Apple A9 SoC

[iroboto]

Yes except your limited by ios which means it will need new apps devoted to it and more so to its $100 pen

I'm taking classes at the local community college and I've started to see the surface pro 3 pop up every where and a lot of people never touch the keyboard and only use the pen or touch screen.

As for being locked to IOS. Yes, that's why I made the comment that developers would make or break the iPad pro. But from what I see of the software presented, it's improved, it's matured - developers are finally figuring out touchscreen as a complete input device.

Keyboards and Stylus has been around on iPad since iPad 2. Apple never got into the game until iPad pro. Rightly so, the resolution of the iPad was never good enough really support a proper stylus until now.

As for people using touch screen and pen at the same time, yes, it's effective. Some people can go with just stylus or finger though, but I'm sure having both together will likely result in more efficiency.

 

[Turbotab]

They kept the 16GB flash starting configuration even with the higher resolution cameras and 4k video so 1GB of RAM would be very disappointing but sadly not surprising.

IRC, iPhone 6 had 2 PoP RAM chips, so they used 4Gb lpddr3. Samsung has had 8Gb LPDDR4 chips in mass production since late 2014, so they could perhaps use just 1 chip for 1GB of RAM, if weight / thickness reduction were a massive concern, over the web performance gains of 2GB RAM, which was one of the most lamented aspects of the 6. LPDDR4 is more power efficient than its LPDDR3, so moving to 2GB should not radically increase power consumption, in fact for heavy web users the regular reloading of tabs might be worse for battery life, than a slightly higher RAM power usage.

The 16GB of nand IMO is just an oldschool marketing ploy, beloved by prestige German car makers, offer a spartan base model, to make the more expensive 64GB look like better value.

 

[anexanhume]

Will be interesting to see how A9X has 2x memory bandwidth. They were already 1600 MHz LPDDR3 128-bit on A8X.

 

[wco81]

Also faster speed to flash storage.

Well the iPad Pro doesn't record 4K video like the new iPhones though. Nor capture Live Photos.

 

[Grall]

I've started to see the surface pro 3 pop up every where

Confirmation bias...

These things aren't "everywhere"; the Surfaces are a niche product, which is why MS is throwing their weight behind iOS app development. If MS's hardware sales were awesome, they wouldn't do that. It's not their standard modus operandi to support competitors' ecosystems unless they have to. (It isn't Apple's either one might observe.)

 

[anexanhume]

Also faster speed to flash storage.

Well the iPad Pro doesn't record 4K video like the new iPhones though. Nor capture Live Photos.

Apple is doing people a favor by not encouraging them to capture video with a 12.9" device

Faster Flash isn't that implausible, though. There are faster flash types out there and they bought their own flash memory controller maker (Anobit).

 

[eastmen]

Confirmation bias...

These things aren't "everywhere"; the Surfaces are a niche product, which is why MS is throwing their weight behind iOS app development. If MS's hardware sales were awesome, they wouldn't do that. It's not their standard modus operandi to support competitors' ecosystems unless they have to. (It isn't Apple's either one might observe.)

you should have quoted my whole point being college campus. They are very nicely priced for students you can get the pro line for $650 without a keyboard at this point in time.

Surface sales keep going up , they just released a new one 3 months ago to replace the RT line and have the surface pro 4 line coming out next month.

MS has been on OSX since its inception so it hardly matters. The real question for MS will be how the unified apps work for them . If it works well then it will be easier for android and ios users to move back over.

 

[Entropy]

Will be interesting to see how A9X has 2x memory bandwidth. They were already 1600 MHz LPDDR3 128-bit on A8X.

In all likelyhood they use LPDDR4 at the same bus width and twice the data rate.
I find the quoted increases in CPU performance to be remarkable, almost implying increases in number of cores, even though that hypothesis is an awkward fit for the A9x increase of 80% vs. the A9 of 70%. No mention as far as I could see about increased bandwidth for the A9.

 

[anexanhume]

In all likelyhood they use LPDDR4 at the same bus width and twice the data rate.
I find the quoted increases in CPU performance to be remarkable, almost implying increases in number of cores, even though that hypothesis is an awkward fit for the A9x increase of 80% vs. the A9 of 70%. No mention as far as I could see about increased bandwidth for the A9.

Yeah I was conflating MHz with MT/s. LPDDR4 is twice. They also made no mention of memory speed for A9. For 1.7x vs. 1.8x, I'm guessing A9X just got a more aggressive clock given laxer TDP.

 

[mczak]

Yeah I was conflating MHz with MT/s. LPDDR4 is twice. They also made no mention of memory speed for A9. For 1.7x vs. 1.8x, I'm guessing A9X just got a more aggressive clock given laxer TDP.

I think Entropy was refering to the fact that a core count increase would represent a quite a bit larger step for the A9 vs. A9X. So, if the A9 had 3 cores instead of 2, that would already be a 50% improvement, leaving only ~5% for IPC improvements + ~7% for higher clocks or so to get to the 70% vs A8. But for the A9X, a core count increase to 4 cores would only represent a 33% improvement, so with the same 5% IPC improvements it would need ~30% higher clock to get the 80% against A9.
Who knows though what metric those 70%/80% increases refer too. It is also possible the chips have quite a bit higher peak clock but need to throttle more than they had to in the past, so using the right workload just about any quoted percentage increase would sound reasonable for both of them, with or without more cores for either of them...
GPU is pretty interesting too, series 7XT always seemed for sure, but can they really get a 90% increase with the 4-cluster GT7400 vs. GX6450? Even PowerVR is saying "up to 60%" improvement vs. 6XT so for some "real" 90% improvement that would probably mean something like 50% higher clocks... So maybe it's a GT7600 after all (and god knows what for the A9X).

 

[Entropy]

Edit: mczak got my point and even did the same calculations while I fell asleep writing my almost identical response. Can't seem to delete my post now.

 

[Entropy]

So far, Apples generational performance multipliers have been surprisingly valid, in spite of their appearance in marketing presentations. FinFET processes should yield decent improvements in power/frequency over TSMC 20SoC, but 70-80% on the other hand seems unlikely witin similar power envelopes. So I'd speculate that we are either looking at numbers derived from great improvements in IPC (20-35%) and frequency (30-40%) or an A9 with 3 cores and a 20% frequency boost or so and an A9x with a frequency boost of 35-40%. I see the latter scenario as the more likely, probably with some minor IPC enhancements a la mczac above.

The GPU claims could possibly be justified by GT7400 and GT7800 respectively, along with some fairly substantial increases in clock. Alternatively GT7600 and GT7600x2 could fit the bill, with the GPU gains of the A9 in either scenario being somewhat constrained by bandwidth not increasing with the increased capabilities of the processor.

Either way, those are some quite substantial numbers Phil Schiller showed on his slides. This will be interesting to see unravelled.

 

[Nebuchadnezzar]

GT7600 makes most sense and while that die shot can't be trusted it did show 6 GPU cores. I doubt they'd let marketing troll the public like that.

I'm more interested in seeing actual CPU numbers.

 

[Entropy]

GT7600 makes most sense and while that die shot can't be trusted it did show 6 GPU cores. I doubt they'd let marketing troll the public like that.

I'm more interested in seeing actual CPU numbers.

Oh, I'm sure the major review sites have gotten their hands on samples already. (* cough *) ;-)
Apparently the iPad Pro has 4GB RAM as leaked by Adobe .

Seriously, I find it a bit of a shame that Apple doesn't let their engineers show off their work even in footnotes on spec sheets. They do great work, and even though the nitty gritty may not belong in marketing presentations, I feel that details of their work should be on display somewhere.

 

[shiznit]

http://bgr.com/2015/09/10/iphone-6s-ipad-pro-ram-memory/

Other sites are claiming 2GB for the phones. Finally! (ram memory....)

 

[anexanhume]

http://bgr.com/2015/09/10/iphone-6s-ipad-pro-ram-memory/

Other sites are claiming 2GB for the phones. Finally! (ram memory....)

A lot of the sites claiming 2GB for phones are referencing a Mobile Syrup article that simply lists it in the specs without any qualifiers of how they got that number, so I wouldn't trust any of them.

However... notice how when they had the Warhammer devs come up and list the cool new stuff they were able to do with the chip, the first thing they talked about was "high resolution textures"? Hmmm.

 

[Nebuchadnezzar]

Oh, I'm sure the major review sites have gotten their hands on samples already. (* cough *) ;-)

Not really, it's still a bit early for that.

 

[wco81]

Well Reddit and blog posts from supposed Apple employees are saying 2 GB of RAM in the A9 of the iPhone 6S and 6S Plus and 4 GB of RAM in the A9X of the iPad Pro:

http://www.gsmarena.com/iphone_6s_and_iphone_6s_plus_feature_2gb_of_ram-news-13948.php

I guess they would need a lot of RAM to load that Autocad model they demo'd, which they said had 360k objects?

Thing is, most iPad Pro users won't be using Autocad on the thing. They will likely watch videos. Maybe in a couple of years they can put a true 4K screen and you can run Netflix and Amazon Video 4K streams.

 

[Entropy]

Not really, it's still a bit early for that.

Gathered as much, given your post. I do hope samples are distributed ASAP. And I'd like to see Chipworks do a die shot and analysis.
It will probably take a while before we can know if Apple is actually dual sourcing the A9.

 

[ltcommander.data]

Not really, it's still a bit early for that.

Maybe not heavy technical sites sadly, but I thought major general news outlets like WSJ and NYT along some general tech sites like The Verge and friends like Gruber get handed review units at the end of the event so they can get a week's usage with it and have their reviews up before it's available to customers? They rarely run benchmarks, much less in-depth technical analysis, which might be a combination of their own interests and skill set and perhaps agreement, tacit or otherwise, with Apple?