I
I think he is wrong. Identifying the SoC with the device itself is a huge mistake (or a beginner's analysis).
While the iPad Pro might not be a killer tablet (in terms of sales), the A9X is surely a monster chip.
Yes, probably the iPad is too big, too expensive and the iOS not at a pro level yet, but the SoC is gorgeous (you can't make it throttling, even under heavy load apps or games, and that's truly impressive).
If we consider the performance per watt per dollar, this guy is miles away from anything Intel can currently offer.
Trying to compare the performance of the A9X with a chip that drains over 15W and requires active cooling, or even worse with chips sold at >280$ is either a mistake or a lie (pathetic I would say).
While the iPad Pro might not be a killer tablet (in terms of sales), the A9X is surely a monster chip.
Yes, probably the iPad is too big, too expensive and the iOS not at a pro level yet, but the SoC is gorgeous (you can't make it throttling, even under heavy load apps or games, and that's truly impressive).
If we consider the performance per watt per dollar, this guy is miles away from anything Intel can currently offer.
Trying to compare the performance of the A9X with a chip that drains over 15W and requires active cooling, or even worse with chips sold at >280$ is either a mistake or a lie (pathetic I would say).
Three things:
- The author has zero semiconductor experience
- The author has zero microprocessor experience
- The author does not even own an iPad Pro
HAhahahahahahahaha...................
This is the one of the reasons why I created SemiWiki, I got seriously tired of people writing about things that they have zero experience with. And did this guy imply that the A9x will not work in a smaller tablet? What chip do you think Apple will use in the next iPad (air 3)?
And in my opinion Apple will not replace the Macbook CPU with an SoC. Apple would do better to create an SoC based iOSbook.
P.S. I brought my iPad Pro with me on the flight to Taiwan and they almost made me check it at the gate.....
Last edited:
I think he makes a points toward a reasonable case why they wouldn't use the same chip in the Air 3...
Based on data he has cut and pasted without verifying. I have an iPad Pro with 32GB and WiFi. I used it for about 12 hours on the flight to Taipei with inflight WiFi (my movie screen was broken!) and landed with 40% left. So I'm guessing about 20 hours of WiFi use. What do the other folks say? An iPad Pro with a modem would last probably half that?
And if not the A9x in the iPad Air 3 then what? The a9?
I am curious to see where things go. We have the Intel Core-M chips operating in the 5W TPD, compared to the i5/i7's at 15W TPD. Eliminating moving parts such as fans and openings that can collect debris is a pretty big thing for electronics. Plus, we are always struggling with battery life. The new MacBook, has batteries shaped to match the contour of the shell. I know Dan likes his iPad Pro and I'm a big booster of the 12 inch MacBook. But the game is not just horsepower, it is how well the software and hardware work together to adapt to the workload. There are lots of hooks in the software these days to scale frequency and turn blocks on and off as needed. This goes to the heart of the whole system integration. iOS is really well thought out in this regard. You know Apple sweated the details here. MacOS is not far behind, but certainly does not impose the kinds of restrictions on Apps that iOS does to manage power.
Having the ability to burst to high clock rates is really important for giving the user a better experience. But keeping within the power envelope of the chip and its heat sink regime is essential.
As for what Apple will do next, it's hard to say. But they are not going to stop designing processors anytime soon. However, this is the company that had their main competitor in the cell phone market building their chips. Pragmatism will win out.
Having the ability to burst to high clock rates is really important for giving the user a better experience. But keeping within the power envelope of the chip and its heat sink regime is essential.
As for what Apple will do next, it's hard to say. But they are not going to stop designing processors anytime soon. However, this is the company that had their main competitor in the cell phone market building their chips. Pragmatism will win out.
Don Dingee
Moderator
So, there was some debate on whether the A9X had 2 cores or 3. AnandTech just published their first look based on Chipworks data, it is apparently 2x Twister at 2.26 GHz. They also indicate it is huge - 147mm2. With the bump to LPDDR4 they have dropped L3 cache, and it has a much improved GPU.
More on Apple?s A9X SoC: 147mm2@TSMC, 12 GPU Cores, No L3 Cache
I really don't like these "Intel vs Apple" angles. The Apple parts are captive, and there is zero chance Intel will ever be in an iPad. If one is comparing GeekBench scores, it's all synthetic. Are we trying to compare an iPad Pro to a Surface Pro 4? iOS 9 vs Windows 10? It makes very little sense, those are two different camps.
More on Apple?s A9X SoC: 147mm2@TSMC, 12 GPU Cores, No L3 Cache
I really don't like these "Intel vs Apple" angles. The Apple parts are captive, and there is zero chance Intel will ever be in an iPad. If one is comparing GeekBench scores, it's all synthetic. Are we trying to compare an iPad Pro to a Surface Pro 4? iOS 9 vs Windows 10? It makes very little sense, those are two different camps.
During my trip to Taiwan this week my iPad Pro was quite the novelty. People even asked me about at the airport. Nobody at TSMC that I met had one either. Everybody ooo and aaaa'd at the style and commented on the thinness and light weight. They also commented on how big it was, and not in a good way. Great for inflight movies! The video stream is better than my Laptop.
If Apple does a similar air implementation it will be a winner for sure. My opinion.
If Apple does a similar air implementation it will be a winner for sure. My opinion.
Bernard Murphy
Moderator
Hardware is one thing, but if Apple wants to compete in any sense with Wintel on laptops, they have to get a lot more serious about enterprise-grade release-to-release compatibility in the OS. If you have a MacBook running professional-level software you know you put off OS upgrades for as long as you can get away with it. My wife runs AutoCAD on her Mac and has ground to a halt several times after upgrades because AutoCAD doesn't work on the latest OS release (when she then has to wait for weeks to get them playing nice again). AutoCAD problem? Maybe, but this rarely if ever happens on Windows. Also stability problems for MS Office on Windows. MS problem? Who cares? As a consumer I want software that works reliably. If Office doesn't work well on the Apple platform, that's an Apple problem. As a consumer I am uninterested in parsing underlying technical reasons.
Then there's the less than stellar file management interface but that's a different problem. In short, Apple is very good at consumer mass-market support and they are very weak at professional support. Fancy hardware is not going to make a dent in that problem.
Then there's the less than stellar file management interface but that's a different problem. In short, Apple is very good at consumer mass-market support and they are very weak at professional support. Fancy hardware is not going to make a dent in that problem.
Bernard Murphy
Moderator
Clarification - when I mentioned file management, I'm thinking of OS X rather than IoS. But the IoS concept of hiding the file interface also isn't going to work for professional environments. Another challenge for Apple
Brianhayes
New member
A10 / A10x surely. That is what I am waiting for.
Speedfriend
Member
I think the fact that no iPad Air 3 was launched now with an A9X is quite telling, just like no iPad Mini was launched with the A8X. iPad Pro will be a low volume seller, so it can't be availability of the A9X that is a constraint. There would be a lot of corporate demand fro the iPad Pro features in a smaller body.
Brianhayes
New member
I wholly agree with Bernard's sentiments as to what Apple needs to do but speaking as one who is looking forward to the day when he can ceremoniously hurl, glass of champagne in hand, his Wintel machine into the Council skip, I hope that Apple is not as far away as he seems to feel. Having done some self taught web design for a small business, which made me detest Internet Explorer, I also had to acquire an iPad and the more familiar I became with iOS the more I realised how flaky Windows is. Without the services of an IT department to keep my machine clean, and without enthusiasm to spend money, effort, and time to do this myself I just want to rid myself of it as soon as I can.
Obviously I have not bought a Mac and the iPad pro is probably not there yet. The title of this thread, however, is the question whether the chip is a big deal and the related article, IMO, simply does not ask a relevant question. We are 9 months away from the A10. When we get to the A umpteen competing with the NextBrandLake, not to mention Snapdragons and Exynoses., I am in no doubt that every chip will be capable of running the vast majority of consumer Apps with ample capacity to spare, with the possible exception of those that require the x86 instruction set to operate to best advantage. In short no chip is a big deal. The OS and the Apps differentiate.
I can quite see how consultants at IBM, some of whom used to be colleagues of mine, have come up with numbers that suggest that Macs have a lower TCO by a few hundred dollars, all of which I believe is due to IT dept costs, without which I, for one, could not recommend Windows to anyone. Since buying an iPad I have lost count of the number of times they have upgraded iOS for me.
Cook, I believe, has declared his hand. It may be wishful thinking but I expect he will win. We just have to wait and see.
Bernard Murphy
Moderator
Hi Brian - in fairness to Apple, I should add that we are a (very nearly) all Apple household (one remaining PC in a sea of iMacs, iPads, iPhones and MacBooks), and largely I'm happy with this. But I'm not trying to run a business where I would have to balance OS upgrades with 3rd-party software upgrades. I hear this works quite well in the web/marketing/image creation/editing domains, less so in the engineering/IT and finance/accounting domains. I'm hoping Tim Cook can figure out how to get software vendors to move more in lock-step with OS X and iOS releases, though this seems challenging to Apple's historical indifference to products developed outside Apple. Of course an effective monopoly (as you suggest) might change that.