I take it both A9 processors are closely balanced unless you're running a particular battery tests in a competition setting.
Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
According to Toms Hardware testing the Samsung is slightly better overall lol.
- Thread starter richard371
- Start date
- Sort by reaction score
best part of the article
If you want a new iPhone 6s, buy it, use it, and don't worry about who made the processor.
If you want a new iPhone 6s, buy it, use it, and don't worry about who made the processor.
And dont care if tom hardware received a big pay cheque
I think this whole chip thing has driven you crazy with the threads you keep starting and the mutterings and tinfoil hat posts you keep posting
what basis do you have to say they were bribed?, or is that just some crazy accusation you are throwing out there for the fun of it?
So every article points to the TMSC being "better" but Tom's got different results? Guess they wanted some much needed hits. Don't matter to me since my phone doesn't have a CPU inside and runs using the superior Apple cloud servers.
It wouldn't be surprise if Samsung is cheating in benchmarks again. Yes, cheating! It's nothing new to them. They have done it before.
http://www.anandtech.com/show/7384/state-of-cheating-in-android-benchmarks
1) On the Exynos 5410, Samsung was detecting the presence of certain benchmarks and raising thermal limits (and thus max GPU frequency) in order to gain an edge on those benchmarks, and
2) On both Snapdragon 600 and Exynos 5410 SGS4 platforms, Samsung was detecting the presence of certain benchmarks and automatically driving CPU voltage/frequency to their highest state right away. Also on Snapdragon platforms, all cores are plugged in immediately upon benchmark detect.
http://www.anandtech.com/show/7384/state-of-cheating-in-android-benchmarks
1) On the Exynos 5410, Samsung was detecting the presence of certain benchmarks and raising thermal limits (and thus max GPU frequency) in order to gain an edge on those benchmarks, and
2) On both Snapdragon 600 and Exynos 5410 SGS4 platforms, Samsung was detecting the presence of certain benchmarks and automatically driving CPU voltage/frequency to their highest state right away. Also on Snapdragon platforms, all cores are plugged in immediately upon benchmark detect.
not at all, as far as I know anandtech and tomshardware are the only ones to have run benchmarks. general consensus seems to be every benchmark performs extremely similar within a few percentage points, close enough that the difference could be down to normal manufacturing variance, that is every benchmark except the geekbench battery test which consistently displays a difference between the two massively in the TSMC chips favour. people don't know why that one benchmark is so different yet
the 'cheating' was software based, unless Apple chose to implement it it's not going to be a issue with the A9 whoever manufactured it.
secondly if you read the article everyone except Apple and Motorola did the same which means the majority of manufacturers did it. while it was pretty a scummy practise the problem did extend far beyond just Samsung
Well I don't trust Samsung anymore and who knows what kind of cheating they may have come up this time.
That's why I was hoping for real world gaming battery results. Geekbench battery results are still mystery... but if games were giving same kind of results, then DING DING! BINGO, we have solved this mystery.
Well I don't trust Samsung anymore and who knows what kind of cheating they may have come up this time.
That's why I was hoping for real world gaming battery results. Geekbench battery results are still mystery... but if games were giving same kind of results, then DING DING! BINGO, we have solved this mystery.
to be fair if the Samsung chip was somehow cheating then surely it would be the one winning the Geekbench battery test and not losing it?
honestly I think the best example of real world usage is to be found in the battery related threads and so far there hasn't been anything to indicate there is an issue in real world usage between the chips., I'm sure time will provide more answers though
Last edited:
how much drain is everyone else getting from the Facebook app?, I have it on mine, check it every couple of hours only for a couple of minutes but over 24 hours it only accounts for 2% of my battery drain and this doesn't seem too bad to me?, or it it prone to going crazy and draining the whole battery?
, have to say I didn't know the 6S got that kind of battery life until I had seen some of the posts here. it's pretty impressive how much drain is everyone else getting from the Facebook app?, I have it on mine, check it every couple of hours only for a couple of minutes but over 24 hours it only accounts for 2% of my battery drain and this doesn't seem too bad to me?, or it it prone to going crazy and draining the whole battery?
I was getting 32% background activity before I deleted it. I'm on 9.1b5. Here's the other half of my usage.
Attachments
So you are refering Primatelabs' Geekbench scandal then? Not APPLE's scandal? Becareful...
ouch, I will keep an eye on it. I know it was prone to doing something like that on my 4S but figured in the 2 years since I used that they might have improved it but I guess not. oh that screenshot reminds me I need to find a reddit app for my phone!
Unless you use both phones exactly the same with the same amount of apps, the same settings and especially the same brightness levels throughout your observation it would be no more than a set of your pants observation.
Even the difference in battery construction and how you charge the phone from a given percentage could result in a percentage difference.
Even the same exact cpu will differ in a point or 2.
Yes, but it's very likely to cause thermal throttling making results hard to compare.