|
||||||||
IPhone 6s breaks battery efficiency records. |
![]() |
|
|
Thread Tools | Search this Thread |
|
|
#76 |
|
Forum Member
Join Date: May 2010
Posts: 11,488
|
Quote:
You really have to look at the history of this when attempting to make any sense of it, suppliers and manufacturers seldom telling the public anything.
What we know is that there are now 3 chip suppliers involved, 2 providing 14 nm and 1 latecome, TSMC providing 16nm. Newcomer Global Foundries's new 14nm was said to be at a low production rate so it forced Apple into last minute using TSMC, Apple having only signed up to Samsung to buy what spare capacity Samsung had available. There is a shedload of missing info here. Are the Global Foundry chips of low quality neccessitating a higher voltage requirement and lower battery life. Or did TSMC give a mixed supply of Finfet and Finfet+, the + version having ULV capabilities for lower wattage? Or are some TSMC ones at fault, running slower when warm, thus improving battery life at the cost of performance? I doubt anyone will find out, and as well as having 3 suppliers and a theoretically 4 possible Socs, we also have that possibility of Apple using various battery supplies, of mildly mixed performace/quality. The only confidence I see is that the 14nm Samsung and Global Foundries silicon runs faster, but not the 10% faster CPU some have claimed. Yet even there, TSMC Finfet+, if used, could be overclocked to compensate for 16nm slowness, and still use less battery! If we are really going to look into history, I would suggest we learn that lots of people who show a pathological dislike for Apple will keep posting claiming it is a huge issue. Click bate articles will be written, and those that own and use the devices will get on with their lives and be very happy with the device they purchased. |
|
|
|
|
Please sign in or register to remove this advertisement.
|
|
|
#77 |
|
Forum Member
Join Date: Apr 2008
Posts: 3,153
|
I've got a 6s+ with the samsung a9 and the battery in real world use is the same as my old 6+.
It just took a few re-charge cycles for the battery to get up to speed. These comparisons are utter drivel as there are 2 many variables involved and the motivation is to create a bend gate style click bait non issue so some sad nerd can make some money. |
|
|
|
|
|
#78 |
|
Forum Member
Join Date: Oct 2003
Location: the wild world web
Posts: 28,132
|
Quote:
...
These comparisons are utter drivel as there are 2 many variables involved and the motivation is to create a bend gate style click bait non issue so some sad nerd can make some money. Im being simplistic in saying someone made it up, but as that is where most of it started, without any proper technical data, we have to assume it mainly fake. What I find totally baffling, is why Apple did not simply use 16nm for the non Plus and 14nm for the Plus. Different Socs for different products would likely have left out all the confusion. |
|
|
|
|
|
#79 |
|
Forum Member
Join Date: Jul 2002
Location: a land filled with trolls
Posts: 12,010
|
The Samsung and Apple fanboys are probably totally unable to cope with the paradox here.
Samsung fans ridicule Apple, but it's a Samsung chip causing the issue. Oops. Apple fans insist it's not an issue, but at the same time it's down to Samsung so how can you miss an opportunity to attack? Ultimately, this issue is apparently serious enough that Apple will seek to drop Samsung completely from supplying the chipsets for the next iPhone - which means that internally this is looked upon as a problem even if officially it's no big deal. Personally, I'm glad such practices are being made more public as this isn't a new thing and certainly not confined to Apple or Samsung - but most people didn't have a clue before and other devices have had a similar 'lottery' where some devices performed better than others. As I said, Galaxy S6 owners have at least two image sensors with different characteristics, and Sony devices can have different displays. I am sure there are many more examples I'm not aware of. |
|
|
|
|
#80 |
|
Forum Member
Join Date: Jan 2013
Location: West Midlands
Posts: 2,450
|
Checked the Mrs 6S yesterday, it's a Samsung chip but I couldn't give a toss to be honest, the battery life is way way better than the 5S it replaced and it's also abit better than my 6.
|
|
|
|
|
|
#81 |
|
Forum Member
Join Date: May 2010
Posts: 11,488
|
Quote:
The Samsung and Apple fanboys are probably totally unable to cope with the paradox here.
Samsung fans ridicule Apple, but it's a Samsung chip causing the issue. Oops. Apple fans insist it's not an issue, but at the same time it's down to Samsung so how can you miss an opportunity to attack? Ultimately, this issue is apparently serious enough that Apple will seek to drop Samsung completely from supplying the chipsets for the next iPhone - which means that internally this is looked upon as a problem even if officially it's no big deal. Personally, I'm glad such practices are being made more public as this isn't a new thing and certainly not confined to Apple or Samsung - but most people didn't have a clue before and other devices have had a similar 'lottery' where some devices performed better than others. As I said, Galaxy S6 owners have at least two image sensors with different characteristics, and Sony devices can have different displays. I am sure there are many more examples I'm not aware of. Why would anyone attack Samsung for a non issue
|
|
|
|
|
|
#82 |
|
Forum Member
Join Date: Dec 2005
Location: Kent
Posts: 8,954
|
Quote:
http://9to5mac.com/2015/10/08/tsmc-i...sung-chipgate/
Differences seem to be linked to specific usage patterns. Not sure I'd want to be caught with the Samsung one though. Wonder why the app has been pulled ? Quote:
Dear users of Lirum Device Info,
We are aware of some serious issues or our Apps with the latest iOS models (iPhone 6S and iPad Air 2). Changes on the requirements for a new update to be approved are also delaying the development process (and our team is very small). In face of such events, in 24 hours, we will take Lirum Device Info down from the App Store - until we can release a decent update. That will take a few months however - but then we promise an entire new user interface, faster updates when a new device model is released, and a lot of new features. Until then, the app will continue to work on models equal or prior to the iPhone 6 and 6 Plus. Best Regards, Rogério Hirooka Lirum Labs |
|
|
|
|
|
#83 |
|
Forum Member
Join Date: Oct 2003
Location: the wild world web
Posts: 28,132
|
While the Samsung bad battery caught the webs attention, Im more getting the impression the problem is battery or software related.
" i have two i6s 64GB, i have and with GeekBench one shows 4:25 the other one shows 3:25, same samsung" That appears a bonafide comment at cultofmac, mimicking similar large differences in battery life for Samsung versus Samsung. Whether one or any of those Samsung chips can actually be Global Foundries chips I'm unsure. |
|
|
|
|
|
#84 |
|
Forum Member
Join Date: Oct 2003
Location: the wild world web
Posts: 28,132
|
http://arstechnica.com/apple/2015/10...two-apple-a9s/
Interesting data, and an honest admission on the many other factors that might come into it. The 14nm is certainly showing more power usage under heavier load, and Im not even sure if that would be down to the 16nm throttling the CPU. Andrew did not cover that question. Theres no mention either of the test running the 14nm any slower to account for its decent extra speed. An interesting comment also suggested using ice packs to see if one is throttling a lot on some of the tests, throttling a common intelligent question. And if those 16nm are the new Finfet+, maybe that is the answer to the differences in battery under heavier load. |
|
|
|
|
|
#85 |
|
Forum Member
Join Date: Oct 2003
Location: the wild world web
Posts: 28,132
|
I might as will add that the stuff I read on TSMC suggested their 16nm Finfets were to be on 20nm silicon size, but obviously the Apple A9 is 16nm size, shown by it being only 10% bigger than the 14nm ones.
I'm wondering slightly more if TSMC were only tooled up for producing Finfet+ at short notice, a 20nm size 16nm Finfet Soc being too large for Apple. |
|
|
|
|
|
#86 |
|
Forum Member
Join Date: May 2010
Posts: 11,488
|
Quote:
http://arstechnica.com/apple/2015/10...two-apple-a9s/
Interesting data, and an honest admission on the many other factors that might come into it. The 14nm is certainly showing more power usage under heavier load, and Im not even sure if that would be down to the 16nm throttling the CPU. Andrew did not cover that question. Theres no mention either of the test running the 14nm any slower to account for its decent extra speed. An interesting comment also suggested using ice packs to see if one is throttling a lot on some of the tests, throttling a common intelligent question. And if those 16nm are the new Finfet+, maybe that is the answer to the differences in battery under heavier load. |
|
|
|
|
|
#87 |
|
Forum Member
Join Date: Oct 2003
Location: the wild world web
Posts: 28,132
|
Maybe Ars will have a follow up report as to why the Geekbench CPU test has 28% better battery for TSMC, way above any possible 10% extra processing done by the Samsung one.
In the closer GPU test I, possibly incorrectly assume the Samsung one has done the most work. So on the face of it there, looks like we only have major discrepancy in the CPU, not the GPU part of the SOC. |
|
|
|
|
|
#88 |
|
Forum Member
Join Date: Jul 2002
Location: a land filled with trolls
Posts: 12,010
|
We won't know until next year, but I'm hearing Samsung won't be a supplier for the next iPhone and this situation was designed to help 'cover' Apple during the transition, and of course the problem now makes it even easier to justify (if justification was needed as Apple doesn't need the support of customers - even if customers will now be 'blaming' Samsung).
It's very odd that people want to defend Apple by saying it's no big deal, when it is clearly a huge deal for Apple and Samsung. Apple will drop Samsung as quick as it can - unless of course Samsung begs for forgiveness and Apple agrees to let them continue but at a much lower cost. Rather like experts predict Qualcomm will give Samsung an amazing price on Snapdragon 820 and we may well see Samsung one of the first manufacturers to use the chipset, after Samsung famously boycotted Qualcomm over the SD810 thermal issues. It's amazing to see how things will be playing out in public and behind the scenes. |
|
|
|
|
#89 |
|
Forum Member
Join Date: May 2010
Posts: 11,488
|
Quote:
We won't know until next year, but I'm hearing Samsung won't be a supplier for the next iPhone and this situation was designed to help 'cover' Apple during the transition, and of course the problem now makes it even easier to justify (if justification was needed as Apple doesn't need the support of customers - even if customers will now be 'blaming' Samsung).
It's very odd that people want to defend Apple by saying it's no big deal, when it is clearly a huge deal for Apple and Samsung. Apple will drop Samsung as quick as it can - unless of course Samsung begs for forgiveness and Apple agrees to let them continue but at a much lower cost. Rather like experts predict Qualcomm will give Samsung an amazing price on Snapdragon 820 and we may well see Samsung one of the first manufacturers to use the chipset, after Samsung famously boycotted Qualcomm over the SD810 thermal issues. It's amazing to see how things will be playing out in public and behind the scenes. |
|
|
|
|
|
#90 |
|
Forum Member
Join Date: Oct 2003
Location: the wild world web
Posts: 28,132
|
OK, seems that both TSMC and Samsung use a 20nm back-end for their chips.
Interestingly Samsung mainly sell their first generation 14nm called LPE, but they seemingly market a lower power higher cost second generation 14nm called LPP. There was even some speculation back in April that the A9 'might' use lower power LPP. edit- this said there might be major differences between TSMC and Samsung. http://www.dailytech.com/TSMC+Hypes+...ticle37298.htm "A compelling question is why top clients like Apple, Qualcomm, and NVIDIA are picking Samsung over TSMC at the 14/16 nm node, if TSMC's process is truly so technically superior. Something is going wrong with TSMC -- be it cost, capacity, or falling short of its amibitious performance claims. But at the end of the day, whatever lead TSMC has on paper over Samsung is largely moot as clients appear to have some inside insight which convinced them that Samsung was the superior fab partner at this node." It certainly starts to look like the TSMC lead is major, at least over LPE. |
|
|
|
|
|
#91 |
|
Forum Member
Join Date: Aug 2005
Posts: 4,206
|
Quote:
Maybe I've misunderstood your post. Why do you think this is a huge deal for Apple and Samsung?
|
|
|
|
|
|
#92 |
|
Forum Member
Join Date: Jul 2002
Location: a land filled with trolls
Posts: 12,010
|
As I said, we'll see with the iPhone 7.
I never said the word 'chipgate' either. |
|
|
|
|
#93 |
|
Forum Member
Join Date: Jul 2007
Posts: 3,985
|
Quote:
"A compelling question is why top clients like Apple, Qualcomm, and NVIDIA are picking Samsung over TSMC at the 14/16 nm node, if TSMC's process is truly so technically superior. Something is going wrong with TSMC -- be it cost, capacity, or falling short of its amibitious performance claims. . Imagine the carnage if TSMC hadn't come through with the chip on time and in quantity and the launch had to be pushed back? With TSMC now having shown they are capable of producing a technologically good product that appears to be superior to a similar product of one of he market leaders it's likely Apple will invest in them and use them more for their next product, as people are already saying. |
|
|
|
|
|
#94 |
|
Forum Member
Join Date: May 2010
Posts: 11,488
|
Quote:
As I said, we'll see with the iPhone 7.
I never said the word 'chipgate' either. I still don't understand what you meant by a huge issue for Apple and Samsung. |
|
|
|
|
|
#95 |
|
Forum Member
Join Date: May 2010
Posts: 11,488
|
Quote:
I think in Apple's case it was probably a decision of not putting 100 million + eggs in one basket with a new supplier.
Imagine the carnage if TSMC hadn't come through with the chip on time and in quantity and the launch had to be pushed back? With TSMC now having shown they are capable of producing a technologically good product that appears to be superior to a similar product of one of he market leaders it's likely Apple will invest in them and use them more for their next product, as people are already saying.
|
|
|
|
|
|
#96 |
|
Forum Member
Join Date: Jul 2002
Location: a land filled with trolls
Posts: 12,010
|
Look what happened to Qualcomm this year. It's like saying everyone is just one pay check away from losing their home.
I am sure Samsung can bounce back but I also reckon Apple how has a major bargaining tool now, especially if it lines up other component suppliers to gradually reduce Samsung's supply to near nil. But, Samsung isn't going to just give up is it? I don't need to be a fly on the wall in Cupertino to know this anymore than you do. |
|
|
|
|
#97 |
|
Forum Member
Join Date: Jul 2007
Posts: 3,985
|
Quote:
I don't need to be a fly on the wall in Cupertino to know this anymore than you do. .
|
|
|
|
|
|
#98 |
|
Forum Member
Join Date: Aug 2005
Posts: 4,206
|
Quote:
I think in Apple's case it was probably a decision of not putting 100 million + eggs in one basket with a new supplier.
Imagine the carnage if TSMC hadn't come through with the chip on time and in quantity and the launch had to be pushed back? With TSMC now having shown they are capable of producing a technologically good product that appears to be superior to a similar product of one of he market leaders it's likely Apple will invest in them and use them more for their next product, as people are already saying. |
|
|
|
|
|
#99 |
|
Forum Member
Join Date: Jul 2002
Location: a land filled with trolls
Posts: 12,010
|
Indeed. There are people who can discuss things without it turning into a flame war.
Hence why the hardcore fanboys are so conflicted here. Here's a problem involving two rivals and already people are saying 'chipgate' and other nonsense. There is a problem for Apple and Samsung, and it will be dealt with. No doubt Apple has had many such issues with component suppliers in the past that we don't even know about, just as any company will have. Nevertheless, I can see on other forums and social media that a lot of people are now trying to swap their phones (and I suspect some will lie about faults to do so) just as Xperia Z1 owners did (and also lied about faults) and Galaxy owners will have for the 'right' image sensor. Not that any company is going to reveal the numbers for obvious reasons. And in a while it will be forgotten about by most people. There is a plus side for manufacturers though. I bet people who get the version they want will take extra care from then on, as they won't want to risk having a replacement. |
|
|
|
|
#100 |
|
Forum Member
Join Date: Sep 2007
Posts: 620
|
I have a TSMC 6S+. I just came from a Nexus 6 which has a larger battery and is running the final version of Android Marshmallow. I have to say, it's only my 2nd day with the iPhone so far (100% @ 10am to 30% at 12am) and it's performing just as good if not better than the Nexus in terms of battery. I was using the phone all day as I just got it, I've installed probably over 100 apps, I've played many hours of games, watched about an hour of YouTube, gone through social apps like Facebook/Instagram/Vine/Snapchat/Whatsapp for maybe 2 hours, and generally scrolled through the settings and opened every app I have on the phone to check it out, oh and a couple of phone calls.
To sum it up, it's done better than I expected considering the size of the battery. But my god, it charges SLOWLY when compared to fast charging my Nexus. I am happy I got the TSMC as I've read it's the better one in terms of battery life. Performance is blazing fast when using the phone and it doesn't seem to warm up at all. The only problem I've noticed is when unlocking the phone, almost every single time, it will lag for the first 3-5 seconds, almost like the A9 chip hasn't fully woken up. |
|
|
|
![]() |
|
All times are GMT. The time now is 17:44.





.