[Ailuros]

Well I wasn't as nuts as I myself thought I am with my 'funky" ideas. Anand is correct in hinting that the timespan for a custom design must have been tight; I didn't myself expect it as early. Apple's engineers must have been chasing their heels for a long time now in order to get it done.

Now I'm of course dead curious what they've exactly done, but since they have their own OS they'd be extremely dumb if the processor wasn't tailored to fit the sw platform's actual needs like a glove.

[Laurent06]

If tbat's their chip, it's definitely an impressive achievement. They bought PA semi about 4 years ago, so they had some experienced designers and probably usable functional blocks. But even then implementing and validating an ARMv7 core is certainly not an easy task.

[wco81]

Maybe the real achievement will be a device which has leading performance, battery life and thinness compared to other A15 or Krait-based devices.

Some other manufacturers may stuff more power-hungry SOCs into thicker, heavier enclosures along with much bigger batteries to get decent battery life.

[OlegSH]

Originally Posted by wco81

Maybe the real achievement will be a device which has leading performance, battery life and thinness compared to other A15 or Krait-based devices.

Some other manufacturers may stuff more power-hungry SOCs into thicker, heavier enclosures along with much bigger batteries to get decent battery life.

Only in fanboys wet dreams. Krait isn't something special on single thread workloads in comparison with A9, I am highly doubt that Apple will be able to beat ARM on ARM's own field anytime soon

[ams]

I think the reason that Apple is not making much of a fuss about a "custom" CPU design on iphone 5 is that, at the end of the day, 2x CPU performance vs. iphone 4s (with 800MHz dual core A9 CPU) is really nothing much to brag about. The overall CPU performance of commercial devices with dual-core Krait or quad-core Tegra is arguably better in comparison to "2x" iphone 4s CPU. Now, regarding benefits of custom CPU design with respect to power consumption, there are far too many variables involved to know for sure what the impact would be. Power consumption is a function of both hardware (including architectural and fabrication process improvements) and software.

[ams]

By the way, with respect to the relatively light weight of the iphone 5, that is partly due to the new in-cell screen technology, and also partly due to the extensive use of aluminum on the chassis. I'm sure we will see more of the same from other companies moving forward. Just as the auto industry is focusing on reducing weight and improving fuel consumption, the mobile industry is focusing on reducing weight and improving power consumption.

[wco81]

What about the TDP allowed by the SOC?

[RudeCurve]

The weight reduction is pretty simple really....

In terms of volumetric area both Stainless Steel and Glass is heavier than Aluminum. Since the chassis is aluminum now and there's no glass back and the front glass is thinner it's logical that it was made lighter even if the battery slightly increased in size.

The next step will probably be carbon fiber composite unibody.

[ltcommander.data]

http://www.anandtech.com/show/6297/i...1gb-lpddr21066

1GB of LPDDR2-1066 in a 2x32-bit configuration confirmed.

[ams]

Hitler finds out about iphone 5: http://www.youtube.com/watch?v=N5AHEd5kbT0 (this video is absolutely hysterical, but please don't even think about shorting Apple right now because it would be incredibly foolish!).

Samsung response to iphone 5: http://www.dailytech.com/article.aspx?newsid=27693

[Ailuros]

Typical marketing which from Samsung's perspective is completely reasonable and nothing I'd personally oppose to. At another level its Samsung that lacks content and the idiotically high profit margins Apple usually enjoys.

That shouldn't mean that things could not change in the future but if then certainly the less foreseeable one.

[RudeCurve]

Big and ugly.

[fehu]

Originally Posted by RudeCurve

Big and ugly.

+1
but it's clearly smaller than the galaxy

[french toast]

Well the iPhone 5 is shaping up to be a special little toy indeed...you bet the efficiency of that software/bespoke A6 processor will be bloody ridiculous...I wonder whether they have increased the battery size?

It would disappointing if not, bizarrely someone stated that the lumia 920 has higher quoted usage times.. you would think that would not be the case..mmm.

I can't wait till we get benchmarks and batterylife consumption

[Pressure]

Originally Posted by french toast

Well the iPhone 5 is shaping up to be a special little toy indeed...you bet the efficiency of that software/bespoke A6 processor will be bloody ridiculous...I wonder whether they have increased the battery size?

It would disappointing if not, bizarrely someone stated that the lumia 920 has higher quoted usage times.. you would think that would not be the case..mmm.

I can't wait till we get benchmarks and batterylife consumption

They are using new chemistry and have increased the LiPo battery to 3.8 V, so it has increased capacity.

Nothing new though, since Motorola have been doing it a year now but every bit counts

[aaronspink]

Originally Posted by OlegSH

Only in fanboys wet dreams. Krait isn't something special on single thread workloads in comparison with A9, I am highly doubt that Apple will be able to beat ARM on ARM's own field anytime soon

Eh? Not like it hasn't happened before (StrongARM) nor that a large number of apple designers come from the same lineage as the designers that did StrongARM.

[aaronspink]

Originally Posted by french toast

It would disappointing if not, bizarrely someone stated that the lumia 920 has higher quoted usage times.. you would think that would not be the case..mmm.

Considering that the 920 has the highest volume to screensize ratio and the highest weight to screensize ratio, you would hope they could fit a big battery.

[Rys]

Originally Posted by fehu

+1
but it's clearly smaller than the galaxy

I'd hope so given it's an iPhone 4 in that pic, not a 5.

[french toast]

Originally Posted by aaronspink

Considering that the 920 has the highest volume to screensize ratio and the highest weight to screensize ratio, you would hope they could fit a big battery.

I'm not talking about lumia size...I'm talking about SOC efficiency ...lumia only has 2000 mha battery... also has much bigger and higher resolution screen...as well as likely higher clocked cpu, with apple having custom built the chip from the ground up perfectly tailored to the now long in the tooth and heavily optimised ios, you would think it would smoke the lumia....I expect it will.

[wishiknew]

Did that Samsung ad list 1 gig of ram for the iphone 5 before the Anandtech article?

[wco81]

Yeah they may have been using some inside info about the A6.

S3 has better standby but what about actual use time? Apple claims 8 hours LTE browsing time?

[ltcommander.data]

Originally Posted by wishiknew

Did that Samsung ad list 1 gig of ram for the iphone 5 before the Anandtech article?

Anandtech found out the size by recognizing a Samsung DRAM part number on the SoC picture Apple provided. It's not surprising that Samsung would recognize their own part number a lot quicker. It didn't have to be the foundry division leaking information to the rest of Samsung, especially since Samsung has been pretty adamant about keeping their component business isolated from Apple/Samsung smartphone politics to keep that component business relationship viable as long as possible.

[Nebuchadnezzar]

Originally Posted by wco81

Yeah they may have been using some inside info about the A6.

S3 has better standby but what about actual use time? Apple claims 8 hours LTE browsing time?

It wouldn't surprise me that it actually will be much superior to the S3; power management on the 4412 in the S3 is horrible; they had claimed per-core DVFS, clock-gating and power-gating, but in reality the chip still is running on a single frequency plane for all the cores, and the only idle power management feature active right now is WFI (clock-gating) per core, which on a 4-core chip, is not enough. Together with the power-hungry AMOLED screen, it doesn't take much for the iPhone to be able to beat it, regardless of the battery size disparity.

[french toast]

Originally Posted by Nebuchadnezzar

It wouldn't surprise me that it actually will be much superior to the S3; power management on the 4412 in the S3 is horrible; they had claimed per-core DVFS, clock-gating and power-gating, but in reality the chip still is running on a single frequency plane for all the cores, and the only idle power management feature active right now is WFI (clock-gating) per core, which on a 4-core chip, is not enough. Together with the power-hungry AMOLED screen, it doesn't take much for the iPhone to be able to beat it, regardless of the battery size disparity.

Well correct me if I'm wrong but that statement is false...my s3 runs on at least 2 voltage planes.

The power management on it is also extremely good, batterylife is superb..especially considering the luminous 4.8 inch hd screen.

Also power gating also has worked flawlessly from the first day I got it...able to shut down 3 cores if need be.

[Nebuchadnezzar]

Originally Posted by french toast

Well correct me if I'm wrong but that statement is false...my s3 runs on at least 2 voltage planes.

The power management on it is also extremely good, batterylife is superb..especially considering the luminous 4.8 inch hd screen.

Also power gating also has worked flawlessly from the first day I got it...able to shut down 3 cores if need be.

You are wrong, I remember you claiming you seeing it run on different frequencies per core, I don't know what program you used to monitor, you might have confused that with core load, and not frequency.

The ARM cores are fed by a single voltage rail from the external PMIC, for it to actually have per-core DVFS it either needs 4 rails on the external PMIC, or for it to have on-chip regulators. In any way the CPUs run on a single frequency plane and you can see that in /sys/devices/system/cpu/cpu0/cpufreq/affected_cpus. Hotplugging the CPUs is a limited and only rough and coarse power management feature and currently it samples every 2 seconds on wether to turn off a core or not, and hotplugging itself is very latency bound by the operating system due to kernel architectures and how it handles migration and creation/destruction of the kthreads, resulting latencies of ~80-150ms for current kernels on consumer devices, compared to actual transition times of dozens of microseconds for which the hardware is capable of.

With newer 3.5+ kernels we now have coupled cpuidle frameworks in the kernel allowing for efficient entry into power-collapse C-states, which affect in most SoC implementations the whole CPU power plane. The hardware itself is not a limitation, but the software is. Precursor to this we could have power-collapse on a core while top (screen on) was running only if that core was the only one running. On the Exynos this is called AFTR/Arm Off Top Running. I'm currently trying to port this myself onto the arch cpuidle driver, but since the system is running on a 3.0 kernel, there is a lot of backporting to be done. The current driver is also dubious in terms of the specified residency times for WFI, being about 1000 times higher than any other architecture out there, and since this differs greatly to the values in the updated driver for the 5250, I think it is a big overlooked mistake in the kernel code.

Did you not ever ask yourself why the device runs so hot in screen-on idle? 40°C CPU temperature at 200MHz while supposedly also clock-gated most of the time, nevermind even hotplugged other cores, is not normal. The phone would have much better screen on battery life if it's power management would be optimized.