My real world experience with TSMC vs Samsung

[daniesy]

I have the TSMC chip and was initially upset bc at first everyone thought that was the better one to have but got over it in about 5 minutes and had no plans to return. But that was thinking it was tiny differences in performance. If it's truly 2 hours and I had a Samsung I'd be exchanging till I got a tsmc. Yes it's not been proven but there's smoke.

There seem to be hundreds of people here and on reddit who have the new 6s+. Why the **** don't you guys test your phones instead of arguing here on the forums? How?
Go on youtube, find a 12h HD or 4k video, set airplane mode on, set the screen brightness on medium and let it play over night. See how much battery you have in the morning. This would perfectly fit the "real world" usage as it can be compared to the official Apple claims.
Instead of acting like a community who stands united, we insult one another and defend a company who's only interest is to take our money. Stop believing everything you're told, use your heads to think and do your own tests.

 

[Gathomblipoob]

There seem to be hundreds of people here and on reddit who have the new 6s+. Why the **** don't you guys test your phones instead of arguing here on the forums? How?
Go on youtube, find a 12h HD or 4k video, set airplane mode on, set the screen brightness on medium and let it play over night. See how much battery you have in the morning. This would perfectly fit the "real world" usage as it can be compared to the official Apple claims.
Instead of acting like a community who stands united, we insult one another and defend a company who's only interest is to take our money. Stop believing everything you're told, use your heads to think and do your own tests.

If I were a company, I'd want to take your money, too. It's business.

 

[Bram1982]

There seem to be hundreds of people here and on reddit who have the new 6s+. Why the **** don't you guys test your phones instead of arguing here on the forums? How?
Go on youtube, find a 12h HD or 4k video, set airplane mode on, set the screen brightness on medium and let it play over night. See how much battery you have in the morning. This would perfectly fit the "real world" usage as it can be compared to the official Apple claims.
Instead of acting like a community who stands united, we insult one another and defend a company who's only interest is to take our money. Stop believing everything you're told, use your heads to think and do your own tests.

Yes, this for example. Or record a total of one hour of 4K video (airplane mode on, brightness on medium) and see how much the battery has dropped starting from 100%.

 

[maxsix]

I'm very happy I kept my 6 Plus. It'll do just fine till next year. Finally if rumors materialize as true, Apple may be getting rid of the prehistoric slow creaky mechanical home button. That alone will be cause to buy a new iPhone.

 

[CEmajr]

At the end of video results are,

Samsung 55%
TSMC 62%

If we then keep on going, and do the math;

Samsung 10%
TSMC 24%

I would call this significant difference in battery life.

You really think 2 out of millions of phones is a large enough sample size to draw an accurate statistical conclusion?

There are way too many variables that factor into battery life for one to draw a conclusion from just two phones. One would need to test several hundred phones before any type of accurate trends or even outside conclusions to be made.

Do people even bother to think for themselves anymore?

Also to the OP, a benchmark is not real world usage. Totally inaccurate topic title.

 

[HEK]

Guys, the question is very simple: Can the Samsung and the TSMC version of this A9 chip do the same amount of calculations in the same amount of time (=performance) with the same amount of consumed energy (=battery power)?

With other words: Is the PERFORMANCE per WATT equal?

Let's ASSUME the Samsung version has indeed a significant worse performance to watt ratio. The next question is how much you will actually notice considering this A9 chip is one of many other components in your phone?

Situations where you read a lot of email, whatsapp your friends all day and make many phonecalls, you will NOT notice any significant difference. These are all phone usages that typically DO NOT require any real processing power on the A9 chip. However they all DO use your screen and your network/radio chip. Let's keep things simple and say the screen and network/radio chip account for 85% of the power used and the A9 chip the remaining 15% in this particular scenario. A less efficient A9 chip will only make 15% of the phones total power consumption less efficient. Apple's claim of 1-3 % average difference could easily be true.

So if YOUR usage doesnt really stress the A9 chip it doesn't really matter if you have the Samsung or the TSMC version. But consider the fact that the more computing intensive tasks you do with your phone, the bigger the difference will be.

The geekbench battery test is obviously on the other side of the scale and thus shows the biggest difference in battery life. In this case the A9 chip might easily consume 80% of the power used and the screen and radio and other components the remaining 20%. A less efficient chip wil play a much bigger role in this scenario.

But what about gaming? Or recording video's and transcoding them to lower quality or editting them? Or making a lot of live pictures during the day? Or using some TomTom navigation app? How do these stack up? Will the phone still be in the 1-3 % range compared to the TSMC chip?

And last but not least: What about the future? A future where mobile apps and games will quickly get more and more advanced and system requirements will go up. What if a game comes out that really utilizes/needs the processing power the A9 chip from the Iphone 6s has? Will we all be in for a nasty surprise where the significant lower battery life isn't limited anymore to some artificial geekbench battery test?

Im very curious if someone tested a few more scenario's except geekbench and a 1 hour youtube video.

I strongly suspect by the time such a game is forthcoming we will be on to the A10 or A11 chip and it won't much matter. For those that have the misfortune to still be using the A9 with this new game, either the battery will already been degraded by recharge cycles and they can do what we all have done as battery becomes depleted. Plug it in. It's not the end of the world, more unusual for any phone so far made.

I spent a couple years using a Mophi battery case on my 4s, did it destroy my use of the phone, NO. Did I come on here and ceaselessly whine about it claiming I was wronged by Apple somehow, NO. I firmly am of the opinion that a few people, that will always find some fault, give a wholly inappropriate weight to something that for the vast majority of users is truly inconsequential. Fact is both chips work, and phones perform within advertised parameters.

There already have been a number of people returning phones for other reasons that have discovered the supposed better battery life chip did not translate into a better battery life for them. In fact the oppose occurred in their daily real world use.

I agree with what you have written, but would add that the variability already demonstrated in real world reports shows that variations can swing completely opposite such that getting the TSMC chip is not a guarantee of better phone battery life. And certainly should not be represented here as such. Giving this issue as much windage as already been done, needlessly places worry and anxiety in people's minds. As click bate news stories go this is right up there.

I'm no defender of Apple, but this is no conspiracy or "gate" of any kind. People without understanding of statistics, manufacturing process, supplier chain process, and the complexity of multi component systems look for simple yes or no, better or worse answers that are really not so simple. And if presented that way, are actually falsehoods meant to advance opinions.

If there were a total failure, lock up, or other catastrophic even associated with a particular component I would be right there screaming bloody murder. This is NOT such a circumstance.

 

[Zimmy68]

Dude, running geekbench tests in not real world.
Come on.

 

[mjschabow]

I have been using the Samsung chip for a few days now and I think my phone might be bent.

 

[HEK]

****************

 

[HEK]

I have been using the Samsung chip for a few days now and I think my phone might be bent.

 

[Mlrollin91]

That simply not true if you know the first thing about statistics. The other variables will be random and effectively cancel each other out with a large enough sample (100 phones should be more than sufficient).

This has nothing to do with statistics, it has to do with systematic testing. Your talking about 100 phones out of millions. Each assembly line at each factory could have a different variable that effects power on the device. You would need at least a dozen phones from each assembly line from each factory to accurately run the test.

 

[Bram1982]

I strongly suspect by the time such a game is forthcoming we will be on to the A10 or A11 chip and it won't much matter. For those that have the misfortune to still be using the A9 with this new game, either the battery will already been degraded by recharge cycles and they can do what we all have done as battery becomes depleted. Plug it in. It's not the end of the world, more unusual for any phone so far made.

I spent a couple years using a Mophi battery case on my 4s, did it destroy my use of the phone, NO. Did I come on here and ceaselessly whine about it claiming I was wronged by Apple somehow, NO. I firmly am of the opinion that a few people, that will always find some fault, give a wholly inappropriate weight to something that for the vast majority of users is truly inconsequential. Fact is both chips work, and phones perform within advertised parameters.

There already have been a number of people returning phones for other reasons that have discovered the supposed better battery life chip did not translate into a better battery life for them. In fact the oppose occurred in their daily real world use.

I agree with what you have written, but would add that the variability already demonstrated in real world reports shows that variations can swing completely opposite such that getting the TSMC chip is not a guarantee of better phone battery life. And certainly should not be represented here as such. Giving this issue as much windage as already been done, needlessly places worry and anxiety in people's minds. As click bate news stories go this is right up there.

I'm no defender of Apple, but this is no conspiracy or "gate" of any kind. People without understanding of statistics, manufacturing process, supplier chain process, and the complexity of multi component systems look for simple yes or no, better or worse answers that are really not so simple. And if presented that way, are actually falsehoods meant to advance opinions.

If there were a total failure, lock up, or other catastrophic even associated with a particular component I would be right there screaming bloody murder. This is NOT such a circumstance.

Yes, I agree we shouldn't complain about things that show only in synthetic situations and we also should not worry too much about things that show only a very minor difference in real life. As long i don't see definite proof there are significant real world disadvantages of having the Samsung chip, I consider this a non-issue.

And yes, Li-Ion batteries degrade over time, in a few years the Apple A11 is the new mobile CPU to drool over and of course you can use an external power bank to charge on the go. However, all these things are part of a different discussion and shouldn't be used to 'downplay' the issue at hand (if proven to be true of course!)

That being said: This Phone should perform equally in equal situations. It should not matter whether it comes with a Samsung or a TSMC chip. On a sidenote: Setting the advertised battery life on par with the Samsung versions is a smart move by Apple, because technically the longer battery life from the TSMC one becomes just an 'extra'. Wether this is a morally good thing to do or wether or not this will result it buyers massively returning their phone in hopes of getting a TSMC one is another story

But still... a benchmark (even if its a synthetic one) that shows that big of a difference in battery life between CPU's is really not something to ignore... Not for the moment! Because unless geek bench battery test uses some very unusual code to stress the A9 chip in a way that no other real world app will ever do, the result might easily be a lot closer to some real world situations than some people here want to admit.

 

[HEK]

Yes, I agree we shouldn't complain about things that show only in synthetic situations and we also should not worry too much about things that show only a very minor difference in real life. As long i don't see definite proof there are significant real world disadvantages of having the Samsung chip, I consider this a non-issue.

And yes, Li-Ion batteries degrade over time, in a few years the Apple A11 is the new mobile CPU to drool over and of course you can use an external power bank to charge on the go. However, all these things are part of a different discussion and shouldn't be used to 'downplay' the issue at hand (if proven to be true of course!)

That being said: This Phone should perform equally in equal situations. It should not matter whether it comes with a Samsung or a TSMC chip. On a sidenote: Setting the advertised battery life on par with the Samsung versions is a smart move by Apple, because technically the longer battery life from the TSMC one becomes just an 'extra'. Wether this is a morally good thing to do or wether or not this will result it buyers massively returning their phone in hopes of getting a TSMC one is another story

But still... a benchmark (even if its a synthetic one) that shows that big of a difference in battery life between CPU's is really not something to ignore... Not for the moment! Because unless geek bench battery test uses some very unusual code to stress the A9 chip in a way that no other real world app will ever do, the result might easily be a lot closer to some real world situations than some people here want to admit.

I seriously doubt that Apple wrote the advertising to set on par with the Samsung chip. The amount of variables affecting battery life is so vast and the general public's understanding of this so poor, I believe there were more pressing concerns about battery life than the chip. As one example the vast differences poor radio reception makes to battery life would be a much more prevalent issue.

As to limiting the discussion to only chip vs battery. I respectfully disagree, as there is nothing save the geek test that any consumer does that is chip only related. Consumer always uses the phone as a whole, and any components, sub systems, and accessories that effect battery consumption and run time are in my mind related.

 

[paulgandersman]

I have a samsung iPhone 6s and I think my phone might actually be defective BEYOND the whole "chipgate" thing. I'm running geekbench battery test now and my phone is losing 1% of battery every 2.5 minutes. There's nothing running in the background either. At these numbers that would put my battery at lasting just over 4 hours of usage in a day.... that seems super super bad. Am I crazy here? or should I return this tomorrow? to me this seems way worse than the samsung specs we've been seeing. (and I'm running the test now because I have been getting surprisingly poor battery life for my first week of use.)

 

[PNutts]

Can anyone suggest me some good games I can play for 8 hours straight on my iPhone? That's the new thing, right?

There's a "Geekbench battery test" I've been hearing a lot about that seems to be getting a lot of play.

My wife and I have different chips in our 6s+ phones. I did some quick benchmarking and came up with the results that 100% of the people in our house don't care.

 

[gsingham]

If the iPhone 7 had 2hrs less battery than the 6S and made official in specs everyone would have an issue but now the TSMC has that difference with the Samsung and some are saying its insignificant...

 

[supervelous]

There seem to be hundreds of people here and on reddit who have the new 6s+. Why the **** don't you guys test your phones instead of arguing here on the forums? How?
Go on youtube, find a 12h HD or 4k video, set airplane mode on, set the screen brightness on medium and let it play over night. See how much battery you have in the morning. This would perfectly fit the "real world" usage as it can be compared to the official Apple claims.
Instead of acting like a community who stands united, we insult one another and defend a company who's only interest is to take our money. Stop believing everything you're told, use your heads to think and do your own tests.

Because I work a 60 hour a week job, go to school at night, and have a kid and a life. I have the TSMC chip as well, so not as much incentive to confirm Geekbench. Also, not like anyone would accept our test anyway. Some will lie about their results. "Medium" brightness is subjective, what if mine is ever so miniscule to the left or right from someone else's. Most of us will have different apps and other things stored on the phone, different carriers, different wifi networks, have charged it more/less than others since buying, buy it on a different date, from different manufacturing runs, with different bands in the phone, different storage capacities which may use different components between them.....etc.etc.etc.etc.

What incentive do I have to do this unscientific test you propose that everyone will just pick apart anyway?

 

[Aston441]

This has nothing to do with statistics, it has to do with systematic testing. Your talking about 100 phones out of millions. Each assembly line at each factory could have a different variable that effects power on the device. You would need at least a dozen phones from each assembly line from each factory to accurately run the test.

This is completely false with modern statistical methods. I'm not trying to rag on you, just starting a fact. Depending on the methodology and statistical model, one can obtain valid results with a very small sample.

 

[Mlrollin91]

This is completely false with modern statistical methods. I'm not trying to rag on you, just starting a fact. Depending on the methodology and statistical model, one can obtain valid results with a very small sample.

Only if this small sample includes all possible sources. And in order to ensure that all outcomes are random, you would need to have phones from every single place they are manufactured.

 

[penajmz]

This test is so flawed. It should be ran on iPhone with no data in them, no emails set up, nothing.

 

[terraphantm]

Only if this small sample includes all possible sources. And in order to ensure that all outcomes are random, you would need to have phones from every single place they are manufactured.

Or one can make the safe assumption that Apple has ensured the different plants have little to no variability and not bother controlling for different plants. And really, chances are a random sampling of phones will be distributed roughly proportionally to the manufacturing output of each plant.

 

[hawon]

Both my wife and I have TSMC and mine always goes down quicker than my wife's in real world experience. We use about same amount when we go out together with our baby and mine always end up 5-9% less than her phone. Do I worry about it? No, at the end of the day before I go to bed phone still has more than enough battery to standby whole night without charging.

I should complain apple for faulty TSMC A9 chip

 

[Bram1982]

Both my wife and I have TSMC and mine always goes down quicker than my wife's in real world experience. We use about same amount when we go out together with our baby and mine always end up 5-9% less than her phone. Do I worry about it? No, at the end of the day before I go to bed phone still has more than enough battery to standby whole night without charging.

I should complain apple for faulty TSMC A9 chip

"Using about the same during the day" can be so widely interpreted it makes indeed a good example why you should NOT complain.

The only real interesting part of this whole story is if there are real world situations that people might actually run into that cause a that significant reduction battery life compared to phones having the TSMC chip.

The real world situations i can think of where this MIGHT be an issue is:

1) If you do a lot of movie making and you like to edit the clips on your phone. For example when you're on holiday and want to send your friends a daily clip.

2) You play a lot of (cpu intensive) games on your phone during your hour and a half trip to work and back.

3) Or in the distant future, when the A9 Chip in this phone is more heavily used because of newer IOS versions / games and apps.

Until someone takes the time to actually test some real world situations, this is all there can be said. If I had two Iphones I'd happily put them to more realistic tests, but unfortunatly I'm still on my HTC one M7 which works fine for the moment.

 

[roeiz]

"Using about the same during the day" can be so widely interpreted it makes indeed a good example why you should NOT complain.

The only real interesting part of this whole story is if there are real world situations that people might actually run into that cause a that significant reduction battery life compared to phones having the TSMC chip.

The real world situations i can think of where this MIGHT be an issue is:

1) If you do a lot of movie making and you like to edit the clips on your phone. For example when you're on holiday and want to send your friends a daily clip.

2) You play a lot of (cpu intensive) games on your phone during your hour and a half trip to work and back.

3) Or in the distant future, when the A9 Chip in this phone is more heavily used because of newer IOS versions / games and apps.

Until someone takes the time to actually test some real world situations, this is all there can be said. If I had two Iphones I'd happily put them to more realistic tests, but unfortunatly I'm still on my HTC one M7 which works fine for the moment.

these are very solid reasons that are fairly common when using a smartphone these days,
i can imagine myself being part of all of them.

 

[HEK]

The only real interesting part of this whole story is if there are real world situations that people might actually run into that cause a that significant reduction battery life compared to phones having the TSMC chip.

Until someone takes the time to actually test some real world situations, this is all there can be said. If I had two Iphones I'd happily put them to more realistic tests, but unfortunatly I'm still on my HTC one M7 which works fine for the moment.

We are looking for input and comments from other non iPhone users. In particular we have not heard from any Razor flip phone users. Please chime in.