Apple iPhone 5s performance review: CPU and GPU speed compared to top Android phones (benchmarks)
0. phoneArena 25 Sep 2013, 06:47 posted on
The Apple iPhone 5s brings is one of the biggest under-the-hood upgrades to ever happen to the iPhone line. Apple’s A7 chip comes with a new Cyclone core that is very well optimized in terms of power consumption and still manages to deliver great performance…
This is a discussion for a news. To read the whole news, click here
95. therealestmc (Posts: 316; Member since: 23 Jul 2012)
What will you say when apple put the A7 in the next and wipe the floor with android tablets and phones?
46. boosook (Posts: 1436; Member since: 19 Nov 2012)
Impressive! But this won't make me switch to Apple... ;)
Btw, there's one test with the nexus 5, and it's the only one that can keep up with the iphone, so we'll see.
47. Eonnaydra (Posts: 217; Member since: 23 Oct 2012)
The benchmarks are important now because they show us how Apple&Samsung team made an amazing SoC clocked just at 1.3 Ghz humiliate Snapdragon 800/other SoCs.
I think that Samsungs executives are too blind to see that the numbers of cores/specs are not important.They chose to put big numbers because stupid people will think that the phone with more cores is faster.
Thats why S4 has so much lag,Samsung must release a clean&modern&well optimized UI and release a new efficient Exynos.
78. TechieXP1969 (limited) (Posts: 10115; Member since: 25 Sep 2013)
What? You don't even understand the tests...do you?
If you knew anything about the phones, you will understand that the UI on Samsung phones or any other phone has anything to do with these tests.
Look at the GPU tests. All the phones but 2 have something in common. TO start, All but the 2 iPhones are pushing 1920x1080 resolutions vs 1136x640. In ANY GPU benchmarking tests where a phone or PC has a less resolution to push, the GPU will work less hardware and can push more frames. A full HD resolutions has over 1M more pixels to push versus the 2 iPhones. As you can see in the tests, the other phones were in a very close range while it appears the iPhones took off. Fact is, is an iPhone had full HD, its scores would also be closer.
Also when it comes to gaming, take 2 gams that are pretty much identical. Here is what I ask you to do. Take one of those above Android phones like the GS4 which I prefer and put it against a iPhone 5S and load Need for Speed Most Wanted. See which one looks the best, plays the best and watch which one drops the most frames.
I have NFS MW for the iPhone and Android and I grabbed an iPhone 5S from one of the lucky guys in my office who got one and I played the game side by side. The differences are like night an day. The only thing about the iPhone was on daytime tracks, the white lights which is produced by the bulb was better than on the AMOLED screen which doesn't produce a nature white. But on tracks that were dark or it was dusk, my Galaxy Note 2 with only just a 1280x720 screen blew the iPhone 5S out the water. And yes Quadcores made a huge difference vs dual.
48. Bootutu (Posts: 227; Member since: 11 Jul 2013)
2 top Galaxy S4 with Adreno 330 are faster than PowerVR Rogue 6430
50. amansingal14 (Posts: 309; Member since: 08 Sep 2012)
Did anyone notice the name - Google Nexus 5, in the last benchmark and final comparision?
52. itsdeepak4u2000 (Posts: 3718; Member since: 03 Nov 2012)
A7 chip is amazing and 5s benchmarks are great with such low specs. One thing is for sure that iOS7 is superbly optimized despite its looks. But what's Nexus 5 doing there?
55. adrianzka (Posts: 60; Member since: 25 Apr 2013)
Where is the triangle test where iphone 5 beats iphone 5s, huh PA. It's in anandtech's review
60. Bootutu (Posts: 227; Member since: 11 Jul 2013)
They are only cherry picking where the 5S done well, GLbenchmark website tells a different story.
57. tryingHARD (Posts: 64; Member since: 31 Dec 2012)
I cant wait to give my s4 to my bro-in-law and buy this beautiful phone. My s4 octa core model lags hard that i cant even answer calls! I really hate android now! Or just samsung. I havent tried other androids but samsung.
58. Eonnaydra (Posts: 217; Member since: 23 Oct 2012)
))) Me too I had such problems.I have used S4 since its lunch,but now Im gonna buy the 5S.My bro has a lot of phones and even the iPhone 4 is faster than S4 in screen latency,fluidity and smoothness of the UI.
59. Spikes9s (Posts: 52; Member since: 09 Aug 2013)
All you cry brats please, stop crying about lag on your phone, open your mind go to your phone developers tool settings and change some stuff there to get rid of the lag... if really there is any lag, do some homework before you bash ur actual phone or one particular platform or the other, you will be surprized what can you do just by getting into some friendly forums.....
67. bucky (Posts: 2563; Member since: 30 Sep 2009)
I agree with you to an extent. If you're spending a huge chunk of cash you shouldn't have to baby sit the thing.
63. vikcytec (Posts: 1; Member since: 25 Sep 2013)
Dual core cpu and 1gig ram performing better than quad core cpu with 2 gig ram. Is it OS optimization or better hardware architecture?
64. OpTiMuS_BlAcK (Posts: 418; Member since: 04 May 2012)
I'd wish more Android manufacturers would start using PowerVR graphics T_T
I'm still betting on Tegra 4 though~
85. PapaSmurf (Posts: 10456; Member since: 14 May 2012)
Galaxy S4 (Exynos Octa) uses the Power VR GPU.
System chip: Exynos 5 Octa
Processor: 8-core, 1600 MHz, (1.6 GHz quad-core ARM Cortex-A15 and 1.2 GHz quad-core ARM Cortex-A7)
Graphics processor: PowerVR SGX544MP3
88. OpTiMuS_BlAcK (Posts: 418; Member since: 04 May 2012)
I know, but not even all S4s use it..
If only TI OMAP was still in the competition then we'd see more PowerVR-powered phones..
75. TechieXP1969 (limited) (Posts: 10115; Member since: 25 Sep 2013)
Hate what exactly? No matter what those tests say, real world usage is what the vast majority are going to consider. Having a faster device on paper doesn't mean nothing. Isn't that what all iFans say? Oh that's right, until your device on paper is actually the winner.
Why do we need to hate? Apple in 2011 sold 72M total phones with 3 selling models. Samsung sold over 50M of just one. Combine all the rest and they sold over 100M phones With more than half being on the same caliper or better than iPhone. Namely GS3, GS4 GN2.
72. TechieXP1969 (limited) (Posts: 10115; Member since: 25 Sep 2013)
When doing GPU benchmarks, you should always offer a disclaimer. With the exception of the iPhone 5, all the other phones are rocking full 1920x1080 displays. That means those tests are pushing 1M's of more pixels which will always make a phone with a lesser resolution seem to be better when factually the numbers would be much closer.
You see in all the tests, the other phones are so much closer to each other because in having similar hardware, they scored much closer.
I real world usage and in reviews, those listed Android phones have beaten even the 5S in usages test. Benchmarks that show hardware capability is one thing, how much of that you're going to use where the difference is actually seen is pretty much ZERO.
Its like saying, my Ferrari is faster than you Toyota. But in real world experience like on a typical major highway that runs through a major city, all those advantages mean nothing when you're sitting in traffic.
74. Lakuta1978 (Posts: 64; Member since: 30 Jun 2012)
dualcore beat quadcore?????? thats awesome.........!!!!
80. TechieXP1969 (limited) (Posts: 10115; Member since: 25 Sep 2013)
The GPU tests are also lame because those other phones are pushing full 1920x1080 displays while the iPhone is pushing 1136x640 which isn't even HD.
If you took 2 TV's and one was 40" and the other was 60", the 60" TV pushing the same resolution would be slower than a 40", simple because the tv is smaller even with the same resolution.
In the case of the iPhone vs the Androids, not only is the resolution lower, the screen is smaller too. That means the GPU in the iPhone isn't working as hard. Its not pushing millions of more pixels and its doing it on a smaller screen too.
Those are facts
87. bigstrudel (Posts: 518; Member since: 20 Aug 2012)
Those are half truths.
CPU and browser optimization are both in play for the Sunspider benchmark. And the CPU does matter.
Screen size literally means nothing. Only resolution. If there are resolution differences; ignore onscreen tests. Offscreen tests are valid between devices because it runs at a simulated resolution.
91. Ninetysix (Posts: 2378; Member since: 08 Oct 2012)
Should be ashamed to name yourself "Techie".
96. therealestmc (Posts: 316; Member since: 23 Jul 2012)
Really, dude? If I did not know about tech I would have thought you knew what you were talking about. Do you know what a Off-screen test mean? Google is your best friend.
79. dancorcal (Posts: 39; Member since: 23 Jan 2013)
I'm impressed. I still hate IOS though...especially the new 7. Looks like crap to me. Good to see nexus 5 is keeping up as that is my next phone. The other makers need to step it up and do some work cause they just got their a$$ handed to them.
83. aleksey4738 (Posts: 10; Member since: 27 Apr 2013)
my favorite thing about iphones. is how innovative they are with their design! it changes so much with every iphone! and their iOS?? Wowowowo its so badass!!! their speakers i mean their little speaker is so loud! their 8+ megapixel camera, wow definetely better than the z1, and lumia 1020, g2 etc. and to put icing on the cake..........their screen resolution!!!!wow. that is impresive!! who needs 1080p or even 720p?? anybody got time for that!! wake up ilovers.
89. bootroot (Posts: 35; Member since: 11 Sep 2012)
Be professional its your job !Iphoen 5s is NO WAY the fastest : offscreen 1080p test at gfx bench
Sony Xperia Z Ultra (C6802, C6806, C6833)
1920 x 1080
6798 Frames (60.2 Fps)
Sony Xperia Z1 (C6902, C6903, C6906, C6943, SO-01F)
1920 x 1080
6793 Frames (60.1 Fps)
NTT DoCoMo FJDEV009
1920 x 1080
6619 Frames (58.6 Fps)
Google Nexus 5
1920 x 1080
6533 Frames (57.8 Fps)
1920 x 1080
6424 Frames (56.8 Fps)
Apple iPhone 5S
1920 x 1080
6399 Frames (56.6 Fps)
Apple A7 GPU
1920 x 1080
6301 Frames (55.8 Fps)
1920 x 1080
6256 Frames (55.4 Fps)
LG F320L Optimus G2
1920 x 1080
6225 Frames (55.1 Fps)
LG D802 G2
1920 x 1080
6190 Frames (54.8 Fps)
99. bigstrudel (Posts: 518; Member since: 20 Aug 2012)
Those results only show that the Rogue GPU in the 5S is just as powerful as the Adreno 330.
But there's nothing wrong with being just as good as the Adreno 330. It's a beast of a graphics chip that fits in large devices with big batteries.
106. Taters (banned) (Posts: 6474; Member since: 28 Jan 2013)
Yup, except the CPU, ram, display, and everything else basically is all crappier on the iPhone 5. If you charge 719 Canadian a pop for the 16gb model, you better have more than just the GPU on par with S800 devices selling for a lot cheaper.
113. bigstrudel (Posts: 518; Member since: 20 Aug 2012)
Actually the A7 Dual core at 1.3ghz beats the Snapdragon 800 Quad at 2.3ghz handily in Geekbench 3 in both single and multi core benchmarks. And that's with a huge clock speed advantage and double the cores.
115. Taters (banned) (Posts: 6474; Member since: 28 Jan 2013)
Making stuff up now. The only Geekbench 3 results I see is the Xperia Z ultra hitting 2800 overall while the iPhone 5s hitting 2556.
It's doing this while pushing the much more intensive OS, browser, and display.
Sounds like you need to learn how to think.
103. jmonteiro829 (Posts: 186; Member since: 13 Feb 2012)
Am I foolish to think that these benchmarks don't mean anything? Essentially we are talking apples to oranges since the each OS in not loaded on the same hardware.
104. JMartin22 (Posts: 1969; Member since: 30 Apr 2013)
Apple's running next-generation SoC architecture that everyone else will be running very soon. Since its the first phone to have it, naturally it's going to be better than tech running outdated hardware.
But as always, Android phones will swiftly surpass it once it gets ARMv8 inside it as well.
105. Taters (banned) (Posts: 6474; Member since: 28 Jan 2013)
Hiding behind a low res screen, safari vs chrome, and a simple is again. When will people learn and not be impressed by them?
Running chrome on both the nexus 4 crushed the iPhone 5 in the browser tests. Any S800 would crush the 5s if both used chrome for the test. The only reason the iPhone wins these sunspider tests is because Chrome is slow as ass and safari isn't.
Also as you see from the off screen, the adreno 330 is on par or better. So if the S800 had a 1160 x 640 resolution screen, lower res than even mid range android devices like the one mini and xperia sp, the 5s wouldn't win a single benchmark.
Also funny how Anandtech and others convienantly never show cpu benchmarks and only gpu and browser benchmarks. Or even read and write benchmarks. Only when comparing Android devices they show non GPU and browser benchmarks. Are they afraid that their poor Apple will get embarrassed? Lol
108. Ninetysix (Posts: 2378; Member since: 08 Oct 2012)
Yep .I mean gee golly, a dual core cpu running at 1.3GHz is beating a quad core CPU running at 2GHz+ is nothing to gloat about. Apple should have used a Qualcomm S800 in their iphones so Taters will be happy.
114. Taters (banned) (Posts: 6474; Member since: 28 Jan 2013)
Except it doesn't beat it in anything. Safari beats Chrome in browser tests and in the off screen, where they take into account important variables such as screen resolution, the S800 wins.
Where exactly is this beating coming from? I wouldn't call hiding being Safari vs chrome and hiding behind a low res screen to artificially inflate your on screen scores as beating anyone at anything.
118. bigstrudel (Posts: 518; Member since: 20 Aug 2012)
And if you're really interested in where the proof is ill just put up a benchmark that PA didn't run.
Geekbench 3 (higher is better)
iPhone 5S: Single Core: 1414
Multi Core: 2564
LG G2: Single Core: 888
Multi Core: 2229
And the A7 does this with half the cores and at 1.3ghz as opposed to the Snapdragon 800 at 2.3ghz. You might not like it but it's a feather in Apple's cap. They did a great job with their SoC. Again.
133. PapaSmurf (Posts: 10456; Member since: 14 May 2012)
Geekbench results: Adreno 330 (2670) vs. A7 (2557)
Xperia Z Ultra (Adreno 330)
Sony Xperia Z1
Single-Core Score: 914
Multi-Core Score: 2784
A7 beats everything in single core operations, but given that multi-core score is higher, you can quietly shut up now.
117. bigstrudel (Posts: 518; Member since: 20 Aug 2012)
So what you're saying is that it's not a fair comparison because Apple has a better stock web browser?
One more thing? Do you know what an offscreen graphics test is?
136. Taters (banned) (Posts: 6474; Member since: 28 Jan 2013)
Of course. The only truly fair test would be chrome vs chrome since both use the same base. That is how objective tests are done, with the same parameters are at least as close as possible in this case, which would be Chrome vs Chrome since both OS have it.
Safari vs Chrome means pretty close to nothing in deciding which CPU is better at Java because Safari heavily skews it in the iphones favour.
So any fair test would be Safari vs Safari or Chrome vs Chrome and the latter is possible.
And when you do chrome vs chrome guess who loses badly? Yup the iPhone.
Yes I know what an off screen test is and the andreno 330 wins in those so those onscreen results Apple fans like to brag about really do not carry much weight at all.
112. AKumar47 (Posts: 48; Member since: 11 Sep 2013)
if all high end android phones uses ios, damn they will all blazing fast, and imagine if apple uses android os, damn it will crawl... haha! its not about specs, it is how you manage your specs...
its a fair deal, all android phones must use greater specs, great power posses great responsibilities...
120. ThePython (Posts: 887; Member since: 08 May 2013)
Android fans on the Octo-core SIV benchmark: Benchmarks are everything.
Android fans on the Apple A7 chip benchmark: Benchmarks are nothing.
142. barondebxl (Posts: 131; Member since: 04 Nov 2011)
I received my 5s this afternoon and this thing is a beast to say the least.
145. Taters (banned) (Posts: 6474; Member since: 28 Jan 2013)
Go compare it with a iPhone 5. Open the message app, phone app, and safari app and hit the home button after you open each. Notice that the icons stutter or hang when flying into view on the 5s but not the iPhone 5.
Yes, a beast that performs worse than the iPhone without pushing anything more than the iPhone 5 does. Beast eh? Lol
Seems like all of them do this. Very beastly. Lol
147. RomeoJDR (Posts: 244; Member since: 09 Dec 2011)
I have always owned and for the foreseeable future will continue to buy Android phones since I like to have some control of my OS. With that said, anyone who does not get credit to the tech marvel that Apple engineers were able to put out by adding their own core design to the ARMv8 architecture sound like pathetic whimpering fools who irrationally argue how the benchmarks lie and their beloved devices are actually faster. Please stop acting like Apple fanatics who sound so dumb when they try to berate the frequent advancements Android phones release over the iPhone. There was one ridiculous comment in this article. Their is no such thing as future proof in technology! Just wait until the Adreno 330 is released next month paired with the S800. Until then give Apple and their fans a moment in the sunlight.
151. Fuego84 (Posts: 344; Member since: 13 May 2012)
And in the real world, the performances would still be unnoticeable. I choose android because it let's you be you. Change wallpaper, icons, widgets, launchers and themes all you want plus you can do lots more on an android that you aren't allowed to do on iOS. Apple IOS=our way or the highway. Android=have it your way.
166. silencer271 (Posts: 254; Member since: 05 Apr 2013)
Yup but to apple fans synthetic benchmarks mean they are better lol
152. Fuego84 (Posts: 344; Member since: 13 May 2012)
Also with these benchmark test let's not forget that the iPhone 5s doesn't even have as much pixels as a flagship android phone which are 1080p resolution and that would mean the iPhone 5s has a much easier time moving the pixels around giving it faster times. If you put the same small screen that the iPhone has on an android flagship the specs on the Android would tear the benchmarks up.
159. RomeoJDR (Posts: 244; Member since: 09 Dec 2011)
That is why they included the off screen benchmark. No need to try making excuses for the 5s beating the pack in speed. Android still reigns supreme in mobile OS and back to your first post, what seems like a large margin in benchmarks won't make a difference in real world performance.
160. xfire99 (Posts: 827; Member since: 14 Mar 2012)
OMFG, I just notice for first time that IP5S/C doesnt even have HD resolution and only 1136 x 640 resolution. This is a joke! High end phone and with low resolution!!! What year are Apple still living at? No wonder all scores are high!
A premium model, with such high price must have HD resolution ffs and really cheap of Apple.
175. vince89 (Posts: 2; Member since: 28 Sep 2013)
Not really. Newer apple product displays are base lined to provide picture quality dependent on typical view distance. These fall under the category 'Retina Display'. If you take their 1136 x 640 'joke' of a resolution and compare it to the screen size of the iPhone 5s, you would find the more dictating parameter to describe HD, pixel density or PPI, and that the iPhone 5s has 326 PPI. This is far beyond HD when scaled up to 13" or 15". Please see the wiki 'retina display' for more detail.
So, on to comparing PPI for other phones:
iPhone 5s: 326
Galaxy S4: 441
LG G2: 423
Yes, competitors have more HD than iPhone but what's it worth if the human eye can't detect it at normal viewing distance? No one in their right mind would pay for these increased PPI's were they scaled to a television size because no one in their right mind watches their 50" TV at 6" from the screen.
177. aries.phills (Posts: 140; Member since: 17 May 2011)
It's amazing and heart-wrenching to see how iphone5s mops the damn floor with my beloveth S4. Well, that was expected. Exactly what S4 did to iphone5. Guess I'll have to patiently wait for S5 to avange this savage.
181. TechieXP1969 (limited) (Posts: 10115; Member since: 25 Sep 2013)
Well the S4 is over 5 months old. I would expect a newer device to be faster. I am sure the GALAXY Note III will mop the floor with the 5S too. I am also sure the Galaxy S V will birn the 5S too.
178. aries.phills (Posts: 140; Member since: 17 May 2011)
Then again, I think if you put all the iphone5s internals in an android device, the benchmark result might even be more pathetic than mr. Crap.