Geekbench is the only test where Samsung is worse. No matter how many times you run it, it's the same test with the same result, whereas every other test or combination of tests shows there's either nothing between them, or (as in the Tom's hardware case) the Samsung is better. How can that possibly be if the Geekbench test is accurate?
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.
My real world experience with TSMC vs Samsung
- Thread starter Markiie
- Start date
- Sort by reaction score
Geekbench is the only test where Samsung is worse. No matter how many times you run it, it's the same test with the same result, whereas every other test or combination of tests shows there's either nothing between them, or (as in the Tom's hardware case) the Samsung is better. How can that possibly be if the Geekbench test is accurate?
In a YouTube video
If more tests come out like tom's hardware then maybe ill believe there's no difference
In that YouTube video there is a 7% difference between those two particular phones after that particular battery of tests, and the vast bulk of that (5%) accumulated right at the start during the first test - the time lapse recording, which is hardly the most processor intensive task. After that both chips are basically evenly matched all the way through, with the Samsung being a touch faster overall. How does that in any way corroborate the Geekbench test which suggests anything from a 20-30% difference across the board?
To me it just suggests that the Samsung phone happened to be up to something in the background at the start of the day, that the other phone wasn't. It finished that background task during the first test and held its own from that point onward.
so far the geekbench battery test is the only benchmark to show a difference, they didn't test it as they don't use it, they did mention the results others have gotten from it and at this point I think there is enough evidence that you can assume results will be similar even if they did run it.
http://www.tomshardware.com/reviews/how-we-test-smartphones-tablets,3894.html
every single other benchmark on anandtech and tom's hardware is nearly identical. you can't just ignore that and keep throwing one benchmark at people while saying the others don't count, check the geekbench chart, there are devices with strange results all over it, like the Galaxy S6 which definitely does not have better battery life then the 6 Plus in real life
every other benchmark seems to back Apple's claims that they perform very close to each other in real life and it seems it's down more to the silicon lottery then which chip you have, a couple of posters on here have returned Samsung devices and gotten TSMC devices and say they have worse battery life.
so far the geekbench battery test is the only benchmark to show a difference, they didn't test it as they don't use it, they did mention the results others have gotten from it and at this point I think there is enough evidence that you can assume results will be similar even if they did run it.
http://www.tomshardware.com/reviews/how-we-test-smartphones-tablets,3894.html
every single other benchmark on anandtech and tom's hardware is nearly identical. you can't just ignore that and keep throwing one benchmark at people while saying the others don't count, check the geekbench chart, there are devices with strange results all over it, like the Galaxy S6 which definitely does not have better battery life then the 6 Plus in real life
every other benchmark seems to back Apple's claims that they perform very close to each other in real life and it seems it's down more to the silicon lottery then which chip you have, a couple of posters on here have returned Samsung devices and gotten TSMC devices and say they have worse battery life.
Geekbench shows a big difference though. And it's only Tom's hardware and one benchmark out of 4 on another site that samsung lasts longer. And it's only people on here saying their samsungs last longer and I bet they're lying.
now they are going to swap again their TSMC phone for a Samsung ...
But still the vast bulk of that gap opens up during time lapse photography. Not the 4K recording, gaming, graphics tests, or movie exporting, where there is negligible or no difference at all. How do you explain that?
One other thing I just noticed - in that video the geekbench test actually suggests that the TSMC chip is slightly faster - I can honestly say that's the ONLY time I've seen geekbench suggest that and I'm pretty sure that if you study the geekbench results for performance you won't find that supported there.
Last edited:
lol .... and you surely are better then them.
SO the test are wrong when they are not in favor of your ridiculous hysteria ?
BTW the A9 doesn't suffer from thermal throttling, as already proved by Anandtech.
that's true but those other results excluding the Geekbench test on Anandtech are within normal variation where 2-7% difference seems pretty common even with identical chips, extreme cases could be more if you are really unlucky, I'm inclined to agree with the comment below that was in the comment section on tom's hardware
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
well I have the Samsung chip and posted my battery results with screenshots as have others which refutes your claim people are lying. so far I haven't seen any results that show the Samsung is inferior in the real world, my battery life is crazy good and seems at least equivalent to any TSMC results I've seen
There are actually ONLY TWO controlled test so far.
One on Ars Technica that titled "Apple is right", and one on Tom's Hardware saying the two chips are almost identical (with a negligible advantage for Samsung's).
But sure, we must trust some ridiculous YouTube videos as an evidence and totally discharge every other serious test.
And again, ONLY GEEKBENCH so far said something different.
It could be the test itself, and so Apple is right: not a representation of real life conditions.
The Tom's HW test is quite accurate.
now they are going to swap again their TSMC phone for a Samsung ...
oh how i wish this day will come!
this would be just epic!
Correct mate, but the whiners here are going to discharge every serious test not supporting their conspiracy theory.that's true but those other results excluding the Geekbench test on Anandtech are within normal variation where 2-7% difference seems pretty common even with identical chips, extreme cases could be more if you are really unlucky, I'm inclined to agree with the comment below that was in the comment section on tom's hardware
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
well I have the Samsung chip and posted my battery results with screenshots as have others which refutes your claim people are lying. so far I haven't seen any results that show the Samsung is inferior in the real world, my battery life is crazy good and seems at least equivalent to any TSMC results I've seen
Tom's HW is in the field of hardware testing since when ? 15 years ? But surely Mr. NoOne on this forum knows better than them how to test hardware.
Lol
The other website with 4 tests the Samsung only won one of them, and by 4 minutes. Tsmc 6s wins all geekbench except 1 anomaly. So yea that's why I don't trust tom's hardware test. Maybe if more tests come around suggesting the same but so far all the tests suggest tsmc is better.
True, but he then records 10 minutes of 4K video (which is surely more intensive?) and there's nothing between them, then spends 11-12 minutes exporting the movie and again there's nothing in it, surely if time lapse burns through a whole 5% in 30 minutes there should be a bigger gap continuing to open up?
In the Ars Technica test all four margins were absolutely minimal and well within typical sample variation with the exception, yet again, of the Geekbench test. Take that out of the equation and there is simply no story here.
Well if more people do some more real world tests on YouTube he'll be exposed. Hopefully everythingapplepro does one. I'd rather this not be an issue than be one.