It's not so much that the iPad is lacking in performance as much as it'd allow the iPad to do even more. Better content creation apps, better games, better multitasking, and so on and so on and so on.
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.
Intel Looking to Cut Power Consumption on Future Ivy Bridge Chips
- Thread starter MacRumors
- Start date
- Sort by reaction score
It's not so much that the iPad is lacking in performance as much as it'd allow the iPad to do even more. Better content creation apps, better games, better multitasking, and so on and so on and so on.
I would assume you mean LED Backlit LCD displays, because technically (or at least in some situations) OLED meets that requirement. Yet Apple isn't using it (in anything).
It's more of, "How do we get Intel to fab our ARM Designs if we don't put an i5 in the Large iPad".
Samsung is done fabbing for Apple 6-30-14. Big issue for iOS.
Lots and Lots of $$ is the only way I could see Intel fabbing a chip for a competitor. Plus doing so would reduce Intel's ability to produce their own CPU's, so what ever they charge it would have to compensate.
So your assessment is that developers is some how holding back? It's my impression that a performance increase would be mostly beneficial in the GPU department, which has never been intel's strong point anyway, as opposed to raw CPU number crunching capabilities.
Two things.
1. Ivy Bridge was late. If Haswell is also late then Intel may need stop gap chips. Even if on time some Ivy Bridge design wins are going to continue even after Haswell appears (with a new socket design and supporting chipset).
2. Even 10 Watts is about 10x higher than the ARM designs. The only competitive designs Intel has for hand held mobiles up against ARM are in the Atom class. Medfield is the latest and has some wins.
http://www.engadget.com/2012/05/31/orange-san-diego-benchmarks/
http://www.anandtech.com/show/6307/motorola-hits-2ghz-with-intel-powered-razr-i
If look at the benchmarks the problem is that the Medfield doesn't keep up with the new A15 ARM designs like Exynos 5
http://www.engadget.com/2012/11/29/samsung-exynos-5-linux-benchmarks/
and A6
http://appleinsider.com/articles/12/09/19/apple_a6_soc_outperforms_intel_atom_in_javascript_test
Intel needs lower power Core i models because quad core A15's are going to start to eat into their core i design wins (e.g., MacBookAir ). Not that Apple is going to shift the iOS devices into x86.
I'm sure that Apple is probably doing a "what if" scenario with Intel testing whether the follow-on Atom alternatives would be a viable candidate in 1 or 2 generations. If their internal ARM design team stumbles badly they'll have a fall back but it likely isn't "Plan A".
In some previous reports Intel made comments as being inclined to do "custom" SoC x86 chips for customers. A custom SoC for a MBA like derivative or even a Surface Pro like clone ( Mac with detachable keyboard) is far more likely than some wholesale switch of the iOS product lines to x86.
Rumors by folks who don't look at chip power specs. The Core i chips are almost an order of magnitude off on power consumption. Intel has a product line aimed at being competitive to ARM, Atom, that is not Core i.
This is yet another one of those rumors that don't even remotely met the common sense test but land on the front page because they generate page views.
subsonix said:
A good developer won't try and make their software beyond the specs of the machine they're targetting. They're holding back because, lets face it, no one wants to use an app that lags like a bastard and crashes their iPad all the time.
Content creation apps are all about the CPU, not the GPU. Think of Zbrush. How well it runs is 90% dependent on the processor. It's having to calculate where millions of vertices are moving to whenever you run your brush across them. That takes a ton of processing power to do. Something the iPad currently isn't anywhere near capable of. And what about Photoshop? It's all about CPU and copious amounts of ram, with the GPU assisting for certain things. The iPad 4 is better at this, but I know on my 3, doing color adjustments on large shots in iPhoto can take 5-6 seconds to process. Doing the same thing on my even my slower PC would be instantaneous. I could do 10 color adjustment operations in the amount of time the iPad can do one. And what about Pages? It's alright on the iPad, but it'd benefit from more raw CPU power. I've seen it get laggy when dealing with higher res pictures.
I'm not saying the iPad sucks or anything. Far from it. But it'd benefit just as much as anything from more processing power.
No one cares about huge cpu dies that can push trillions of computations per second. It's simply not needed for normal consumer usage scenarios. All people need is to browse the web, take a picture, play some music, videos or games and check email. That's it.
Intel are part of the old school tradition heavy power usage in power scenarios when they were the only game in town. They easily could have captured the low power market with their resources, but didn't have the foresight or interest to do so until the market changed hard and far enough to take serious action.
Microsoft and Intel are the same peas in a pod, old behemoths trying to reinvent themselves. They are no longer no1 in these respective new markets. In the same way Microsoft is clinging onto ancient x86 software architecture base, Intel is clinging on to x86 hardware. Google and Apple would sooner do away with these old technologies as soon it's possibly feasible.
Intel are part of the old school tradition heavy power usage in power scenarios when they were the only game in town. They easily could have captured the low power market with their resources, but didn't have the foresight or interest to do so until the market changed hard and far enough to take serious action.
Microsoft and Intel are the same peas in a pod, old behemoths trying to reinvent themselves. They are no longer no1 in these respective new markets. In the same way Microsoft is clinging onto ancient x86 software architecture base, Intel is clinging on to x86 hardware. Google and Apple would sooner do away with these old technologies as soon it's possibly feasible.
Yeah I don't know about switching to Intel for iPads. At least not for iPads in their current form factor. I can see the convergence of iPads and Macs though and perhaps that device would use Intel processors.
In the end it doesn't matter. Software is the key.
In the end it doesn't matter. Software is the key.
I take it you are a developer and are speaking from experience. The point is, do you perceive a lack in performance or capability. Your point is obvious, the question is related to the cost/performance ratio, with the quoted ten times increase in price, is it worth it if the added power if it's just used for headroom.
Rumors on rumors...
The speed at which iOS and OS X are marching to unification indicates the amount of time that Intel has left to get its ducks in a row. Apple is going to move to a single OS, and it stands to reason, that they will only want one architecture.
Likely, they're going ARM. Apple has purchased it, and if they weren't going ARM, that's a LOT of wasted money. Intel's time will likely be up in 2014. Then all the Macs will be on ARM processors, and OS Xi (or XiOS, pronounced Zeye Oh Es).
Anyway, I may be completely wrong, and Apple may decide to buy AMD.
The speed at which iOS and OS X are marching to unification indicates the amount of time that Intel has left to get its ducks in a row. Apple is going to move to a single OS, and it stands to reason, that they will only want one architecture.
Likely, they're going ARM. Apple has purchased it, and if they weren't going ARM, that's a LOT of wasted money. Intel's time will likely be up in 2014. Then all the Macs will be on ARM processors, and OS Xi (or XiOS, pronounced Zeye Oh Es).
Anyway, I may be completely wrong, and Apple may decide to buy AMD.
No, but I can see my iPad doing so much more.
That's the big question, and I think you've got a point there. The first and second release of an Intel based iPad wouldn't take immediate advantage of the processing power unless Apple were to start what's practically a new platform from scratch. The biggest advantage with them sticking to ARM processors is that Apple can scale performance slowly over time without relatively cheaply and keep older app compatibility. It'd be a huge pain in the ass for them to switch to Intel now.
Really, I can see it happening for Apple's possible theoretical future Mac tablet, but not for the iPad.
Or not :
http://www.gsmarena.com/lenovo_k800-4445.php
I think a lot of people are quite delusional. ARM CPUs are efficient because they compromise a lot of things x86 desktop/mobile chips don't. If you make the same compromises on the x86 side, you end up with efficient chips. If you squeeze out the performance and features from an ARM based CPU, you lose the efficiency.
ARMv7 and x86 are instruction sets, not chip designs. I bet most people think x86 chips are still CISC when they have been using RISC cores for more than 10 years.
But the Surface RT is a "Windows" machine that doesn't run classic Windows apps. It was bound to run into 'rapid growth' problems coming out of the gate due to the disconnect with legacy software. The RT market so far isn't that far off of year 1 iPhone OS sales ( another new derivative OS that started out with a limited software library... namely no 3rd party apps. and only widgets.).
As long as Windows 8 , Windows RT, and Windows 7 combine to make Windows XP (and Vista) disappear Microsoft is on track to continue to be successful.
If Intel's Atom line continues to make steady, substantive improvements some RT ARM designs may switch back to Windows 8. If not then ARM will get more design wins in 2013-14 as the RT oriented software library builds.
I can imagine more experience at the 22nm process node could lead to slight improvements but this is seriously no leap to Haswell. We already have configurable TDP that can push an ULV processor down to 14W on Ivy Bridge. This feels more like a play to keep interest in Ivy Bridge down the line.
It's more of, "How do we get Intel to fab our ARM Designs if we don't put an i5 in the Large iPad".
Samsung is done fabbing for Apple 6-30-14. Big issue for iOS.
Intel could just dust off the ARM fab it once had :
Retina, OLED and IGZO have all been getting a lot of coverage, as has the iMac screen improvements. Display optimisation has been at the forefront in recent years.
Sharp's IGZO technology should slash display power consumption.
How do you figure ? Mac compatibility has nothing to do with instruction sets. The problem is frameworks. iOS exposes UIKit as a high level framework, OS X exposes AppKit. Both are quite different.
Not an Apple Design. Apple is only interested in the Fab.
The point was that Intel has experience in fabbing for ARM designs, having made ARM cpus back in the days.
These super-ultra-low TDP chips could be used for the ultimate mobile Mac: 400 to 600 g and as small as possible (ideally, pocketable) with whatever form factor (clamshell, slider or tablet). Not for heavy work on them, but being the ultimate Keynote and PowerPoint presentation tool. The Mac in your pocket. Always.
----------
I mean virtualization compatibility as now possible between Mac and Windows or Linux on x86 via VMware Fusion on Mac (run Windows on Mac), for instance.
----------
I mean virtualization compatibility as now possible between Mac and Windows or Linux on x86 via VMware Fusion on Mac (run Windows on Mac), for instance.
This AND MAYBE ...a Air with touch display and quad core (plus HD5000).Lower power requirements = retina MBA
Would be sweet.