iPhone 6s: Samsung And TSMC A9 SoCs Tested

[Gefahr]

FAKE ＴＥＳＴ
SHAME ON YOU

 

[Ching-Han]

How about this? Seems pretty scientific to me.

http://tscholak.github.io/bayesianism/apple/chipgate/2015/10/14/chipgate.html

 

[Todd_Sung]

Video #1
https://www.youtube.com/watch?v=52AXbKJO_ss

Video #2
https://www.youtube.com/watch?v=k5etZpmn6AM

Video #3
https://www.youtube.com/watch?v=_m-nKcoIAzc

 

[show_me_the_monkey]

And the paid shills have now invaded... I suspect a few "damage" control accounts already have been made just to comment on this article.. If you guy's know it all then PROVE it with scientific testing, if not quit slandering Tom's Hardware.. I personally would like to have scientific, reasonable, and thought out arguments instead of obvious troll shill fud..

Why so many new accounts - include mine - are registered to make comments on this test? It's because Tom's Hardware did this test badly and made a ridiculous conclusion, not any company pays people to do this. BTW, Samsung did pay people to do this(I think they are still doing this), was fined NTD 10,000,000 (USD 298,000) by Taiwan Fair Trade Commission. If any company want to hire people to defend the company, to attack others, they should pay to people who can speak English fluently, not guys like me or these new users.

What did Tom's HW do wrong?

1. Silicon Lottery and Sample Size
There is a very good article explains this on AnandTech website:
Analyzing Apple's Statement on TSMC and Samsung A9 SoCs
Quality of chips from the same wafer are different, Tom's HW might get a 6s with good Samsung chip and another with bad TSMC chip. Tom's HW did test 2 different model, but only 1 phone for each model, this kind of test is meaningless.

Many people say GeekBench's test results are not trustworthy, because who run it do not set their phone in the same condition(lcd backlight, airplane mode, etc.). Yes, that's true, but if you get enough samples, says 30 samples, you could still make some reasonable conclusion. someone post a good article using statistical way to check if there is any difference between Samsung SoC and TSMC SoC, and the answer is yes.
Torsten Scholak's Chipgate

2. Thermal Picture Problem
1st, anyone can easily point out that the phone with Samsung chip is hotter, because the red area is larger, almost covers the whole phone. how could a hotter phone got a ice cold apple mark? the editor made an explanation, but is not persuadable. Why not post a thermal test video, not just a picture only?
2nd, Tom's HW measure the 2 phones at 2 different position, the Samsung one is closer to Apple mark, not the hottest spot on the phone.
3rd, why there is a larger area which has higher temperature at the bottom right of the phone? If the SoC is near the Apple mark, why Tom's HW didn't get a picture with more heat at the bottom left of the phone?

 

[joonee88]

How about this? Seems pretty scientific to me.

http://tscholak.github.io/bayesianism/apple/chipgate/2015/10/14/chipgate.html

One thing - why didn't you guys test using the Geekbench battery life benchmark?

Considering it is the one that started this whole debate, it seems like it would be the first test to run...

Arstechnica found similar results to yours, except for Geekbench, where the Samsung performed a lot worse.

So the important question IMO, is why that benchmark gets such a big difference, and does it reflect any real world usage scenarios - even if not experience by most people.

We did not run the Geekbench battery test because it's not part of our usual test suite, and we're not familiar enough with how it works or with the accuracy of its results.

I saw the Ars Technica article. It seems a bit odd that every test except Geekbench battery shows minimal difference, while Geekbench shows a 20+% gap. This is a red flag to me that something might not be working right with this test and is another reason we did not include those results.

- Matt Humrick, Mobile Editor, Tom's Hardware

 

[Pieter_2]

To make it short,

Most of Asian users very concern the CP (Cost Price) Value.

All Asian pacific countries including millions of USA users are returning iPhone 6S (with Samsung A9 Processor) to swap those with TSMC processor.

Be Smart, use the best, you pay the same, why use Inferior processor??

 

[joonee88]

To make it short,

Most of Asian users very concern the CP (Cost Price) Value.

All Asian pacific countries including millions of USA users are returning iPhone 6S (with Samsung A9 Processor) to swap those with TSMC processor.

Be Smart, use the best, you pay the same, why use Inferior processor??

You know that Taiwan is not "All Asian pacific countries"? Heck, its not even recognized as a country in many places. But that's a topic for another day. And no, no one is returning their phone here in the US for this -_-...

 

[Solandri]

2. Thermal Picture Problem
1st, anyone can easily point out that the phone with Samsung chip is hotter, because the red area is larger, almost covers the whole phone. how could a hotter phone got a ice cold apple mark? the editor made an explanation, but is not persuadable. Why not post a thermal test video, not just a picture only?
2nd, Tom's HW measure the 2 phones at 2 different position, the Samsung one is closer to Apple mark, not the hottest spot on the phone.
3rd, why there is a larger area which has higher temperature at the bottom right of the phone? If the SoC is near the Apple mark, why Tom's HW didn't get a picture with more heat at the bottom left of the phone?

As I (attempted) to explain before, temperature and heat generation are not the same thing. You cannot simply snap a picture, find one component is a higher temperature, and declare that it's generating more heat. Unless both phones were (1) turned off and allowed to cool to the same temp, (2) turned on at the exact same time, and (3) run the test load for the exact same time before the thermal pictures were taken, you'd expect their temperatures to be different even if they were generating the exact same amount of heat.

I won't speculate as to how the heat from the SoC would transfer inside the back of the case (would need internal schematics and to model air circulation inside). But it's incorrect to assume the red area being larger automatically means that chip is overall hotter. Hot water freezes faster than cold water after all.
http://www.wired.com/2014/05/whats-up-with-that-mpemba-effect/

It could simply be that the 118F phone was on longer, so the air inside it started off at a higher temp, making the rest of the case heat up more than the 120F phone. But directly above the SoC where the temperature was measured, the 118F was in fact at a lower temperature because the SoC was generating less heat. Or maybe the phone on the left was fiddled with in the hand for longer prior to the photo being taken, resulting in more heat being transferred from the hand to the case, and less heat being transferred to the table surface.

There's a lot of ways to misinterpret what's going on here. A steady-state temperature reading is usually the most reliable, because it's taken after all of these variable effects have (presumably) normalized. So the temperature reading you're getting is actually indicative of the amount of heat generation.

As for the logo, I don't have an iPhone, but isn't the Apple logo a polished reflective surface? If so, then the temperature reading you get off of it is really going to be for whatever is being reflected off of it, not necessarily the temperature of the logo itself. Mirrors have a way of doing that. Based on the pics, I'd speculate the phone on the left had the ceiling reflected off the logo, while the one on the right was probably reflecting the reviewer's face.

 

[Simon_5]

I don't know. More I look at these pictures, more strange they seem to me.

1. The phone on the right has fuzzy edges all around and looks like natural heat radiation. The phone on the left has clear edges except right bottom where there is a huge shade. This can only be explained by air flow from left top corner....

2. The phone on the left has overall larger dark red area as well as orange color area.

These images are indeed questionable.

 

[darkcrayon]

No one believes me when I tell them Samsung makes apples CPU

Perhaps you could be more complete, and tell them that although Apple custom designs their CPU, Samsung or TSMC manufacture it for them.

 

[Mitch94]

as much as this sucks, you can't honestly say apples is winning when then comepetition makes their phones,Well some of them