Here is the deal though. Apple is already having to pump out what, 80 million+ A11 chips for the iPhone? Do you think they could make another 10-20 million for the iPad? The iPhone is already supply constrained as it is.
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.
My beef with A10X
- Thread starter macduke
- Start date
- Sort by reaction score
Here is the deal though. Apple is already having to pump out what, 80 million+ A11 chips for the iPhone? Do you think they could make another 10-20 million for the iPad? The iPhone is already supply constrained as it is.
The X-class is the key here because even if the A11 is better than A10X in benchmarks, I doubt it will be better in GPU performance.
If I recall, A10 is 10% better than A9X in CPU performance but is still not an improvement in GPU performance.
If I recall, A10 is 10% better than A9X in CPU performance but is still not an improvement in GPU performance.
Some strange gripes, I must admit to thinking...
I don't want anything held back just to keep another chip in a spotlight longer, I want things to move at full bore at all times. The die size of A10X is actually conservative for an iPad, so possibly A11 could gun for even larger cores, GPUs, etc.
The iPad will likely still have faster graphics for the wider memory bus, as is usually the case, but since A10X is just 3 A10 CPU cores, there's room to change things yet again on the CPU side.
I don't want anything held back just to keep another chip in a spotlight longer, I want things to move at full bore at all times. The die size of A10X is actually conservative for an iPad, so possibly A11 could gun for even larger cores, GPUs, etc.
The iPad will likely still have faster graphics for the wider memory bus, as is usually the case, but since A10X is just 3 A10 CPU cores, there's room to change things yet again on the CPU side.
I know the original post in this thread was written over a month ago, but I'm just now reading it. I have to say: it was very well-written and made an excellent point. Thanks for sharing your thoughts with us. I think you're spot-on in your assessment!
Considering that the X series of chips are meant to be used with the Pro devices, I wonder if the delay is intentional as Apple needs to weed out any potential issues with the chip when using with the iPhones before creating their X variant for the Pro devices, which supposedly for people who cannot afford to have downtime with their pro devices
Then again, the iPhones are the ones that cannot afford to go down...
Then again, the iPhones are the ones that cannot afford to go down...
Not like the Macbook "Pro" 13 goes through Xeon like validation or has ECC memory, and the iPad would be less wont to be used in mission critical apps than the Macbook Pro. The Radeon Pros in the 15 however do go through twice the thermal testing, from the microsite.
I don't expect A10X goes through any more validation than a regular Apple chip, being GPU-enlarged, higher clocked, sometimes more core-ed, versions of the previous iPhone chip. I am however also sure that Apples regular validation standard is already high.
Anyways, this should be of interest to this thread:
The A10X, with its stunning performance, is actually conservative for iPad die size. 10nm fab with the largest 165mm2 die size would be something to see...
A11, on a more mature 10nm process, and should they choose to go big, will be a monster.
[doublepost=1499895978][/doublepost]
It actually very slightly edges it out
https://arstechnica.com/apple/2017/...-people-who-have-never-upgraded-their-tablet/
From that pattern, the next iPhone having near the GPU performance of the A10X would be pretty stunning. But we're getting used to that year over year.
Do you think Apple will release an A11 that is inferior to the A10X? I think the hue and cry of such an act will be worse then if Apple rolled out a next generation CPU that is better then the prior generation. In fact I think that's the conventional wisdom.
I very much doubt it.
Every new generation processor was better than the previous X version because they were smaller architecture with more computing power per watt. A10X is 10nm, as opposed to 14nm A10 (first time X version is on a smaller architecture)
A11 will also be 10nm, even if more refined i doubt it'll be faster than A10X.
A11X however should show an increase, but i don't think it will be as high as the difference between A9X and A10X.
I know, It was more of rhetorical type of question because why would apple release a brand new cpu that is actually slower or inferior to the prior generation.
I'm not sure why people would complain if the A11 is actually faster then the A10X, because that's the nature of computing.
I think this mostly stems from the fact that Apple has always tried to position the iPad Pro's in the sphere of computers. Having your top of the line tablet then being beat out by a smart phone in a year seems pretty... Meh. I agree with the sentiment of the post - it just seems weird having your top of the line iPad soon to be beat out by a smart phone. It would seem less strange if the chipsets were different - that way, you wouldn't really mind about the iPhones, and just compare the relative performances between the tablets.
But at the end of the day, Apple is proving itself to be one of the strongest in producing these chips, in midst of Intel and AMD.
But at the end of the day, Apple is proving itself to be one of the strongest in producing these chips, in midst of Intel and AMD.
Two-thirds of Apple's revenue comes from phones. Not only that, but the bill of materials for a similarly priced iPhone is lower than an iPad. Apple doesn't compare the performance of iPad to iPhone in marketing materials so the problem is mitigated. But in the end, the iPhone deserves Apple's greatest attention.
Intel and AMD are still doing most of the pioneer work for the rest of the industry through multiple generations of x86 processors. The bleeding edge technology lead by Intel because of the experience and collaboration with academic researchers. Apple is able to close the gap quickly because they are adopting published methodologies for extracting performance. This is no different than China building stealth fighters to match F-22. It has been done, but the step beyond that is most revealing.
I will never stop being amazed by people who think time should stand still once they buy a specific device. The A11, A11x, A12, A12x, etc. will all be better than the A10x in some way or another. As soon as the A11x comes out for the next iPad Pros, the current iPad Pros will not be the most powerful iPad Pros any more. But for the next 12-18 months, the current iPad Pros WILL be the fastest iPads being manufactured. And after the A11x comes out, they will still be screamers.
For me personally, most of the time what spurs me to upgrade is changes in form factor. I was totally going to stick with my 9.7" iPad Pro after the 10.5" was launched, but it started bugging me more and more and more that I would not have the most up to date form factor of the smaller sized Pro. When it became clear that the RAM was also doubled in the 10.5" model, it was a no brainer. But the A10x was NOT a factor. The 9.7" Pro would still have been fast as hell (and is currently being fast as hell for whoever now owns it.)
For me personally, most of the time what spurs me to upgrade is changes in form factor. I was totally going to stick with my 9.7" iPad Pro after the 10.5" was launched, but it started bugging me more and more and more that I would not have the most up to date form factor of the smaller sized Pro. When it became clear that the RAM was also doubled in the 10.5" model, it was a no brainer. But the A10x was NOT a factor. The 9.7" Pro would still have been fast as hell (and is currently being fast as hell for whoever now owns it.)
I don't see anything in iOS 11 that would drive people to buy iPad devices.
You are probably right that 10X was aimed at or built to coincide with WWDC, but thats not really a surprise.
The only people who should be miffed in my eyes are the IPP 12.9 G1, the 9.7 G1 had many better features and i would have felt short changed.
You are probably right that 10X was aimed at or built to coincide with WWDC, but thats not really a surprise.
The only people who should be miffed in my eyes are the IPP 12.9 G1, the 9.7 G1 had many better features and i would have felt short changed.
This gets the award for the most ridiculous comment I've seen on this forum in a really long time.
Thanks - i'll PM you my address for the prize. What is it by the way?
if its iOS11 - keep it.
if its iOS11 - keep it.
No, that’s not what I was saying. While possible due to the difference in cores, usually the increase in core performance makes up for that. What I was saying is that I hope it has around the same performance (I.e. not “better than the A10X”) and the A11 isn’t a major leap ahead which might be possible given the 10th anniversary of the iPhone and the fact it’s an S year. The only reason why it would be disappointing to to some is because they will have just bought a NEW $649-1229 iPad PRO and so would hope that the next iPhone, released 3 months later, wouldn’t outperform their device.
Last edited:
That is crazy. How come Intel has not invented a mobile chip that is as powerful as the A10 with the same power draw? True Apple is just closing the gap in terms of performance of Intel’s laptop class PCs. But the true magic Apple is doing here is that they are closing the gap with a mobile class chip meant for phones. That is more bleeding edge in terms of where we are going in tech than Intel.
Apple is creating chips that can be used for AR and VR in a form factor of 5 inch phones, not too mention running all the sensors and updating us in real time. Low end Intel laptops need a fan to stream Youtube clips.
Intel is the king of the hill in terms of history , and bleeding edge performance. But for so long they have tried to create a mobile chip for phones but failed. Now Apple is encroaching their territory with these same mobile chips.
So the Chinese analogy does not fly with me.
Last edited:
Intel doesn't want to compete in a segment where they can only charge $10-$40 per processor. Just as Apple doesn't want to create phones that cost $100 and live off the crumbs. Intel would rather develop general purpose processors that are more powerful and sell for $300 and up. It doesn't sound like you realize Intel sells Core M that consume 4.5W or less.
If Apple processors were so powerful and had the bandwidth to perform general computing tasks, Apple would have converted their MacBooks to A10X by now.
I agree if the apple chips were really so powerful, they would already be in the current macbooks etc.....
Intel did try but failed miserably.
What are you on about? This revisionist history is ridiculous.
Intel lost billions of dollars pushing their mobiles chip that were underpowered and no one wanted. They finally gave in the towel last year by abandoning the braxton line of mobile processors.
https://www.businessinsider.com.au/intel-kills-its-next-smartphone-and-tablet-chips-2016-5
And no - the processor in MacBooks draw way too much power (Core m) to be useful in mobiles phones. Their high turbo boost speed they would burn a mobile phone. Sure tdp is 4.5w but maximum draw greater than 10w when boosting.
It's not just a matter of putting the A's in there. They're a complete different architecture, requiring significant rewrite. Remember migration of OSX to Intel from PPC?
Regarding performance, the A-series are comparable to intel chips:
https://arstechnica.com/apple/2015/...-all-the-virtues-and-limitations-of-ios/4/#h2
If Intel was pioneering the mobile age then they would want that $10-$40 sale per processor, rather than giving it up to ARM. The smartphone industry is too big a segment too pass. They just could not do it even if they wanted to.
Wintel tried to get into mobile and they failed, and that is not because for lack of trying, they just did not have it in them.
Core M is slower than the A10X and even slower than A10/A9X. You cannot stick it on a phone not just because of the power draw, but because it will mean a slower phone.
So you're saying that you're hoping Apple releases a new CPU that is not faster then the A10X? That makes no sense at all. Why even release a new chipset at all, if there's not going to be any improvements?
It would make no sense at all if I hoped for an A11 that performed around the same as the A10. But I am not saying that, I am hoping for the month old 2017 iPad Pro processor to not be drastically outperformed by the next iPhone processor in 2 months. To end this, I better make it clear: It doesn’t mean I expect it, it doesn’t mean I want it to perform exactly the same, it doesn’t mean I don’t want to see improvements.