Sidebar Sidebar
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.

Controlled Testing Supports Apple's Claim of Nearly Equal iPhone 6s Battery Life With TSMC and Samsung Chips

Markoth said:
Pretty horrible article.

WiFi Browsing test: The CPU is asleep most of the time. Network access is incredibly slow by CPU standards. Any impact on battery life from the CPU during WiFi browsing would be negligible, as is shown here.

WebGL: GPU-based test. The GPU is an independent part, sourced from a third party. Perhaps customized somewhat knowing Apple, but the claim was that the samsung CPU was using more energy, not the GPU. This test is mostly irrelevant.

Geekbench: The only true CPU-based test isn't valid, because it isn't based on real-world usage? Interesting. Well, then I guess this whole article was a waste of time, since this is the only CPU-based test we have to go on.

GFXBench: Another GPU test. Not relevant, because the CPU is the item in question here.

What a poorly thought-out article. There's basically no data here. Geekbench is the only one that's of any real value, but it truly isn't indicative of real-world usage. Regardless, it still shows a clear difference in battery life between the two CPUs, and at the very least shows conclusive proof of this issue. There is a difference in battery life, and if whatever your doing is taxing the CPU, you WILL notice it. That's what this has proven. Real world usage is subjective. Not everyone spends their time surfing the net.
Click to expand...


I trust more Arstechnica than you since you don't seem to know what you're talking about.

The GPU is on the same die as the CPU (NOT a separate part), you know the one SAMSUNG DID.
So, if there is an issue with the process it will turn up there too. Got that.

When building the pages, the CPU works, with modern complex pages, it can work pretty substantially.
Most of those pages have 50 links in them all being transferred/built asynchronously too.
There's a reason most smart phone stutter when loading pages (but not IOS), because it is not a trivial test.

Everything else you said is just blah blah blah.
 
  • Like
Reactions: brianjsw, Duane Martin, JackANSI and 2 others
Sgt.ThroatPunch said:
I have two 6s iPhones, a Sammy and a TSMC. I'd like to do my own testing and record the results. Any tips?
Click to expand...
Return that Sammy for a refund, then sell your TSMC on eBay for a 56% profit, then buy two Sammy iP6Ss at a 28% discount.

Pocket the extra $$$ and save the $0.99 not buying Geekbench 3.
cat-playi.gif
 
  • Like
Reactions: smoking monkey, Carotdo, apolloa and 2 others
If you are an average user the impact will be minimum. If you know that your usage patterns are based on heavy cpu usage, your at a disadvantage battery wise with the Samsung chip.

Upto the individual to make the assessment.
 
  • Like
Reactions: smoking monkey and bevsb2
Nobody raised the point that Samsung phones have twice bigger batteries 3000mah and last the same time as Apple twice smaller batteries... What is the difference? Samsung chip :)

Anyway reading this article I feel like : I'm using my phone wrong...

Sorry that I can install games from apples tore, I think if gaming is not considered as normal use then all games have to be removed from the store... I can understand that cell network coverage has an impact, screen brightness, cpu... But such difference... Well even 60 mins can make a difference...

And again why all this? Because people are clearly requesting better battery life APPLE!!!! Stop making flat phones and please listen back to your customers. Thanks
 
  • Like
Reactions: Benjamin Frost, Aston441, tvbi and 1 other person
Keirasplace said:
I trust more Arstechnica than you since you don't seem to know what you're talking about.

The GPU is on the same die as the CPU (NOT a separate part), you know the one SAMSUNG DID.
So, if there is an issue with the process it will turn up there too. Got that.

When building the pages, the CPU works, with modern complex pages, it can work pretty substantially.
Most of those pages have 50 links in them all being transferred/built asynchronously too.
There's a reason most smart phone stutter when loading pages (but not IOS), because it is not a trivial test.

Everything else you said is just blah blah blah.
Click to expand...

if you are going to attack someone, Avoid blah blah blah yourself . I see you don't understand why a CPU can peak when it comes to webpages, it does not build them.....
 
MH01 said:
If you know that your usage patterns are based on heavy cpu usage, your at a disadvantage battery wise with the Samsung chip.
Click to expand...
You actually believe this??? Amazing. Since there is NO way of knowing which chip you'll get with a new phone, clearly any "heavy cpu" user should NOT purchase an iPhone.
 
SSD-GUY said:
Quick Apple PR payment to ArsTechnica and we get this.

P.S - How are people saying this isn't a problem. The one test that constantly pegs the phone, Geekbench, shows around a 30% difference between the two chips. There are times when I can use my phone at full load. If I'm commuting, sitting through something boring and play a demanding game etc.
Click to expand...

Because it isn't a problem. The variation you are seeing here isn't any different than what is seen in laptops with Intel or AMD chips in them. You can have significant differences in thermal performance depending upon where on the wafer the chip was cut from.
 
  • Like
Reactions: apolloa, dk001 and CraigGB
dstud208 said:
They should have just put 100% tsmc chips in 6S and 100% Samsung in 6S+, then there wouldn't be any real reason to suspect a difference. In reality the difference is too small to really measure. Splitting the manufacturers between the devices would have just prevented internet people from freakibg out over nothing by eliminating "equivalent" devices to compare against.

Of course apple couldn't have known (maybe have guessed) that this would have exploded like it did.
Click to expand...

Apple knew, but probably thought they could "deal" with it like before. Probably by the end of year, all 6s and 6s+ units being manufactured will only have the TSMC A9.
 
paulrbeers said:
No one would take your offer, not because of the CPU, but because you've used it. I don't know what you did or didn't do to your phone. That's a terrible offer. For all I know you got it wet and will stop working in 2 months and I'm boned....

If you and I walked into an Apple Store at the same time to buy our 6S's, then I wouldn't care which one I got. I'd just be happy I can afford an iPhone 6S and gladly use the crap out of it.

This is seriously #FirstWorldProblems
Click to expand...

That's seriously the best you can come up with?


Its brand new, still in factory plastic. The only thing that was done was a check for what kind of cpu and a factory reset. Now all of sudden it is suspected of water damage?

With logic like that, you should be in politics.

I could send it back to Apple today and they would give me all my money back without a single question! In fact, I'll even return this one and pick up a brand new sealed one and trade you for your used, known-TSMC one.

Still no?

Face it. It is a big deal. It does have bearing.
 
Last edited:
  • Like
Reactions: Benjamin Frost and smoking monkey
Markoth said:
Pretty horrible article.

WiFi Browsing test: The CPU is asleep most of the time. Network access is incredibly slow by CPU standards. Any impact on battery life from the CPU during WiFi browsing would be negligible, as is shown here.

WebGL: GPU-based test. The GPU is an independent part, sourced from a third party. Perhaps customized somewhat knowing Apple, but the claim was that the samsung CPU was using more energy, not the GPU. This test is mostly irrelevant.

Geekbench: The only true CPU-based test isn't valid, because it isn't based on real-world usage? Interesting. Well, then I guess this whole article was a waste of time, since this is the only CPU-based test we have to go on.

GFXBench: Another GPU test. Not relevant, because the CPU is the item in question here.

What a poorly thought-out article. There's basically no data here. Geekbench is the only one that's of any real value, but it truly isn't indicative of real-world usage. Regardless, it still shows a clear difference in battery life between the two CPUs, and at the very least shows conclusive proof of this issue. There is a difference in battery life, and if whatever your doing is taxing the CPU, you WILL notice it. That's what this has proven. Real world usage is subjective. Not everyone spends their time surfing the net.
Click to expand...

Exactly, this "test" smells more like some PR for Apple. The Geekbench test is the only component of the test that measures the battery performance of the CPU and it shows a real difference. It's written off because it doesn't conform to "real world usage" which is a term I find nebulous to say the least.
 
  • Like
Reactions: Benjamin Frost and Aston441
CEmajr said:
Wow. Too bad for the gullible saps who thought there was a noticeable difference.
Click to expand...

You guys always think it's normal, within 1-3% variance. On Geekbench itself, the delta as high as 51% on another member's test 344min vs 227min.

How many time Steve Jobs said in their commercial while he was CEO: "Think Different"
Now apple fan choose to forget that huge difference?
 
Whilst this consumer info is useful, it's also at the same time utterly useless.
Because their is no way what so ever you can tell what chip is in what iPhone without unboxing it and then taking it apart or installing a app to find out.
And their is no point taking it back to get another phone if you have a Samsung chip but want a TSMC one because you will be in the exact same situation. And you can't keep exchanging perfectly good iPhone.

That's my thoughts on it, chipgate is a non starter I feel as the stats do seem to indicate in real world use their is a very very small difference which no one will be bothered by in general.
 
  • Like
Reactions: 5105973
Not to give credence one way or another but the title of this 'Article' is complete hype. "Controlled Testing" says Macrumors. Nowhere in the Asrtechnica article do they themselves even call their test 'Controlled' In fact they say pretty much the exact opposite.

"It should be noted that no matter how many variables we try to control for, any testing of a sample size this small can only tell us so much about a problem that's this complex."

There is enough compiled data from multiple sources to safely say there is a difference is battery consumption. Even Apple admits this. The is NOT enough data to confirm how significant the difference may be and under exactly what conditions would it be the most noticeable.

Anandtech, clearly spell out why you cannot even begin to judge based on such a small sampling size due to the 'silicon lottery' that affects all semiconductor manufacturing. They do agree that there is enough composite data to indicate a difference.

http://anandtech.com/show/9708/analyzing-apple-statement-for-tsmc-and-samsung-a9
 
  • Like
Reactions: 5105973 and apolloa
wizard said:
Because it isn't a problem. The variation you are seeing here isn't any different than what is seen in laptops with Intel or AMD chips in them. You can have significant differences in thermal performance depending upon where on the wafer the chip was cut from.
Click to expand...

Your comparison to Intel and AMD is wrong because this is the same model iPhone with the exact same price not 2 different computer models from 2 different manufactures.
 
MrGuder said:
Wow this made its way back to Apple...see the power of the web and forums? It shouldn't be a shock this made news because does Samsung really have our best interest?
Click to expand...

That's such a hilarious comment, it is Apple who placed the chip orders and tested them and damn well fully know the exact minute differences between them, yet to be cool you blame Samsung? So funny.
I'm sure Apple has your best interest at heart and it is the Samsung scum that are to blame.... Being sarcastic by the way.
 
  • Like
Reactions: SirCheese
33man said:
Nobody raised the point that Samsung phones have twice bigger batteries 3000mah and last the same time as Apple twice smaller batteries... What is the difference? Samsung chip :)

Anyway reading this article I feel like : I'm using my phone wrong...

Sorry that I can install games from apples tore, I think if gaming is not considered as normal use then all games have to be removed from the store... I can understand that cell network coverage has an impact, screen brightness, cpu... But such difference... Well even 60 mins can make a difference...

And again why all this? Because people are clearly requesting better battery life APPLE!!!! Stop making flat phones and please listen back to your customers. Thanks
Click to expand...


No! If you want to play games, just get one of these covers with battery installed and you will have more than a day battery life. I don't want a bigger phone and will never install a game. For me battery is great!
 
tvbi said:
Your comparison to Intel and AMD is wrong because this is the same model iPhone with the exact same price not 2 different computer models from 2 different manufactures.
Click to expand...

All current model iPhones have the A9 chip and they all meet or exceed the specs they were sold as. Show me where one model fails to live up to the specs as promised by Apple and you might have a story.
 
Last edited:
  • Like
Reactions: cupcakes2000, apolloa and OC40
You must log in or register to reply here.
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.
Top Bottom
Back