He obviously means lower power Core.It does, and it does quite a bit. Low power x86: Atom
Yes.So is it a 20nm chip, or what?
Because it likely doesn't. If those numbers you say do turn out to be right, it's probably because the design is optimized for higher density. Also note that Intel's transistor count is schematic; if you count the transistors on a Core M die, it will be more than 1.3B.how is it they doubled the number of transistors but the performance hasnt gone up by all that much?
also how is it that they have higher density in it that core m???
1.3 / 82mm^2 vs 2 / 87 mm^2?
Because it likely doesn't. If those numbers you say do turn out to be right, it's probably because the design is optimized for higher density. Also note that Intel's transistor count is schematic; if you count the transistors on a Core M die, it will be more than 1.3B.
SPEC2K is not very representative, except its gcc component which pushed branch prediction and slightly stresses the instruction cache. But I agree, it's better than Geekbench.Whereas soon as you switch to SPECInt 2K (low level, but meant to reflect a variety of actual integer algorithms and hence somewhat representative of actual performance) the Intel performance relative to ARM almost doubles.
Cross platform Javascipt comparisons are hard for a very simple reason: you can have vastly different results depending on the browser you use. At least Geekbench doesn't rely on external software at runtime. But I wouldn't say one or the other is better, both are interesting.You can easily see the same thing happening in that chart with a comparison between Geekbench 3 and Google Octane v2.0. While there are plenty of flaws in javascript benchmarks I don't understand why Geekbench is any better
Die size is 89mm² compared to 102mm² for the A7 according to AT. The density looks good.Also, A8 has 2B transistors. I wonder what the die size is.
i dont understand. they did say the soc had 2bln transistors and was 17% smaller or something than the a7.
is the transistor count not apples to apples?
As mentioned, the #1 problematic use memory wise is a core app written by Apple - Safari.
It seems esp. bad with 64-bit iDevices, presumably because 64-bit apps and OS require more memory.
Die size is 89mm² compared to 102mm² for the A7 according to AT. The density looks good.
I don't even know what that means. What is the OS limitation?Safari isn't using 1GB of memory. That's an OS limitation.
Only if you can run Geekbench or SPECint on it.Also what about the processor in the Apple watch? The S1? Does that count as a CPU and should it have a separate thread?
If those numbers are true, that would imply that 20nm demolishes 14nm in density.
So Intel sacrificed traded density for power consumption with some designs in order to break into markets they haven't had much success in: ultra-low power mobile. There's nothing particularly controversial about that assumption.
Even more annoying is Apple announcing its things at the same time as IDF, quite immature of them.
LOL wut?
The last several iPhone announcements have been the second Tues or Wed in September. I doubt they gave IDF a single thought - frankly if Intel wanted solo attention, they need to move now, not Apple.
http://newsroom.intel.com/docs/DOC-1186
Apple knew Intel has IDF at the same time, they decided to take the same date.
I wonder if they load one ship with 50 million iPhones and just send it to the US. Or would putting a couple billion dollars on one boat be a bad idea? What if it sank? No iPhones until February?
Only if you can run Geekbench or SPECint on it.
There is no way this would happen, one ship. Remember that 1.5 million dollars of ipads were stolen at jfk airport (they were coming from china via plane). 3600 ipads minis.