Apple says battery life gap only 2-3 percent in TSMC, Samsung A9 chips

AppleInsider · October 8, 2015 4:04PM

Trying to quell rumors, Apple on Thursday issued a statement claiming there is little difference in the battery life of iPhones using A9 chips made by TSMC versus those made by Samsung.

Various anecdotes and Geekbench tests posted online suggested that the gap could be as much as two hours, but an Apple spokesperson told TechCrunch that the variance is small.

"Our testing and customer data show the actual battery life of the iPhone 6s and iPhone 6s Plus, even taking into account variable component differences, vary within just 2-3% of each other," the person said.

The representative blasted "certain manufactured lab tests" as "unrealistic," since they force processors to run at high CPU usage until a battery is drained.

"It's a misleading way to measure real-world battery life," Apple suggested.

Accounts have generally pointed to TSMC-based iPhones lasting longer. The supplier uses a 16-nanometer FinFET process to build its version of the A9, whereas Samsung is believed to use 14-nanometer technology. Energy efficiency normally increases as an inverse function of semiconductor fabrication size.

Apple's dependence on two different A9 manufacturers is likely a result of needing enough chips to meet demand. The strategy may also be a way of forcing price competition, and having a fallback in case of production problems.

suddenly newton · October 8, 2015 4:08PM

Nanometergate.

jfc1138 · October 8, 2015 4:19PM

Quote:

Originally Posted by Suddenly Newton

Nanometergate.

Meaning, really, REALLY, tiny gate.

mstone · October 8, 2015 4:27PM

Quote:

Originally Posted by AppleInsider

Energy efficiency normally increases as an inverse function of semiconductor fabrication size.

I was under the impression that the opposite was true. So smaller die size results in worse energy efficiency?

seankill · October 8, 2015 4:30PM

Quote:

Originally Posted by mstone

I was under the impression that the opposite was true. So smaller die size results in worse energy efficiency?

The article and you are right (your impression, that smaller consumes less energy). As the die goes down in size, the energy efficiency goes up; thus, an inverse relationship.

I hope TSMC can produce enough A10s for the iPhone 7.

jfc1138 · October 8, 2015 4:36PM

Quote:

Originally Posted by mstone

I was under the impression that the opposite was true. So smaller die size results in worse energy efficiency?

Were they both to INCREASE together then 'the opposite would be true'. efficiency INCREASE: size DECREASE: INVERSE.

And they don't and it isn't.

chadbag · October 8, 2015 4:36PM

Quote:

Originally Posted by mstone

I was under the impression that the opposite was true. So smaller die size results in worse energy efficiency?

NO, smaller die size does not result in worse efficiency. Die size is not the only parameter -- process, or how they make the chips, also matters. Die size on the same process should result in more efficient chips, but dies size across 2 different processes does not mean much.

ireland · October 8, 2015 4:47PM

#FWPgate

thewhitefalcon · October 8, 2015 4:48PM

mstone wrote: »

I was under the impression that the opposite was true. So smaller die size results in worse energy efficiency?

Not normally. It just means Samsung's 14nm process isn't as good as TSMC's 16nm process.

rob bonner · October 8, 2015 4:52PM

Anyone know how to determine the source of a chip in a device?

wood1208 · October 8, 2015 5:00PM

Those who are skeptical, iPhone 6s battery life may vary slightly or more depending on who uses, how long and for what tasks.

netmage · October 8, 2015 5:04PM

rob bonner wrote: »

Anyone know how to determine the source of a chip in a device?

I believe you would need to x-ray the device.

kpom · October 8, 2015 5:14PM

Quote:

Originally Posted by Rob Bonner

Anyone know how to determine the source of a chip in a device?

There were some untrusted apps that purported to be able to tell the difference.

aplnub · October 8, 2015 5:16PM

Quote:

Originally Posted by Rob Bonner

Anyone know how to determine the source of a chip in a device?

MacRumors posted link to an (unverified) app that will tell you. Another may have popped up but I have not kept up with it. I am not that curious at this point.

wizard69 · October 8, 2015 5:19PM

mstone wrote: »

I was under the impression that the opposite was true. So smaller die size results in worse energy efficiency?

This isnt a question that is straight forward to answer. What happened a few nodes ago was that leakage increased significantly so chip makers had to make significant changes to processes. Effectively the new process are not ditectly compatable to the old processes.

Now we have this new technology in the teen processes that is dramatically different yet again. FinFET is just one difference that again makes direct comparison between nides impossible. In fact there is good reason to believe s direct shrink was possible. At least not without a power payoff swamped by leakage.

aldobaie · October 8, 2015 5:19PM

A very possible reason is heat; As smaller chips generate more heat and thus consume more power.

rogifan · October 8, 2015 5:20PM

Part of me thinks Apple should not have responded to this. Apple's response is going to give it even more attention, attention it probably wouldn't otherwise have received. This wasn't really a "gate" yet and was mostly just floating around tech/rumor text sites. Now Apple has legitimized it as a story and I won't be surprised if some idiots who ran an app to find out what processor they have will run to an Apple store demanding a phone with a TSMC chip. And I won't be surprised if some lawyer is out there just waiting to file a class action lawsuit claiming Apple lied about battery life. :rolleyes:

webweasel · October 8, 2015 5:28PM

All things being equal, a smaller process will produce a more efficient chip but I think Apple are saying the test is flawed because it maxes out the power to see which chip runs exhausts the battery first. In this case the larger TSMC chip is better at dissipating heat so it runs out of power faster than the Samsung which is throttled by it's relatively smaller size. It is not a scenario that is likely to be seen in real life.

9secondko · October 8, 2015 5:33PM

Normally the smaller process would be more energy efficient. It just seems that Taiwans process at 16nm is that much better than Samsung at 14.

Score one for TSMC.

Good news actually.

9secondko · October 8, 2015 5:35PM

webweasel wrote: »

All things being equal, a smaller process will produce a more efficient chip but I think Apple are saying the test is flawed because it maxes out the power to see which chip runs exhausts the battery first. In this case the larger TSMC chip is better at dissipating heat so it runs out of power faster than the Samsung which is throttled by it's relatively smaller size. It is not a scenario that is likely to be seen in real life.

It wasn't heat dissipation. It was the fact that no one is using 100% cpu all day.

So of course it will exaggerate the variance.

But still interesting insight into the superiority of TSMC design.

An interesting read:

http://www.fool.com/investing/general/2015/05/18/heres-how-samsungs-14-nanometer-transistors-compar.aspx

boltsfan17 · October 8, 2015 5:58PM

Quote:

Originally Posted by Rob Bonner

Anyone know how to determine the source of a chip in a device?

You can find out with this app:

https://itunes.apple.com/us/app/lirum-device-info-lite-system/id591660734?mt=8&ign-mpt=uo=4

After you install the app:

1) Look under model in the main page

2) If model is N66AP or N71AP, sorry, you’ve got a Samsung chip. If model is N66MAP or N71MAP you’ve got TSMC.

EDIT: looks like the app has disappeared off the app store. Here is link to other way to find out. Use at your own risk:

http://demo.hiraku.tw/CPUIdentifier/

Apple says battery life gap only 2-3 percent in TSMC, Samsung A9 chips

Comments