options as third party code was licensed with a clause of not placing the source code on a network.
That is interesting, I did not know that happened. Would that also apply to something like what IBM did with their Swift Sandbox website?
 
I wonder how long I have to wait until Apple puts a quad core into the 13". Which I prefer would become a 14" without the bezels.

Do we need 10um nodes for this to happen? By the change of the cycles it will take longer.
 
And were would the 'cloud' be? Apple's servers?

I don't think many companies would like very much sending off source code to some remote server... oh, and then there's the reliability issues:
"Sorry, we can't compile your code, please try again later".

Or,
"You request is number 50,000 in the queue.. estimated compilation time, 5 hours..".

I don't see it working too well.
I agree, it was just an idea. I would not be surprised if Apple, or someone else, tired it though.
 



In its latest 10-K annual report (PDF) filed last month, Intel confirmed the end of its long-heralded "tick-tock" strategy of delivering new microprocessors to the market. Intel originally introduced the product cadence to the world in 2006 with the launch of the "Core" microarchitecture, alternating "ticks" of shrinking chip fabrication processes with "tocks" of new architectures.

Over the past ten years, Intel has successively delivered new processor families based on this tick-tock cycle on a nearly annual cycle from its 65 nm manufacturing node all the way up until recently. The tick-tock release cycle allowed Intel to reestablish dominance in both the consumer and enterprise CPU markets and had given OEMs such as Apple a regular update cycle to rely on for annual product updates. But with chip updates stretching about beyond a yearly cycle in recent generations, Apple's product launch cycles have started to be affected.

In the face of the difficulties in maintaining the tick-tock cadence, Intel has announced that the launch of Kaby Lake this year as the third member of the 14-nm family following Broadwell and Skylake will mark the official end of the tick-tock strategy. Instead, Intel will move to a new "Process-Architecture-Optimization" model for the current 14 nm node and the 10 nm node.

Tick-tock-dead-1.png
This development is not unexpected, as semiconductor foundries have had increasingly tough times creating smaller process nodes as fabrication of smaller transistors has become increasingly expensive and complex. Transistors are rapidly approaching the physical limits of traditional semiconductor geometries, and the famous Moore's Law regarding transistor density has been formally acknowledged to no longer be valid.

Intel has no doubt moved to this new release model in an attempt to get back to a regular product and platform cadence as it struggles with the technological challenges of bringing new fabrication nodes to volume production. As noted in our Mac Buyer's Guide, many of Apple's Macs have gone without update for the longest time since we began tracking them, though Apple has yet to update to the available Skylake microarchitecture for its Mac line. Some product uncertainty is due to continue as the launch of Intel's Kaby Lake microarchitecture has been recently delayed to the second half of 2016 after Skylake suffered similar setbacks last year.

Article Link: Mac Update Cycle Faces Uncertainty as Intel Abandons Tick-Tock Strategy
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.