The physics should at least show a bump up because of the clock speed bump if it is hitting the CPU. If it is running off of the GPU, I would expect a bump also.
If there really is no bump in performance, I'd like to see the battery life comparable to the other systems.
Uh, Tablet will destroy it. The new iPads aren't going to have some magical unicorn GPU in it. The Shield Tablet has the K1, which will not long after be followed up by Maxwell. There's simply no catching NVIDIA at this point when it comes to their mobile GPU performance. It's not a contest.
This only shows that A6 and A7 are not far better than the other phone in the market. Give me reason then to buy one. /Craig from http://www.topreport.org/phones/
The new iPads might not have a magical unicorn, but they *could* have an A8X with 10-12 of these PowerVR cores... I like how there's "no catching NVIDIA" even though they were pretty poor for a pretty long time in their smartphone/tablet GPUs.
problem is that there is no PowerVR 6XT GPU that have more than 6 clusters, i.e. GX6650 is the top of PowerVR GPUs. And GX6650 can compete with K1 only on frequency higher than 600MHz(theoreticly). We'll know for sure in the end of october.
A bigger problem is that nobody is buying the K1 NVIDIA Shield, while tens of millions of people are buying A7 and A8 iPhones and iPads.
There is a concept of profitability and capital reinvestment that shifts where technology goes. Google and even Samsung (more than half of Android) are laser focused on cheap, low end devices, in large part because their attempts to sell expensive tablets (Honeycomb tabs were more expensive than iPads) failed, and even attempts to sell cheap tablets and phones failed. So now they are in India trying to sell $100 phones.
Meanwhile, Apple sells 80% of the $500+ phones in China, 40% of all smartphones in US, and most of the smartphones in Japan.
Anyone who thinks one struggling GPU vendor will save Android despite the fact that the vast majority of those "80% of phones" making up the Android "platform" are using ARM Mali cheap GPUs (including many of the "premium" Samsung models), should prepare to have their expectations dashed.
Apple was winning in smartphones and tablets before they had their own advanced silicon. Now its more of a joke. TI left the OMAP chip business, Samsung is struggling, and ARM isn't catching up to Apple's aggressive support for Imagination Tech - and conspicuously, nobody else is using high end PowerVR CPUs because they're targeting the third world with cheap devices. Didn't work out too well for PCs did it?
You do realize Intel uses the same if not beefier GPU in their cell phone APU's right? Put that together with a real die shrink at 14nm and x86 compute power and it is only a matter of time before Apple puts x86 chips into the iPhone.
Haha you are possibly the most delusional person that I've seen posted here, ever. I don't even know where to start. Calling nvidia a "struggling" GPU vendor or calling google and Samsung "failed" when the android cellphones are outselling iPhone more than ever. Selling $100 phone to a crowd who cant ever afford a smartphone otherwise is desperation? Man its just easy to call you stupid
Here's the reality. Android isn't one big happy family. Guess what ? Android makers don't just compete against Apple - they compete against each other. Samsung's sales hurt HTC.. HTC's hurt motorola's.. and LG's ! WOW Guess what ? They all use Android but only Google makes money from Google Play, only Google makes money from the Google searches and Google ads on Android phones. This is one reason why the best selling Android Phone maker: Samsung is trying to make it's own Android like operating system - like Tizen. See Samsung has this dream to one day not have to kiss Googles ass. They want t run their own OS ( like Google and like Apple ) and their own store.
And guess what ? Only TWO phone makers make more money EVERY YEAR. Apple and Samsung. Most Android Phone makers had flat growth ( see none ).. some even lost money.
Apple losing market share? Technically it looks that way yeah - I'll give you that. But only because the market is growing so fast that it appears Apple's share has shrank some.
The reality is that is only true if you compare ALL of Android against IOS. This would be great if all of Android made a profit together- but they don't. This is not true when you compare each phone makers market share to the others regardless of operating system. Its like trying to compare to sale of matches to lighters. Most of the Android phones in poor Asian/Indian countries are cheap phones with little functionality. This is why Android dominates in those countries. Say what ? You think those countries have 4g? Most have 2g if they are lucky. This is why guys like you don't understand how they can still sell a iPhone 4 over there. It's because an iPhone 4 is as fast as a Samsung G5 ona 2g network. Countries like China block Google services - you think they are getting your Google experience? Haha.. over there the phone is junk.
Apple has outsold itself ( SEE SOLD MORE PHONES EACH YEAR ) every year since the original iPhone ( One year before Samsung started copying the iPhone and before the Google guys got kicked off the Apple Board for stealing technology ) .. This is also why Apple got a headstart on tablet sales - the google guys were kicked off and didnt see the tablet idea coming !
Apple just sold 10 million iPhone 6's in 3 days.. The next best selling phone is a Samsung G S5 and it took 28 days to sell 10 million.
You can knock Apple all you want. It won't stop people from buying their products. The only phone makers that actually matter are Apple and Samsung. Android versus Apple is a joke. Especially when you look at how they all want Apple's profits.
You fan boys on both side crack me up. Does it make your dick bigger if the iPhone is the No. 1 selling phone and you have one? Do you all realize how physiologically damaged you are? Why can't you just buy the phone you like and shut the fuck up? Why must everyone bend and come to the same conclusion you did when purchasing your phone? Fact is, you're a joke. You can sit here and spout all the numbers you want claiming whatever you'd like. The fact is, people will buy the phone that suits them better. I am currently using a Galaxy S4 from my work and an iPad 3. I'm buying the Galaxy Tab S 10.5 and my work is getting me an iPhone 6. They both have their uses. They're both great products. Why must people stick a fucking flag in the ground behind one of these companies who don't get a flying fish turd about you? You make yourself look like a moron. Jebus fucking christ you people drive me nuts.
What have you been smoking? Nice presentation of twisted and inflated marketshare numbers (what is that, 80% of the +$500 market in China? their Iphone copies cover almost every segment and cost a quarter of an Iphone) though.
Regardless, the only country in which Apple really holds ground is the US. In Europe and Asia it will not last and the market share has actually decreased year over year.
You fan boys on both side crack me up. Does it make your dick bigger if the iPhone is the No. 1 selling phone and you have one? Do you all realize how physiologically damaged you are? Why can't you just buy the phone you like and shut the fuck up? Why must everyone bend and come to the same conclusion you did when purchasing your phone? Fact is, you're a joke. You can sit here and spout all the numbers you want claiming whatever you'd like. The fact is, people will buy the phone that suits them better. I am currently using a Galaxy S4 from my work and an iPad 3. I'm buying the Galaxy Tab S 10.5 and my work is getting me an iPhone 6. They both have their uses. They're both great products. Why must people stick a fucking flag in the ground behind one of these companies who don't get a flying fish turd about you? You make yourself look like a moron. Jebus fucking christ you people drive me nuts.
According to what I've read, a 950Mhz GK20A in the K1 chips (either one) produces less FLOPs than a ~650Mhz GX6650. Not to say that that's all that matters, but it's a pretty good indication that GX6650 is quite competitive. I wouldn't jump to conclusions, even if they're "only" using 6 clusters.
With that being said I've often defended Nvidia's Tegra chips. I think they are very good overall performers, especially given the price. They usually undercut comparable ARM SoCs, and as a result provide very good performance (especially graphics) within a given price range.
I'd be a lot more interested, frankly, if Apple used Wizard. Get all those Apple developers onboard with hardware-assisted hybrid raytracing. Might encourage others to develop similar GPUs.
But are we talking about same FLOPs? GK20A produces 365GFLOPS in FP32 at 950MHz. GX6650 produces ~250GFLOPs in FP32 at 650MHz, but in FP16 it produces 500GFLOPs. I don't know how many FLOPs GK20A produces in FP16.
iPad Air destroys Nvidia Shield Tablet with...what? Games for K1 using OpenGL 4.x, not mobile OpenGL ES 3.0. And iOS doesn't have games like Half-Life 2, Portal or Trine 2: Complete Edition. BTW, games with Metal(like Asphalt 8 or MC5) have only a little more particles. Besides, i was talking about hardware, not software.
Uh you don't know what you are talking about. These are synthetic benchmarks. Throw in games developed with metal and they will destroy everything else out there.
You fan boys on both side crack me up. Does it make your dick bigger if the iPhone is the No. 1 selling phone and you have one? Do you all realize how physiologically damaged you are? Why can't you just buy the phone you like and shut the fuck up? Why must everyone bend and come to the same conclusion you did when purchasing your phone? Fact is, you're a joke. You can sit here and spout all the numbers you want claiming whatever you'd like. The fact is, people will buy the phone that suits them better. I am currently using a Galaxy S4 from my work and an iPad 3. I'm buying the Galaxy Tab S 10.5 and my work is getting me an iPhone 6. They both have their uses. They're both great products. Why must people stick a fucking flag in the ground behind one of these companies who don't get a flying fish turd about you? You make yourself look like a moron. Jebus fucking christ you people drive me nuts.
Well, either way. There's nothing that will match the HTC One M8 overall (see http://www.consumertop.com/best-phone-guide/ for example) currently. Apple just can't do it, instead they lag behind.
Physics test is CPU and is very light on GPU. It might be something that Metal could help with, but it's one area where having multiple cores helps (im not sure to what degree it would help in real world games though).
> Adding this to the mix means that the iPhone 6 is just 10% faster than a good old iPhone 5 according to physics test.
That's the 5s, not the 5. The 5s processor, released last year, was the first 64-bit ARM CPU and a true performance monster fully twice as fast as the chip in the orriginal iPhone 5. In terms of architecture the chip in the iPhone 6 is a relatively minor iteration on that epic CPU.
A5, A6, and A7 were all more like tocks. Four in a row would be too much to expect I guess. A8 is definitely a tick, modest performance gains, mostly efficiency gains.
I wouldn't be as pessimistic as this. There are a number of improvements that Apple can add to their existing CPU/SoC that would improve performance substantially without compromising power. Most obvious (but by far the only) improvement would be a really robust and powerful turbo-ing system like Intel uses.
The real reason for the lack of improvement, IMHO, is that it was not necessary this year, so Apple directed priorities in a different direction; most obviously I expect their best and most experienced designers were working on the S1 (for the aWatch) and the A8 was given over mostly to less experienced engineers as a learning project.
Next year (once the A57 devices are actually available, along with Denver and the various ARM server cores) there will be more compelling competition for Apple, and I'm guessing they already have a more substantial CPU/SoC boost planned for that.
The physics tests is one of the test that benefits heavily from multiple cores. Unless Apple decides to go beyond dual-core, their phones will lag in this test. It seems the small improvement in performance is due entirely to clock speed. Hopefully the detailed review will make this completely clear.
If you read the futuremark news release they specifically say that the A7 (and now A8) have poor random memory performance resulting in low benchmark scores. Arranging memory objects sequentially the A7 (and supposedly A8) shows the expected performance increases. However, this is a non trivial thing to do in a physics engine as objects are constantly entering and exiting the scene and you need physical memory space to place them in. This also shows in a few geekbench sub tests and is one of the 'glass jaws' flaws in the CPU design.
This is by no means an A7/A8 specific issue; it's well known that doing heavily random data access is orders of magnitude slower than a sequential array even on Haswell (there's a well known whitepaper in the gamedev community that discusses this). The difference is mostly that the performance "cellar" is a lot higher on HWL than on mobile chips thanks to the higher clocks.
The tests are not quite correct - Samsung Galaxy S5 from February 2014 (not Galaxy Note 4) against Apple iPhone 6/6+ from September 2014. Let's wait for the Samsung Galaxy Note 4 tests.
The javascript test based on Apple Socs always high since it use direct javascript engine on chip called Nitro. It is more like "SSE" for javascript. In other hand for calculation matrix, etc, cpu just run the physics test like other instruction, thats where A7 / A8 loses to other high clocked cpus.
No Nitro is just a JIT version of the Apple JavaScript Engine (like V8 in Chrome). It has nothing to do with SSE for JavaScript. A7/A8 executes more instruction per cycle than competing CPU (in 64 bits in particular).
That battery test is used with a web script of some kind. Depending on bluetooth, wifi, signal strenght, brightness and a few other factors and as well what web browser that is used and how well optimized the browser is for each phone and how power hungry each browser is. That's why I don't trust the webscript battery test so much.
The Xperia Z3, for example, a phone that does extremely well on that test, has a 3100mAh battery, as compared to the 1810 of the iPhone6. So the iPhone6 gets more "active life" per energy unit than the Experia Z3 does (and probably any other phone on their list).
It's a matter of choice what to make of this. APPLE has prioritized thinness and weight over other things --- they appear to have target battery lives in mind, for laptops, iPads, iPhones (and probably watches) and when they hit the target they start shrinking the device. If YOU don't like this and think more battery is warranted, of course you're going to prefer a thicker device. What is NOT a matter of choice but an empirical fact is the power efficiency of the device, and Apple appears to be the winner here with the current crop of flagship phones.
But in fact, from a developer point of view, and if we look at the low level stacks of the operating system, iOS is has much optimized as their SOC: the LLVM compiler rocks, the 64 bits ABI, the implementation of Swift, the work done on vectorization, the display optimization, the fluidity of the UI, animations and physics in the UI... in all these iOS shines...
Ive used every OS out there even had crackberry.. There is no advantage to the mass public to use Android. A smartphone is a internet appliance. The only thing that matters is the apps and both Android and iOS enjoy well over 70,000 apps a piece - and most of those apps are the same. The most popular apps on both are usually the same. A cell phone most important uses- texting, facebook, gps.. ? I suppose now you're going to tell me you can see the difference between you 13 MP picture and the iPhones 8 MP camera? What do you see when you take a picture on a 13 Mp camera picture and view it on your smartphone screen? or 60 inch HD TV? You see a 2 Mp picture because thats all your screen can display ! HAHAAH ! And you really thought you could see !! A 5 thousand dollar 4k SUPER HD TV - the best screen out there - can only display 8 Mp's .. This is why Apple and HTC have not played the marketing hype game of Samsung. A high quality sensor matter much more then Mp's .. Mp's are great for printing posters i suppose..
well said.. if apple had used a 3100mAh battery they would've had beast battery life.. lets not say battery life doesn't matter, but the truth is most of us plug in our phones much more then we admit. I work in healthcare and we have a healthy mix of Android and iPhones. I see alot of phones being charged through the day - both sides
You gotta keep in mind the iPhone 6 has one of the the smallest batterys of the pack.. A8 is very power efficient and the iPhone LCD has always been tweaked to use less power - which you expect as they are doing it with AMOLED's as well. Still.. if Apple had made the iPhone 6 just a hair thicker with a bigger battery - they could've easily had a 14 hour battery life..
Ive been reading at Anadtech for years.. and guess what ? Each product is reviewed without bias. And each product is reviewed not just for the bad - but the good things found in each device. When it comes to being able to come here and read about ANY product - I know I'm getting a real look at the device. If you don't like a company that's fine. This is no different then Ford versus Chevy except for some reason the Android boys get more offensive then anyone. They cry "Apple sheep" yet they are the ones acting FANATICAL and like little kids because post all the stupid crap - that won't change anyones mind at places like this and at blogs - ALL FOR A PRODUCT THEY DONT EVEN OWN ! Wow.. some of those Android guys really need a change in life style !
I thought it was selective to leave out the fact that NVIDIA cleaned house. Don't care about the fact that it's a tablet. Still worth acknowledging that what they bring to the table is epic and is poised to change the landscape of mobile gaming in ways that nobody else in the industry is even close to achieving.
But for the user to really wins, developer must support the new platform and it's particularity (OpenGL 4.0 in this case). Has there already is a lot of different GPU capacity on Android, developers are having hard time to optimize their games for all the hardware.
Contrary to what android users think, all those really high specs on paper, are still lacking in overall performance. It's because the operating system is clunky, and all those extra handwaving features and megapixels that 90% of the world will never use actually get in the way of a great user experience. But, all those Asian companies will sell some phones because they will prey on the ignorant in America. "Our phone has a quad core processor, more is better". Wrong Samsung! They're just cheap Asian knockoffs of the iPhone.
Buy American people, and stop falling for the stupid "the numbers are higher so they must be better" BS. And people say Apple fans are fall for their marketing...geeez
What's American anymore? Apple uses non-American components (this included Samsung) and android was developed by an American company. They assemble the iPhone in China along with just about anything else you're going to buy these days.
Ok. I just looked at Tom's benchmark briefly and noticed that the first benchmark (display brightness) is very misleading.
According to Displaymate, Samsung Galaxy S5 has peak brightness of 696 nit vs Apple iPhone 6's 558 nit. Dr. Soneira remarks that "The Galaxy S5 has the Highest Brightness and Best Contrast Rating for High Ambient Light that we have ever measured."
Now, the S5 has three different display modes (professional, adaptive, cinema) and can be configured to respond to different ambience lights / picture levels at different nits. I'm assuming that these benchmarks are designed to gauge top, peak hardware performance. So why is Tom's G comparing Apple's top nit against the S5's lowest nit figure?
I'm not going to dissect the rest of Tom's G flawed review and I certainly don't want to flame any fanboi war here -- but if you want to demonstrate that the new wares from Apple are indeed better than Android's, you would have to bring legitimate, well thought-out reviews.
Of course it rigged and testcases choosen to highlight the software aspects of iOS rather than try to do real hardware benchmarking. Who do you think pays mr. Anands salary these days :)
Hardware is used through the software. Users doesn't only the hardware part of their phone. They use their phone through two filters: the operating system (it must be optimized for the hardware) and the apps (games, video, music, others...) If one of these two filters is not optimized for the hardware (it could be from a screen resolution standpoint of from a performance standpoint) the user won't get much. That's the biggest force being the AppStore ecosystem: developers implements new iOS feature in droves to encougare users to stick to their apps. It's a really dynamic market.
Anand commented in passing in the iPhone 5s review. The physics portion of 3DMark is multi-threaded and capable of hitting multiple cores, this is where iPhone performance is penalized due to Ax remaining dual-core vs SnapDragon et al. going quad-core.
LOL. I love how people can't accept the fact Apple can do 'something' right. They're not the most valuable tech company in the world because everything they make is 100% crap.
They do get a great deal of shit right. But annoyingly, they get the most simply shit wrong. Like being 2-3 years late in making a phone with a usable on-screen keyboard.
I concur, at phonearena.com, the battery test were OK for 6 n 6plus. They couldn't beat the s5 and HTC m8, but here the 6 plus is beating the s5 n m8 by 3 hours.lol
I agree 100%. Thin is nice, but to make the 6 just thick enough to cover the camera bulge would have given it 30% more battery room, and still be thinner than the iPhone 5c. 3 weeks of standby on the 6 Plus!! I'd liked to have seen the results on that setup.
Their chosen config is cheaper for them tho (better profit margin), and enough for most users (going by our current expectations anyway), and also make the devices obsolete sooner (promoting future sales).
I've got very little interest in an iPhone, but I'd love to see Apple build one that absolutely destroys any and ALL Android phones in battery life, something where they can genuinely claim 2-3 days of battery life.
That would certainly light up a fire under everyone's butt and stop this obsession with thinner and thinner devices. Something 5-7mm thick isn't even particularly easier to hold than a "chunkier" 9mm phone which tapers at the edge or whatever.
Moving away from the first 12mm thick bricks (hello OG EVO) was great, but Apple and Samsung are just over doing it now. At least Samsung still packs in decent sized batteries, something Moto started with the Maxx but has now shied away from.
The IPhone 6 is the hands down the best phone I've ever used. My wife has a cell phone store, and I've used them all. Android for 8 years, too. The 6 is smooth as butter, and a crazy good camera to boot.
As far as I can tell the results of our testing are accurate. The web browsing test has been carefully designed to ensure that all SoCs have some level of sleep in order to avoid penalizing SoCs for being faster.
So it's not the same workload? Maybe detail the battery test some more? For the GPU the perf doesn't seem much. Snapdragon hasn't done a major update in quite a while and the A8 is on 20nm. With 810 arriving the Android bunch should leap ahead. Anyway, as far as i can tell the GPU is some 19mm2 vs 21mm2 last year and the cores with the cache some 12.5mm2 while the SRAM seems to be some 4.5mm2 . One would assume it's the same 4MB but last year the SRAM area looks like it was 6mm2 so it would be just a 25% shrink for the SRAM. The cores do seem rather huge for ARM and i do wonder how much that SRAM helps.
But designing a test to have some level of sleep... is that more for an SoC comparison or is that more for real world usage? I think the important thing here is to consider how these results are representative of real world battery stats.
Your test method should reflect a real world use scenario or something similar. Testing a certain because its good to normalize for other SoCs isn't alone a sufficient reason.
It does, but my point is the test should be designed not just to be fair for SoCs as Joshua has said, but to be representative of real world use. I think in many cases he's pushing too hard for apples to apples synthetic benches that real world use doesn't see.
For example, most users I know use auto brightness. I'd argue that a test setup with autobrightness under CONTROLLED ambient lighting conditions makes more sense than pegging the brightness at 200 lux. Here's a good example. The Nexus 5 is known to have a autobrightness curve on the bright side. Who here actually gets 9 hours SOT on a Nexus 5 as shown by Anandtech? Not many. I'm guessing the autobrightness curve probably penalizes the Nexus 5's battery life. If you look at an apples to apples bench, you might expect the N5 to do well, but in real world use, most users find the N5 to be mediocre at best for battery life. Where's the discrepancy? Probably the autobrightness has something to do with it. That information is completely ignored in the battery life test.
I'd argue that in this case, you have a benchmark that isn't representative of what the phones actually behave like. If you end up being too focused about normalizing everything including SoCs, then you start creating unrealistic benchmarks for the sake of apples to apples.
maybe it's the other way around? Maybe this is the only fai way to compare apple vs apple? Anand has been doing those test from a while, but when it comes to Apple it's never fair?
Auto brightness is a software feature(for the curve). The manufacturer can change the curve with an update, thus invalidating all results previously recorded. On android, you can probably even tweak it yourself. You yourself have acknowledged the curves differ from a phone to phone, how does it then make sense to use auto-brightness, when you get different luminance on different phones ? That's what i would hardly call a fair comparison. What you call real world comparison here can rather be called "ignorant user comparison". There is no point in running tests for that.
The same with the SoC sleep. Would you penalize one phone in a battery life test for being quicker and loading 10 pages in a time it gets the other one to load 3 ? It makes a perfect sense for a device to load a page, put SoC to sleep as we expect the reader to read the page anyway - and only then load another one.
That's probably it then; phonearena probably just let all candidates power through no matter how fast they were, never considering that the iPhone displayed the web page 300 times vs. 175 for some lesser phone. or whatever.
I've been there, and phonearena appears to have their own agenda.
Yeah, but phone arena has some strange battery life testing in there... They runned it down in 5:22 which is less than a half compared to anandtech... What the heck they do with that phone :D.
But for example the samsung galaxy s5 lasts significally more than the samsung galaxy note 3 and I call that strange as well.
We will need to wait for more battery life tests from other sites and somehow make a average from those as battery life testing is naturaly wildly inconsistent across the Internet.
The S5 lasting longer is the improvements made in samoled tech. Also makes sense how the note 3 looks bad on battery life - most webpages are mostly white with black lettering. That much white for an older samoled puts the note 3 in a disadvantage- it basically gives you the worst possible scenario for battery life.
The note 4 should do much better with all the improvements in samoled tech shown in the S5's numbers.
There hasn't been much of an improvement in Samsung's AMOLED screens as far as efficiency goes. AMOLED still has serious problems in that area. It's why AMOLED is about half as bright as the average LCD screen.
Displaymate disagrees, check out the Note 4 AMOLED tech. Given the current rates of progress it could very easily be LCD that has the efficiency problem next year. It has been confirmed the Apple Watch will be AMOLED, I would love to see Apple phones and laptops with AMOLED in the next few years.
I question battery results as well, nobody no matter what is getting 8hours screen on time, much less while wifi browsing, out of their Nexus 5. I posit (well, bet) that these tests were run on 4.4.1 or 4.4-- battery lives dropped significantly, on the order of 35-55mins screen on time (thanks "Android System") when 4.4.2 came out
The battery results should be accurate. It's not a question of their testing methodology, but a measure of their precision. Anandtech tests all their phones the same way (high precision), ergo, whatever differences that arise from one website to another, assuming they also standardize their testing methodology, would be from the testing method (function of accuracy). Therefore, any attributable difference from one website to another is due to how they test the phone, however, relative standings for the phones should still hold (e.g.: % difference in battery life should still hold per each phone per website, assuming proper standardization and control of variables)
Its relative, but I doubt it scales that well. CPUs scale differently and also different generations of CPUs and network chips will act differently when you compress the time scale. Race to idle benefits the faster CPUs in the end.
So its hard to take an Anandtech benchmark and say that a phone that lasts 10 hours will last twice as long as once that lasts 5 hours in their benchmark.
Anandtech's numbers are good for comparing against other phones, but in the end you have that difficulty in translating those numbers into real world usage figures. TBH this is where GSM Arena's endurance rating comes in. Its not the cleanest test method, but its the best we have right now and the only site that tests idle use. GSM Arena's test results are relatively consistent with what you hear out there. The N5 does well on AnandTech but almost every user agrees its mediocre at best in performance. Most users see S5s and M8s outperform their Nexus 5s in typical daily use scenarios in terms of battery, which is reflected very accurately in the GSM Arena benchmarks. The 1 or so hour difference in WiFi life from Anandtech's site between the N5 and other flagships isn't very realistic I don't think. I also find my iPhone 5 can easily outlast my Nexus 5 in screen on time, and the difference is pretty startling.
You shouldn't treat these tests as something similar to real use. Loading one web page after another is VERY different from browsing (where you actually scroll on pages), since it just downloads and renders a page and then sits there sleeping until the next load. And the phones are pristine, with no services running in the background, no accounts being synced...
But this is the only way to have a controlled setup that creates scores you can compare phones on.
I can get 4-5 hours under Wi-Fi in real world conditions with brightness on auto and all kinds of background stuff, I'm not surprised their controlled/standardized test can double that on lower brightness and a clean environment.
As others said tho, the point isn't to show you real world results anyway, the point is to compare hardware on as level a playing field as possible in a repeatable and precise way.
Anandtech's numbers are only good for comparing against Anandtech's numbers, but that's true of many precise and repeatable benchmarks or tests on many site/publication regardless of the product we're talking about.
I don't compare Tyll's FR graphs for headphones from those from Headroom and I don't compare Top Gear's tests with Car & Driver's. I still value both, but cross comparison is pointless.
Do you have a scientific assessment of why the PhoneArena measurement is more valid and accurate? Or is it just, "I want the iPhone to be worse."
We do have a difference. Not the same as rigged. Plus, even if one is rigged, how do you know which one was? I plan to wait for more info as to where things stand.
It depends on your usage. I have a Galaxy S5 and I may be use it different than you. Did you know the iphone can last 5+ days? Some android phones can only last 3 hours
As far as Phone Arena is concerned, one test is statistically invalid. Poor math, really poor math. Anandtech shines again on their great testing abilities.
And? They don't disclose their testing methods and every device runs differently.
Furthermore, it's been noted by people who have an iPhone 6 while at crowded cellular locations like Disneyland that through extensive usage they got 1 full day and had over 30% remaining when they plugged in at 11pm. The 6+ got through 2 full days with 13% left.
I don't trust PhoneArena. They simply don't disclose their testing methods. They don't talk about screen brightness, type of connection, etc. It's honestly ... bad journalism and reviewing on their behalf. It's why everyone trusts ArsTech and AnAndTech over every other blog.
Another thing that I find none of these other site's battery tests seem to account for is actual screen brightness. Anandtech has always used 200 nits as it's baseline brightness for all phones in its testing. Some sites don't even mention the brightness they use for testing while others say 'we had the phone at about 50% brightness,' which is also kind of useless. The iPhone 6's get over 500 nits, where as some other phone might get 400 and some odd nits at max brightness and the 50% level of each is not comparable!
If there's anyones battery life #'s I'm generally using as a good guage (because it's always a YMMV situation) it's AnandTech.
A PhoneArena writer here. We do disclose our battery testing method: all our tests are run with device displays set at 200 nits, and we run a script that simulates real world test (we have looked at average use stats and how much activities draw, and simulate this via a web script), looping it at a set interval. Test results are fairly consistent and in the iPhone 6 case we have run the test more than three times to ensure the results are indeed correct. As to the difference with AT's results - I'd just say different testing methods.
I assume older iOs devices were running on iOs8 in these test? What about Android devices? Are they all using Chrome or whatever browser the phone maker has provided?
"Normally an 1810 mAh battery with 3.82V nominal voltage would be quite a poor performer, but the iPhone 6 is a step above just about every other Android smartphone on the market."
Yeah and just about every other android phone on the market would be 1080p and not some goofy scaled 1334x750. Screen resolution makes a difference: see LG G3.
How come these charts exclude the Note 3? Apple force your hand? Makes the most sense to compare these two devices...
Yeah, it's Apple's fault for not putting in an unnecessarily high PPI and taking the performance and battery life hit for it. I'm not sure what you mean by "goofy scaled" either, it's the native resolution which is what you see. An *app* could be scaled but that's a different thing obviously.
The 1334x750 resolution is the native resolution of the panel and is the native resolution that iOS is rendering at on that device. Not sure what you mean by "goofy scaled".
In the case of the iPhone 6+, the display is 1920x1080 but iOS is rendering to an intermediate frame buffer that is 2208x1242 and then scaling down to 1080p at display time. This may qualify as "goofy scaled", but that wasn't what you were referring to in your post.
I have my 5s by my bed. It plays fine to a bluetooth speaker in the bathroom while also connected to and being controlled from my Pebble over Bluetooth from the shower. This feels like a really long way away, I'm always surprised but it works. I also annoy my wife by playing my podcasts at her to the bluetooth in her car from my car when we're both driving in traffic together.
I don't have much frame of reference for comparison, but I've been mightily impressed with Bluetooth range.
Samsung is great at marketing their phones through specs, not so great on choosing hardware that actually gives superior performance. Also, consider that iOS is optimized and designed specifically for the hardware used in the iPhone.
It has to do both with hardware and software. On the hardware side, the thermal headroom and maximum power consumption is very limited in a phone. It also has to do with the CPU architecture.
On the software side - Android simply is slower, and multi-core programming is very difficult, and in many cases theoretically impossible (not all problems can be parallelized).
That was just some of the reasons, explained very briefly.
Very little non-graphics/scientific software scales well to many cores. The iPhone's poor showing in the physics benches is probably due to that test being able to scale to the extra cores common on most Android devices. Extra memory only matters for heavy multi-tasking. I suspect a test that loaded switched between enough browser tabs that the iPhone and budget android phones were forced to unload/reload due to memory pressure but android flagships weren't would generate very different results.
That's like asking "How can a 4 cylinder engine be better than a 6 cylinder engine?"
The count of cores has very little to do with how a SoC performs. Also note that this is NOT as in the Intel days when everybody just bought the very same chips. Apple's SoCs are utterly different from what others come up with and Apple is one of the very few companies that not only license ready ARM designs but have an actual license to design their own cores (and whatever goes between them).
The difference would be displacement. If I had a 6 cylinder 2L displacement for 400HP versus a 8 cylinder 3L displacement for 400HP, i would choose the 6 cylinders for the efficiency.
You could also extend the metaphor to the car is advertised as 400HP, but brake horsepower could actually be 250BHP vs 300BHP.
Apple makes up for the frequency difference with more IPC, and most smartphone workloads aren't threaded enough yet to notice a difference beyond dual core.
If you go back to last year's iPhone 5s review, you'll find this tidbit at the beginning:
"Then there’s the frequency discussion. Brian and I have long been hinting at the sort of ridiculous frequency/voltage combinations mobile SoC vendors have been shipping at for nothing more than marketing purposes. I remember ARM telling me the ideal target for a Cortex A15 core in a smartphone was 1.2GHz. Samsung’s Exynos 5410 stuck four Cortex A15s in a phone with a max clock of 1.6GHz. The 5420 increases that to 1.7GHz. The problem with frequency scaling alone is that it typically comes at the price of higher voltage. There’s a quadratic relationship between voltage and power consumption, so it’s quite possibly one of the worst ways to get more performance. Brian even tweeted an image showing the frequency/voltage curve for a high-end mobile SoC. Note the huge increase in voltage required to deliver what amounts to another 100MHz in frequency."
Mobile devices just aren't very good at parallelization yet. You're constrained by the amount of power you can draw and the amount of heat you can dissipate. Additionally, a lot of problems that we solve on our mobile devices (email, web browsing, etc.) just don't break up into parallel problems well. However, I expect this is why the Android phones all crush the iPhones on the physics tests--those problems CAN be solved by throwing more cores at them.
Performance with multiple cores all depends on the task at hand. Think of building a house. There are a lot of tasks going on at once and so having more people/cores to work on that will get it done faster. If all you have to do is change a lightbulb, then having 8 people/cores will just result in one changing it while seven look around doing nothing. Having two really fast cores so you can do that task faster (while one handles background tasks) will work better.
Software design is much further along than it was a decade ago when it comes to multi-core workloads, but there are still many tasks that just don't scale well to multiple cores. This is what hampered the PS3 as well, in that it had tons of cores but was harder to program for than a system with fewer, faster cores. There is no single right choice, as each has benefits and drawbacks, but it's why you'll see different systems win different tests as they're optimized towards different workloads.
Basically those extra cores in an 8core system are wasted, designed to fool ignorant people into thinking they are better because it has bigger number.
The thing about engineering, is that a bigger number isn't better.
That's not correct. For example Android doesn't even allow you to run network stuff in the same thread as UI stuff. Most Apps are actually multi-threaded without you doing anything nowadays, because the libraries and APIs you call already do the whole scheduling stuff for you.
Also, even if a App is single threaded you usually run several Apps on your phone at the same time, doing background tasks, you have tasks checking your connections, GPS, audio etc. All things that benefit from more threads.
I'm surprised all these people claiming their phone(s) have 8-octo Cores doesn`t understand the concept behind the implementation of the BIG.little configuration. Most implementations actually doesn`t use all 8 cores. At any one time, EITHER the `BIG`(4) cores are processing heavy threads or the `little`(4) cores are processing light threads. So technically, all these marketing gimmicks of so called `Octocore` SoCs are in fact `Quadcore` systems.
Intel i5's with 4 cores beat out AMD 8 core CPU's all day long, with a much better IPC. 2 big cores make a lot more sense in a mobile device. It's had to get software that takes advantage of 4 cores in your desktop computer, let alone your phone.
"In in GPU benchmarks, we generally see a pretty solid lead over the competition for the iPhone 6/A8."
Actually, it looks to me about on par with its contemporaries (SD 810, Exynos 5433)
Also your battery tests need revising as they don't reflect real world usage at all. I can tell you from experience the Moto G is not in the same league as the Note 3 in terms of battery life.
Hi there. I know this is Anandtech and you're few steps above the others, but may I kindly ask you to check the performance over time of this new iPhones? I heard that what's special about this new chip isn't the increase of performance, but the power/heat management (maybe I even heard about it from this site). Thank you!
Regarding battery life: don't you need to compare "apples with apples" (as it were)? You can't just list a device's name: both the size of it's battery and, more importantly, the weight of the device, plays into this. Yes, the Ascend Mate 2 has a longer battery life, but it also weighs in 202 g as opposed to the 129 g of the iPhone 6 and the 172 g of the 6+). So while you get a longer battery life with the Mate 2, you have to carry around the extra weight. This is an important trade-off and I think it should be in there with the other stats.
Not really. If what you're concerned with is the absolute life of your phone throughout the day, and that's your main consideration when buying a phone, it's likely that you'll eschew other factors just to get the most life out of the phone that you can. A lot of people online seem to cite the swappable battery and SD card in the Samsung as reasons to buy that phone, regardless of the merits of other phones. If you want your phone to last all day, the power to weight ratio probably isn't going to impress you much.
I don't think I've every seen battery life compared like that. If no one wants to read it then it is probably a waste of time to create, I'd rather websites I read publish interesting things. No one is stopping you from publishing that article though.
It's pretty much pointless comparing an Ascend Mate 2 with its 6.1" screen to a small 4.7" iPhone 6 anyway. They aim at different segments of the market. I wouldn't personally even consider a 5" or larger mobile phone, they are simply too big to be very useable, may as well get a tablet. Compare it to an Xperia Z3 Compact instead, as that's a similar size phone, aimed at the same market segment. Gives a much more interesting result. Unfortunately Anandtech doesn't appear to have tested the Z3 Compact yet, so how it would compare in those battery life tests is yet an unknown factor.
FACT: Apple has been forced to copy Android in style and size for years because people abandoned their tired, moribund and fossilized devices for superior and innovative Android devices.
Steve Jobs said no one should want a 7" tablet until everyone went and bought Android devices forcing Apple to copycat with the iPad Mini. Apple didn't think anyone wanted a phone screen larger than a business card until they all bought Androids thus forcing the arrival this week of the iPhone Galaxy and iPhone Galaxy Note clone phones.
Swipe down notifications that don't interfere? Copied from Android and WebOS. Siri? Bought and ruined from a private developer; Google Now crushes it. 3rd-party keyboards? Welcome to 2010, iChumps! Widgets? Welcome to 2009 except you can't place them on your home screen. Live wallpapers and hidden icons? Maybe Apple will get around to copying those in iOS X in 2016. Who knows.
Apple lacks creativity and honest people acknowledge it. Steve Jobs gets credited as an innovator when all he was was a huckster who'd spot someone else's tech, polish it up nicely, then slap a gnawed fruit logo on the back, charge a premium price and wait for the rubes like Jim Smith to hand over their cash like the good iSheep they are.
But after that initial iteration, Apple is incapable of actually innovating something new. They literally cannot make a product until someone else shows them how and they copy it. They are also unable refine things because they believe to improve is to admit something was imperfect the first time. (This is why QuickTime 4 had a legendarily terrible UI that was never changed through QT7 a decade later.) All they can do is make things incrementally thinner or faster but it's just minor refinements since they can't invest their way out of a wet paper bag.
For all their squealing about Retina displays, they never even had a HD display until now; 8th time is the charm, though you need the iPhone Galaxy Note to get the 1080p that many Android users have had for at least a year and is now considered bare-minimum spec. At the rate Apple drags along, QHD screens should arrive in 2018. Maybe. A graphic went around after the reveal comparing the iPhone Galaxy to the Nexus 4 from 2012. Exactly.
TL;DR: Apple is finally (and desperately) copying where Android OS and phones have been for the past 2-4 years.
You've been soiling yourself in public with your iTard blather all through these comments, but you totally beclowned yourself here, you sad pitiful foolish fucking cunt.
Your ranting insult makes no sense at all, you are wasting space, oxygen, time, help make the world a better place, go and die.
Go back to your den and still dream about your device to be the best. your worst nightmare is here in front of you - apple pawned your super spec'ed sammy phone. the benches performed by anandtech is right in front of you.
Where in the world did that diatribe come from? Much more to a display than resolution, especially at a million pixels + @ <6"! U sir have spent WAY to much time, alone with your spider named Andy and phone named Android. I use both and am quite lost with what you've added to a discussion based on this year's A8 performance. Not features. iPhone, first out of the gate in 2007. Android 2008. First with 'retina' or HiDPI displays. Check. Now an OEM or two has decided we need qHD on our 5" phones when most are quite content with their 1080p 65" televisions in their living rooms? I'm all for HiDPI, but you can go too far thus exhausting battery life. Android has copied, Apple has copied, as has Windows. The latter two of high have matching and wide spread vertical and horizontal aggregation and integration with your 'other' everyday tools like the laptop or desktop. Who's carrying s Chromebook these days? And if you believe the A8 second gen 64bit mobile SoC coupled with the new APIs, XCode and it's free update along with the marriage with OS X AND Windows 8.1 with continuity and handoff with OSx specifically, it's not JUST the phone! I'm slamming this response out on my year old lowly A7 Air with half the cores running at half the speed and using a third the RAM as my Note 3 while sporting a HELLUVA lot more pixels! Wonder why it seems ...after a year of using both everyday, that regardless of the task, gaming, photo/video manipulation, or productivity...when you can find an app in parity, let's use Asphalt8. Doesn't matter which device I use. Note 3. Air. Second gen iPad mini, or my 5s...not a single one fails to SLAM the performance of my Note 3. Graphically, efficiency (this drains my note very quick!) wise and the Apple support network, vertical and horizontal support in my personal and professional life and NO Need to use KIES to organize, sync and support my media, I can count on update for a good two to three years well ahead of the next update necessary and unbelievable resale value! Did I see ten mill sold in one opening weekend this morning? That's. Amazing!
Sorry, but your comment deserves to be smacked down and hauled out to the trash. Anand, who's now passed the torch to Brian have kept ALL of us informed for the better part of a decade now in the mobile sector all the while maintaining an 'objective' testing standard, call outs to companies 'cheating' their benches, and educating a LOT of us that don't have the time to relearn/graduate silicon, electrical and computer engineering degrees in our 40s with families.
Again, while doing all for 'nothing'. YOU, me and the rest of the world are allowed to learn for nothing and enjoy their reviews. It wasn't but six months ago five or take they were praising Samsung's latest SupAMOLED display technology now meeting the best performance of LCD and IPS/LED. Displaymate, another reputable site seems to agree.
Contributing is one thing. Even 'questioning' a result. But what you've said is almost as ignorant as one can get.
Yes, I actually saw the first Android prototype; it looked a lot like a Blackberry, even in UI. Credit to Google though for immediately recognizing the potential of Apple's new hardware and UI design where Nokia, RIM, MS and others scoffed at it. In 2008 they had it reworked.
I had to wait for Android 1.5 beta (released 2009) to get a software keyboard needed for the EBook reading device we were designing that used Android.
Anyway, everybody borrows from each other all the time.
In the Plex has said that they had TWO phones in the works, one, the blackberry-like interface, and one, called Dream (iirc), that was only used a touchscreen for input.
2 comparatively fast cores can be faster than 4 comparatively slow cores. And the RAM doesn't matter as long as whatever job you're doing doesn't require more than you have.
No test is perfect. However I really like this test as it shows what a realistic interaction of the phone would be. To say that the new iPhone is archaic or antiquated is amiss. The phone by most modern standards will be a good experience and apples continued improvements will be a boon to its users. KAI-ZEN
I think something is wrong with the iphone 6+ benchmark. I downloaded the gfx app and ran the manhattan on screen on my iphone 6+ and I got a score of 31.2 fps. But the app was crashing a lot. I think the app needs to be updated to provide an accurate measurement!
I'm still surprised at how dense people are over hardware specs.
It's a comparable concept to cars. You could have 1,000hp in a car, but if it weighs 10,000lbs it's not going to perform as well as a car with a 1/4 of the weight and half the power.
Another example with cars, you could have two cars with exactly the same specs and weight. But one is 4 wheel drive and one is 2 wheel drive. The 2 wheel drive version will be faster, because of higher efficiency in the delivery of power. Which is comparable to this situation, in that apple designs their software and hardware together to boost efficiency. Most times android phones just don't have the luxury of doing this, so they have to take a premade software package and adapt it their hardware after the software has already been designed. Which leads to the need for higher specs to get the same level of performance.
Common sense (or maybe I should call it uncommon sense?).
Yeah, I'm not sure you're right at all. I know that this has been the "common sense" view, but I don't think the evidence supports that conclusion. In particular, when you normalize for screensize, battery, and process node you typically see similar results (for battery----third party apps vary in quality far too much to make useful comparisons). In this case, apple has beaten everyone to a completely new node, so I'd certainly hope they have a lead over everyone else (save that one huawei device I'd never heard of). IMHO, where "controlling" the hardware/software comes in most handy is performance of drivers. Apple+Imgtec have, hands down, the best drivers around.
"The 2 wheel drive version will be faster, because of higher efficiency in the delivery of power."
Not if the 2WD car can't get its power to the ground and it just sits there spinning tires all day.
Apple's integration would be like the manufacturer of the performance car cherry-picking the (or designing their own) grippiest tires for the car, implementing an amazing traction and launch control system, and their own designed dual-clutch gearbox.
The Android implementation would be a bigass twin-turbo LS motor making 1500hp on an average platform running regular tires with no traction control and a manual gearbox. It'll just spin through 4th gear.
It's vastly more powerful in terms of raw output, sure, but the chassis can't utilize the power for any useful work.
BTW when it comes to higher powered cars, it's more efficient to put power down through 4WD than 2WD...cuz you have twice the amount of rubber applying tractive force to the road, despite the inherent drivetrain efficiency losses. A more efficient drivetrain doesn't do jack squat if your tires are melting.
"Another example with cars, you could have two cars with exactly the same specs and weight. But one is 4 wheel drive and one is 2 wheel drive. The 2 wheel drive version will be faster, because of higher efficiency in the delivery of power."
Not relevant to the discussion, but I can't let you slip this by. :P Two cars with the same HP at the crank, the 4wd will be faster because of the available traction at all 4 wheels. Ask anyone with a STi or an EVO.
I'd say battery life is OK, but not stellar. If they'd not been so keen on thin, and just bumped up the thickness to match the camera bulge and put in a bigger battery, I think we would have seen super stellar battery. Then the iPhone would be a MUST have. I calculate 3 weeks standby on the 6 Plus. WOW!
Rather have thinner than a battery that may last me 1.5 days instead of 1. I'm going to charge it at the end of the day either way. It honestly doesn't matter as long as it lasts the day of heavy use, which it is confirmed to be doing as long as you aren't blasting it with Bioshock for 4 hours straight.
gfxbench.com is working now. Did you notice that both iPhone 6 and 6+ are using iPhone 5's resolution(1136x640) in Onscreen tests? That's why they have over 30fps in Manhattan Onscreen and 18-19fps in Manhattan 1080p Offscreen.
Scaling these days seems to be pretty negligible in terms of performance; all digital monitors and TV's do scaling whenever your input resolution doesn't match the native display of the panel.
It's a lot less work to render at a lower resolution and upscale to native vs. the other way around.
Ok you have obviously never actually used a WP8/8.1 device. The speed difference alone was enough to make me give up on my Nexus 5 after 3 months & go back to my Lumia 920. It was much faster to load & use apps in general not to mention the complete & utter lack of any OS stuttering which I was connately hounded by on the Nexus 5 no matter what I did (enable ART, Cyanogen, AOSP).
The consensus is Lumia's have the BEST cameras in a phone out there. They also happen to have the best Navigation suite (HERE Maps, Nav+, etc.) with the most accurate & up to date maps & in my experience the best usability (live tiles, Glance Screen), the best outdoor readability & Incredible build quality (Go youtube Lumia 920 Hammer Test).
Lumia's are great phones. They are fast, like iOS, unlike Android. The cameras take awesome pictures. The shot to shot time, focus speed, burst speed, are appallingly slow compared to even the iPhone 5s. This may or may not matter depending on what you take photos of. But with a Lumia you may miss the shot. (Eg the Bee flys from the flower before you can get a second shot in) Also, low light performance at Night time is quite terrible on some models.
Is there a video you can point me to that will prove your point? My experience is that win8 performs very similarly to android on the n5 (in particular). The input latency tests I've seen also tell the story that win8 and android have similar results.
No I don't have any videos to point to but just go to a store & try one for your self you'll see the speed difference (at least compared to the Nexus 5 class hardware)
I have, that's why I said "my experience". It's entirely possible all the win8 devices i tried were crap, so I asked if you have something I can refer to verify your claim
I got the 6. I had a palm pre and then all the htc evos and ones. I've never liked Apple much but liked apple care. When the screen got big I decided a year of Apple was worth a shot. This phone is great. They probably should have made it slightly thicker and put in a massive battery. But it's a great phone. Apple haters should give it a shot.
guys... seriously, just check the Bench part of this website, your Nokia Lumnia's are in there even if no full review of them has been done. Seriously, just check the bench site - it's all there.
Yes but the fact that there are no WP8/8.1 devices included in these tables is kinda bias by neglect.
For a lot of people comparable performance is rightly or wrongly the biggest determine factor when buying a device. So when a entire class of devices unmentioned not only do the people looking for an honest comparison of ALL there options lose out but it also works to downplay the the platform to those reading it. Out of sight out of mind.
Guys for those of you saying that the article seems to make Iphone 6 better than it actually is and particularly in light of the recent site's founder move to Apple (unclear if Anand still owns Anandtech but likely as he did not said that he has sold it). Lets be fair - the test presented are typical of what the site was presenting (in terms of tests used) before Anand left. Yes, I personally felt that the Apple coverage under Anand was biased but not in the sense that any of the tests were manipulated but more in the sense of what tests were used.
I always felt they were done more from the point of view of an Apple fan be-lovingly playing with its shiny new toy while trying to be objective regarding to the actual data recorded during the tests. For example, the coverage may include how fast you can open a web site but does not tell you how much the relatively smaller (4 inch until now) screen sacrifices in terms of content vs. a 5 inch (Nexus 5) or 5.7 inch (Note 3/4) phone. The default browser was always used and many times such test become more of a browser test under different platform than a device test.
Where I see Iphone 6 lagging is in terms of memory. In the full review I hope we can see tests of how many safari tabs one can open before the OS starts reloading them every time (in other words how good is multitasking under iOS 8). The same test should be replicated under Android L using Nexus 5. Also it would be great to include in the full review comparison of Android L vs. iOS 8 of things like a) sharing content to Dropbox, Google drive, different e-mail ( i.e. default and Gmail). b) uploading and downloading files from NAS (you can use Note 4 as a comparator) c) comparison of what can be done in Anroid L (but not in iOS 8) and vice versa.
Multitasking is better than on Android. I can run 8 or more soft synths at the same time on my ipad air connected to a controller without problems. Apple only must add multiwindow that is in the works. Maybe ios8.1.
So one thing I noticed the reviewer did not mention is information about each phone. Are these with the current OS/browser level available? Or the version the phone shipped/reviewed with? I just ran Kraken/SunSpider on my N5 (ART RT). My results for SunSpider 1.0.2 were 784.3ms total and 6694.1ms Total for Kraken. As Anandtech knows for performance ART>Dalvik maybe some sort of information about that in such a heated post. Or maybe there test was with ART and I have a great N5?
I am not sure, but that is where it would be good if we knew if the device scores are from initial review. Many years ago when I would go to THG they would show benchmarks for graphics card drivers. Not sure if they still do as it was early 2000s, but it seems like Anandtech might want to look at a similar article with different variations of the same device.
Sure, the version of chrome would make a difference. So, that's a fair point. This is yet another reason why I wish AT would use a third party browser (my preference would be ff, but any would do) so we can actually get useful data about the hardware instead of having to factor in oem optimizations.
Can I just say that while I generally respect Anandtech's articles for detailed analysis of technical details, can we stop trying to release these piecemeal articles? If you truly want to write an article from start to finish with the typical analysis that Anandtech provides, then do it. If it takes 2 weeks, then wait 2 weeks. Please don't do these preview, then follow up then slight follow up articles.
I can point to a bunch of examples (ahem GS4 Part 2 review?) where reviewers promise to follow up and never do. There's plenty of examples from the computer hardware world too where you guys promised a DDR3 roundup article, an i7 overclocking article, and a lot of other stuff following a single review article so we kept on waiting and waiting. Either do it now or never.
This. I myself am checking the website a couple of times a day (work-related) to see whether the iPhone 6 review is ready, but this half-baked piece just looks amateurish and out of place. Anand has set the bar high, and you guys with Andrei Frumusanu (I loved his last review, hope I got the name right) certainly have what it takes to keep it up.
There is something absolutely crazy going on with the web browsing battery life test.
On no planet can the Nexus 4 ever achieve 6+ hours of screen on time. (Source: I own one)
There is zero chance that my LG G2 can ever get 10+ hours of screen time.
I get that the test is supposed to give a relative idea of battery strength. However, a better representation of an actual use case would be greatly appreciated!
Seems like those benchmarks gauge browser performance, not CPU.
Microsoft for instance claims IE11 (139) performs better than Firefox 25 (205), Chrome 30 (195), Opera (188) on a Dell Optiplex with a 3.0 GHz Core 2 Duo Intel processor, 4GB RAM, Intel Integrated Video. That's about 40% difference in performance.
I also just ran Octane test on my rMacbook Pro (mid 2012) -- Chrome 37 (24530) outperforms Safari (20853), Firefox (20718) by about 20%.
Probably because this test performs CPU intensive operations but there are many caveats. As I understand, operations in this test can't be multi-threaded (hence CPUs with smaller number of large cores do better in it). Secondly it is very browser dependent so the same CPU (phone) can produce vastly different results using different browsers. Because of that it is mostly useful in just two scenarios: * running the same browser on two different Android phones * running the same browser (same iOS version) on two different iPhones
Cross-platform comparisons are not very useful (at least not as CPU comparisons)
Sure, I can see that it'd be useful look at it within one product line/browser, but, as you indicated, it seems nonsensical to use this composite "web/javascript benchmarks" to gauge "CPU performance."
I just did a quick test of my own on my rMacBook 15" (mid-2012, 2.3Ghz Core i7)
So they are obviously not browser neutral, even within one same platform, so apparently they are not just measuring "CPU intensive operations." It's one thing to say the iPhone 6's browser or javascript engine is faster, but to say the A8 is faster than those used by competitors is laughable.
Sure, my limited sampling is by no means representative, but according to Microsoft's recent announcement, IE 11 (139) outperforms Firefox 25 (205), Chrome 30 (195), Opera (188) on a Dell Optiplex with a 3.0 GHz Core 2 Duo Intel processor, 4GB RAM, Intel Integrated Video in SunSpider scores. That's about 30% difference in SunSpider performance.
So these benchmarks are highly dependent on webbrowser/javascript engine implementation (duh!) -- and they make no false pretense about measuring CPU performance as AnandTech does here.
Hey Josh, a forum member noticed that the 3DMark Unlimited "Physics" graph appears to be off for the NVIDIA Shield Tablet (it appears to be using the Shield Portable data point instead): http://forums.anandtech.com/showpost.php?p=3673681...
On a side note, have you taken a look at the actual gfxbench website for data on iPhone 6 and 6 Plus? It appears that these phones are consistently rendering at the lower iPhone 5S resolution of 1136x640 in the T-Rex and Manhattan On-screen tests (and most other reviews reported those numbers too): http://gfxbench.com/compare.jsp?benchmark=gfx30&am...
I have repeatedly run both the Manhattan onscreen and T-Rex onscreen tests on my iPhone 6 Plus and have received much better results than in this test. 37.1 fps and 52 fps respectively. I'd suggest to the author they re-run the benchmarks again.
Did you notice, what resolution is used? I know there was 1136x640 resolution. If it was native resolution, there wasn't be much difference between offscreen and onscreen results.
Is there a reason you aren't using the basemark battery test? I'd asked about this before and I was told that you used it often (and basemark clearly works on ios).
I'm surprised all these people claiming their phone(s) have 8-octo Cores doesn`t understand the concept behind the implementation of the BIG.little configuration. Most implementations actually doesn`t use all 8 cores. At any one time, EITHER the `BIG`(4) cores are processing heavy threads or the `little`(4) cores are processing light threads. So technically, all these marketing gimmicks of so called `Octocore` SoCs are in fact `Quadcore` systems.
its the gx6650 the a7 used a 4 cluster rogue gpu apples added 2 more clusters to gain the 50% jump in performance, it's the exact same architecture and same gpu just with 2 more gpu clusters
So you pick and choose the benchmarks that show iphone6 ahead to say it beats other Android phones? Where's antutu and geekbench, oh wait those show phones such as 6 month galaxy s5 faster than iphone 6 plus lol.
They are so Apple bias, that's why the Indian guy that started the site is now working for Apple.lol. Tim Cook was like ' good job Asian man for making us look good on your site".
I hope I am wrong but it is possible that the author's choice of benchmarks was heavily influenced by the fact that Anandtech Inc owner now works for Apple.
Arstechnica have just released their own battery testing. The author notes day to day use was similar to a 5S and the android flagships actually outlasted the 6+ in their own metric test. Interesting. http://arstechnica.com/apple/2014/09/iphone-6-and-...
The author of the Arstechinca author notes this about their battery testing, "But the short answer is I know our current test is limited and we're working on better ones, I just keep1 getting2 sidetracked3 by other stuff4."
They also note the battery life is better than the sheer numbers would indicate:
"Chalk it up to something our Wi-Fi test can't really measure—Apple is pretty good at tamping down idle power usage, at least when your signal is good. I've had an iPhone 6 Plus sitting on my desk all day, buzzing at me only sporadically as new Twitter and Messages notifications come in. It's been sitting at a 90 percent charge for at least the last two hours"
Because real people don't browse for 10 hours straight, they are doing stuff in-between. The less power consumption happens during those periods, the longer the real-world battery life is.
You should include a benchmark chart showing how many browser tabs can be opened without thrashing and reloading when switching between them. Maybe this would finally convince Apple to put a modern amount of RAM in their new devices.
Maybe you should devise that test and post your results? 1) Design a fake website with controlled webpage sizes 2) Load your webpages in tabs, checking everytime if some of your others tabs had to reload 3) Register the number of max tabs open before reloading become necessary 4) Repeat test multiple times to try to average result 5) Repeat test on different devices and platform 6) Prepare illustrations of your results 7) Write article analysing your results 8) Publish
or
1b) Admit that you don't have the patience, time or technical skills to do any of those things, reaching the conclusion that you have no standing to demand for an engineering spec whose implication you cannot judge.
of course there is always 1c) Keep trolling... (optionnal: put a tin-foil hat before adding something about Anadtech being completely controlled by Apple)
I'd love to see them address that. It's a big failing, and it has left me scratching my head about how Craig, Jony, Phil & co. use their phones, if they haven't been frustrated by this already to the point where adding RAM was a no-brainer. It was a big missed opportunity to further improve the experience of using these devices.
Since a huge proportion of 6/6+ buyers will be coming from the iPhone 5, it would be useful if that phone's performance was always included in the graphs.
Great work, but could you spare the one extra line per graph and include the iPhone 5? In the form of the 5c it is still for sale (and in fact this year as a 5c will probably outsell a number of the models you list above). The 5 and 5c is frequently used and together have outsold even the Galaxy S5, I believe. So it is valuable to see that info. Also your readers are frequently going to be looking at this data to compare the merits of upgrading after two years. (Obviously very common time frame for phones.) So it would be super useful to see this. And if possible, could you do this going forward for all your Apple reviews? Looking back two generations really is part of the story you are trying to tell, especially since two year ownership before upgrade is very common for phones and might be more common for tablets. It is just an extra line in your tables (and you even did it for the battery test) and lets face it that Apple products have a different position in the market from many of the other units you review.
Man, android fans are livid with this preliminary results. Here is the thing android fans, the fact is the current mobile world have apple's hand all over it, whether you want to admit it or not. All current smartphones are using templates from 2007, nothing and I mean nothing has change from that fundamental truth. The current Arm chips that's in everyone smartphones have apple written all over it, it was founded by apple and ARM holdings and served as the basis for the Apple Newton in 1994. Apple sold a lot of their stake when the company ran into financial trouble in 1997. Apple still own a big stake into ARM even today, which might explain why they were able to get 18-24 months jump on everyone for the 64bit ARM chip. This is why most android fans are mad, they just don't want to give apple any sort of credit no matter what. Look, unless android change the current mobile market as far as how we are currently using our smartphones, Apple will always have those things as the pioneer of the modern smartphone. I will not hold my breath for it either since no one in the android camp have the clout for that kind of industry change movement. Have you ever notice what android fans count as revolutionary, NFC, bigger screen, 3gb ram, really! How about market defining things, like you know, forced the industry from bb and their clones blackjack to touch interface.
Well said, sir. The whole modern smartphone revolution is based on Apple's iPhone. But competitors are now trying to rewrite history, as if Apple didn't start the revolution.
Sorry for my english I wanted to ask you a question that I can not figure out between 6 and iphone 6 iphone plus I noticed a difference in the performance of the GPU being equal resolution. I wanted to know if they have a difference in the frequency of the GPU or know what it depends on the difference
There is at least one difference, the 6 plus as being physically larger for cooling and able to rely on a larger battery doesn't need to throttle down its clock speed as far, as quickly as the 6.
Surely the reason why the physics test sees little improvement is that it's limited mostly by memory and nothing else. Of course it should still see some improvement, but perhaps it's not limited by either CPU n'or GPU.
And are those resolution numbers right? 2208x1242? That's a 1,15x scaling factor for the 1920x1080 display.
I have a problem with the Battery test. I own an iPhone 5 and i love it, but the battery is shit, I cant belive that this device does better than iPhone 5S 5C and Galaxy S5.And how come the MOTO G CAME ALMOST ON TOP, this phone has mediocre battery.
This is interesting why many other sites write about poor battery life and you say otherwise. Sorry, but there isn't any tech out there that allows good battery life from a 1810 mAh battery, period. Maybe just take the phone and test it in real life for a few days. (http://www.phonearena.com/reviews/Apple-iPhone-6-R...
But you didn't mention what device that was... :) My iPad 2 has pretty awesome battery life, I sometimes leave it weeks between use, but I wouldn't compare it to the small iPhone...
yeah indeed good over other smartphone, ans with specification too likewise talk about weight of the Apple's new smartphone iPhone 6 and iPhone 6 plus, via.. howtoisolve
Hi, I have stock of Brand New Apple iPhone 6 - 64GB Unlocked phones for sale at $650 only, sealed in box with 1 year warranty. Available in Gold , silver and space grey colors
We’ve updated our terms. By continuing to use the site and/or by logging into your account, you agree to the Site’s updated Terms of Use and Privacy Policy.
Facebook
Twitter
RSS
316 Comments
Back to Article
Valis - Monday, September 22, 2014 - link
Excuse me for saying, but this looks rigged.RattyUK - Monday, September 22, 2014 - link
Yeah, I wonder if the Physics test has been written for the A8 processor, or it's just running in emulation.GC2:CS - Monday, September 22, 2014 - link
The physics test between A6 and A7 showed basically no improvement at all.Adding this to the mix means that the iPhone 6 is just 10% faster than a good old iPhone 5 according to physics test.
eanazag - Monday, September 22, 2014 - link
The physics should at least show a bump up because of the clock speed bump if it is hitting the CPU. If it is running off of the GPU, I would expect a bump also.If there really is no bump in performance, I'd like to see the battery life comparable to the other systems.
kron123456789 - Monday, September 22, 2014 - link
Nvidia Shield Tablet is still more powerful. But it's still a tablet. I wonder how it would compete with the new iPads that coming within a month.Giraffy - Tuesday, September 23, 2014 - link
Uh, Tablet will destroy it. The new iPads aren't going to have some magical unicorn GPU in it. The Shield Tablet has the K1, which will not long after be followed up by Maxwell. There's simply no catching NVIDIA at this point when it comes to their mobile GPU performance. It's not a contest.lucam - Tuesday, September 23, 2014 - link
Fair enough, I wonder if you can use the Shield for 12-15 hours without stopping. And moreover if you can ever plug a K1 inside a phone.kron123456789 - Tuesday, September 23, 2014 - link
Well, in wi-fi browsing it can last about 11 hours. In games...not so much.craighamilton - Saturday, December 6, 2014 - link
This only shows that A6 and A7 are not far better than the other phone in the market. Give me reason then to buy one. /Craig from http://www.topreport.org/phones/blackcrayon - Tuesday, September 23, 2014 - link
The new iPads might not have a magical unicorn, but they *could* have an A8X with 10-12 of these PowerVR cores... I like how there's "no catching NVIDIA" even though they were pretty poor for a pretty long time in their smartphone/tablet GPUs.kron123456789 - Tuesday, September 23, 2014 - link
problem is that there is no PowerVR 6XT GPU that have more than 6 clusters, i.e. GX6650 is the top of PowerVR GPUs. And GX6650 can compete with K1 only on frequency higher than 600MHz(theoreticly). We'll know for sure in the end of october.danieleran - Tuesday, September 23, 2014 - link
A bigger problem is that nobody is buying the K1 NVIDIA Shield, while tens of millions of people are buying A7 and A8 iPhones and iPads.There is a concept of profitability and capital reinvestment that shifts where technology goes. Google and even Samsung (more than half of Android) are laser focused on cheap, low end devices, in large part because their attempts to sell expensive tablets (Honeycomb tabs were more expensive than iPads) failed, and even attempts to sell cheap tablets and phones failed. So now they are in India trying to sell $100 phones.
Meanwhile, Apple sells 80% of the $500+ phones in China, 40% of all smartphones in US, and most of the smartphones in Japan.
Anyone who thinks one struggling GPU vendor will save Android despite the fact that the vast majority of those "80% of phones" making up the Android "platform" are using ARM Mali cheap GPUs (including many of the "premium" Samsung models), should prepare to have their expectations dashed.
Apple was winning in smartphones and tablets before they had their own advanced silicon. Now its more of a joke. TI left the OMAP chip business, Samsung is struggling, and ARM isn't catching up to Apple's aggressive support for Imagination Tech - and conspicuously, nobody else is using high end PowerVR CPUs because they're targeting the third world with cheap devices. Didn't work out too well for PCs did it?
Guest8 - Tuesday, September 23, 2014 - link
You do realize Intel uses the same if not beefier GPU in their cell phone APU's right? Put that together with a real die shrink at 14nm and x86 compute power and it is only a matter of time before Apple puts x86 chips into the iPhone.happycamperjack - Wednesday, September 24, 2014 - link
Haha you are possibly the most delusional person that I've seen posted here, ever. I don't even know where to start. Calling nvidia a "struggling" GPU vendor or calling google and Samsung "failed" when the android cellphones are outselling iPhone more than ever. Selling $100 phone to a crowd who cant ever afford a smartphone otherwise is desperation? Man its just easy to call you stupidrobbie rob - Monday, September 29, 2014 - link
happycamperjack,Here's the reality. Android isn't one big happy family. Guess what ? Android makers don't just compete against Apple - they compete against each other. Samsung's sales hurt HTC.. HTC's hurt motorola's.. and LG's ! WOW Guess what ? They all use Android but only Google makes money from Google Play, only Google makes money from the Google searches and Google ads on Android phones. This is one reason why the best selling Android Phone maker: Samsung is trying to make it's own Android like operating system - like Tizen. See Samsung has this dream to one day not have to kiss Googles ass. They want t run their own OS ( like Google and like Apple ) and their own store.
And guess what ? Only TWO phone makers make more money EVERY YEAR. Apple and Samsung. Most Android Phone makers had flat growth ( see none ).. some even lost money.
Apple losing market share? Technically it looks that way yeah - I'll give you that. But only because the market is growing so fast that it appears Apple's share has shrank some.
The reality is that is only true if you compare ALL of Android against IOS. This would be great if all of Android made a profit together- but they don't. This is not true when you compare each phone makers market share to the others regardless of operating system. Its like trying to compare to sale of matches to lighters. Most of the Android phones in poor Asian/Indian countries are cheap phones with little functionality. This is why Android dominates in those countries. Say what ? You think those countries have 4g? Most have 2g if they are lucky. This is why guys like you don't understand how they can still sell a iPhone 4 over there. It's because an iPhone 4 is as fast as a Samsung G5 ona 2g network. Countries like China block Google services - you think they are getting your Google experience? Haha.. over there the phone is junk.
Apple has outsold itself ( SEE SOLD MORE PHONES EACH YEAR ) every year since the original iPhone ( One year before Samsung started copying the iPhone and before the Google guys got kicked off the Apple Board for stealing technology ) .. This is also why Apple got a headstart on tablet sales - the google guys were kicked off and didnt see the tablet idea coming !
Apple just sold 10 million iPhone 6's in 3 days.. The next best selling phone is a Samsung G S5 and it took 28 days to sell 10 million.
You can knock Apple all you want. It won't stop people from buying their products.
The only phone makers that actually matter are Apple and Samsung. Android versus Apple is a joke. Especially when you look at how they all want Apple's profits.
Havekk - Friday, October 3, 2014 - link
You fan boys on both side crack me up. Does it make your dick bigger if the iPhone is the No. 1 selling phone and you have one? Do you all realize how physiologically damaged you are? Why can't you just buy the phone you like and shut the fuck up? Why must everyone bend and come to the same conclusion you did when purchasing your phone? Fact is, you're a joke. You can sit here and spout all the numbers you want claiming whatever you'd like. The fact is, people will buy the phone that suits them better. I am currently using a Galaxy S4 from my work and an iPad 3. I'm buying the Galaxy Tab S 10.5 and my work is getting me an iPhone 6. They both have their uses. They're both great products. Why must people stick a fucking flag in the ground behind one of these companies who don't get a flying fish turd about you? You make yourself look like a moron. Jebus fucking christ you people drive me nuts.Vayra - Friday, September 26, 2014 - link
What have you been smoking? Nice presentation of twisted and inflated marketshare numbers (what is that, 80% of the +$500 market in China? their Iphone copies cover almost every segment and cost a quarter of an Iphone) though.Regardless, the only country in which Apple really holds ground is the US. In Europe and Asia it will not last and the market share has actually decreased year over year.
Havekk - Friday, October 3, 2014 - link
You fan boys on both side crack me up. Does it make your dick bigger if the iPhone is the No. 1 selling phone and you have one? Do you all realize how physiologically damaged you are? Why can't you just buy the phone you like and shut the fuck up? Why must everyone bend and come to the same conclusion you did when purchasing your phone? Fact is, you're a joke. You can sit here and spout all the numbers you want claiming whatever you'd like. The fact is, people will buy the phone that suits them better. I am currently using a Galaxy S4 from my work and an iPad 3. I'm buying the Galaxy Tab S 10.5 and my work is getting me an iPhone 6. They both have their uses. They're both great products. Why must people stick a fucking flag in the ground behind one of these companies who don't get a flying fish turd about you? You make yourself look like a moron. Jebus fucking christ you people drive me nuts.Alexvrb - Tuesday, September 23, 2014 - link
According to what I've read, a 950Mhz GK20A in the K1 chips (either one) produces less FLOPs than a ~650Mhz GX6650. Not to say that that's all that matters, but it's a pretty good indication that GX6650 is quite competitive. I wouldn't jump to conclusions, even if they're "only" using 6 clusters.With that being said I've often defended Nvidia's Tegra chips. I think they are very good overall performers, especially given the price. They usually undercut comparable ARM SoCs, and as a result provide very good performance (especially graphics) within a given price range.
I'd be a lot more interested, frankly, if Apple used Wizard. Get all those Apple developers onboard with hardware-assisted hybrid raytracing. Might encourage others to develop similar GPUs.
kron123456789 - Wednesday, September 24, 2014 - link
But are we talking about same FLOPs? GK20A produces 365GFLOPS in FP32 at 950MHz. GX6650 produces ~250GFLOPs in FP32 at 650MHz, but in FP16 it produces 500GFLOPs. I don't know how many FLOPs GK20A produces in FP16.darwinosx - Tuesday, September 23, 2014 - link
The Nvidia based tablets use a crap os and development tools. iPad Air destroys it now in real world and the next one even more so.kron123456789 - Tuesday, September 23, 2014 - link
iPad Air destroys Nvidia Shield Tablet with...what? Games for K1 using OpenGL 4.x, not mobile OpenGL ES 3.0. And iOS doesn't have games like Half-Life 2, Portal or Trine 2: Complete Edition. BTW, games with Metal(like Asphalt 8 or MC5) have only a little more particles. Besides, i was talking about hardware, not software.djgandy - Tuesday, September 23, 2014 - link
Yeah no catching them in their sales figures either....darwinosx - Tuesday, September 23, 2014 - link
Uh you don't know what you are talking about. These are synthetic benchmarks. Throw in games developed with metal and they will destroy everything else out there.Havekk - Friday, October 3, 2014 - link
You fan boys on both side crack me up. Does it make your dick bigger if the iPhone is the No. 1 selling phone and you have one? Do you all realize how physiologically damaged you are? Why can't you just buy the phone you like and shut the fuck up? Why must everyone bend and come to the same conclusion you did when purchasing your phone? Fact is, you're a joke. You can sit here and spout all the numbers you want claiming whatever you'd like. The fact is, people will buy the phone that suits them better. I am currently using a Galaxy S4 from my work and an iPad 3. I'm buying the Galaxy Tab S 10.5 and my work is getting me an iPhone 6. They both have their uses. They're both great products. Why must people stick a fucking flag in the ground behind one of these companies who don't get a flying fish turd about you? You make yourself look like a moron. Jebus fucking christ you people drive me nuts.ColinByers - Monday, September 29, 2014 - link
Well, either way. There's nothing that will match the HTC One M8 overall (see http://www.consumertop.com/best-phone-guide/ for example) currently. Apple just can't do it, instead they lag behind.blackcrayon - Tuesday, September 23, 2014 - link
Physics test is CPU and is very light on GPU. It might be something that Metal could help with, but it's one area where having multiple cores helps (im not sure to what degree it would help in real world games though).simonh - Monday, September 22, 2014 - link
> Adding this to the mix means that the iPhone 6 is just 10% faster than a good old iPhone 5 according to physics test.That's the 5s, not the 5. The 5s processor, released last year, was the first 64-bit ARM CPU and a true performance monster fully twice as fast as the chip in the orriginal iPhone 5. In terms of architecture the chip in the iPhone 6 is a relatively minor iteration on that epic CPU.
apunari - Monday, September 22, 2014 - link
Wait for A9. The tick usually introduces new design. The tock boosts the performance. Just like intel.paravorheim - Monday, September 22, 2014 - link
Isn't the tick a process node change usually, and a tock being a new design?tipoo - Monday, September 22, 2014 - link
A5, A6, and A7 were all more like tocks. Four in a row would be too much to expect I guess. A8 is definitely a tick, modest performance gains, mostly efficiency gains.stingerman - Monday, September 22, 2014 - link
Performance per Watt and on that Metric, the A8 was a big improvementtipoo - Monday, September 22, 2014 - link
I think you have it reversed by the way, tocks are the big changes.http://www.dvhardware.net/news/intel_tick_tock_roa...
bigstrudel - Monday, September 22, 2014 - link
That's what he meant. A6 and A7 were ground up creations and "Tocks"KPOM - Monday, September 22, 2014 - link
You can only move to ARM v8 64-bit once. I'm surprised they improved as much as they did.name99 - Monday, September 22, 2014 - link
I wouldn't be as pessimistic as this. There are a number of improvements that Apple can add to their existing CPU/SoC that would improve performance substantially without compromising power. Most obvious (but by far the only) improvement would be a really robust and powerful turbo-ing system like Intel uses.The real reason for the lack of improvement, IMHO, is that it was not necessary this year, so Apple directed priorities in a different direction; most obviously I expect their best and most experienced designers were working on the S1 (for the aWatch) and the A8 was given over mostly to less experienced engineers as a learning project.
Next year (once the A57 devices are actually available, along with Denver and the various ARM server cores) there will be more compelling competition for Apple, and I'm guessing they already have a more substantial CPU/SoC boost planned for that.
yakuba - Monday, September 22, 2014 - link
The physics tests is one of the test that benefits heavily from multiple cores. Unless Apple decides to go beyond dual-core, their phones will lag in this test. It seems the small improvement in performance is due entirely to clock speed. Hopefully the detailed review will make this completely clear.whyso - Monday, September 22, 2014 - link
If you read the futuremark news release they specifically say that the A7 (and now A8) have poor random memory performance resulting in low benchmark scores. Arranging memory objects sequentially the A7 (and supposedly A8) shows the expected performance increases. However, this is a non trivial thing to do in a physics engine as objects are constantly entering and exiting the scene and you need physical memory space to place them in. This also shows in a few geekbench sub tests and is one of the 'glass jaws' flaws in the CPU design.Arbee - Tuesday, September 23, 2014 - link
This is by no means an A7/A8 specific issue; it's well known that doing heavily random data access is orders of magnitude slower than a sequential array even on Haswell (there's a well known whitepaper in the gamedev community that discusses this). The difference is mostly that the performance "cellar" is a lot higher on HWL than on mobile chips thanks to the higher clocks.user777 - Wednesday, September 24, 2014 - link
The tests are not quite correct - Samsung Galaxy S5 from February 2014 (not Galaxy Note 4) against Apple iPhone 6/6+ from September 2014. Let's wait for the Samsung Galaxy Note 4 tests.realbabilu - Tuesday, September 23, 2014 - link
The javascript test based on Apple Socs always high since it use direct javascript engine on chip called Nitro. It is more like "SSE" for javascript. In other hand for calculation matrix, etc, cpu just run the physics test like other instruction, thats where A7 / A8 loses to other high clocked cpus.FrenchMac - Tuesday, September 23, 2014 - link
No Nitro is just a JIT version of the Apple JavaScript Engine (like V8 in Chrome). It has nothing to do with SSE for JavaScript. A7/A8 executes more instruction per cycle than competing CPU (in 64 bits in particular).The Hardcard - Monday, September 22, 2014 - link
To be excused, you must be specific as to what you mean.Valis - Monday, September 22, 2014 - link
Check the score of iPhone 6 here, scroll down a bit. Compared to the others.http://www.phonearena.com/news/All-bow-to-the-new-...
Kureno - Monday, September 22, 2014 - link
That battery test is used with a web script of some kind. Depending on bluetooth, wifi, signal strenght, brightness and a few other factors and as well what web browser that is used and how well optimized the browser is for each phone and how power hungry each browser is.That's why I don't trust the webscript battery test so much.
robbie rob - Monday, September 29, 2014 - link
agreedminifi - Monday, September 22, 2014 - link
So? Just because this comparison rates the iPhone much higher "it looks rigged"? Why, because you want it to?designerfx - Monday, September 22, 2014 - link
Not everyone performs their tests the exact same way. However, it is notable that iphone is not leading the pack for battery life, even brand new.name99 - Monday, September 22, 2014 - link
The Xperia Z3, for example, a phone that does extremely well on that test, has a 3100mAh battery, as compared to the 1810 of the iPhone6. So the iPhone6 gets more "active life" per energy unit than the Experia Z3 does (and probably any other phone on their list).It's a matter of choice what to make of this. APPLE has prioritized thinness and weight over other things --- they appear to have target battery lives in mind, for laptops, iPads, iPhones (and probably watches) and when they hit the target they start shrinking the device. If YOU don't like this and think more battery is warranted, of course you're going to prefer a thicker device.
What is NOT a matter of choice but an empirical fact is the power efficiency of the device, and Apple appears to be the winner here with the current crop of flagship phones.
darwinosx - Tuesday, September 23, 2014 - link
The difference is nobody buys the Xperia and Apple sold 10 million iPhone 6's in three days. Quality wins over synthetic benchmarks every day.jettto - Tuesday, September 23, 2014 - link
because we have 10 millions of retarded sheep?!I hate iOS but ... DAMNNN! their soc OWN!!!
FrenchMac - Tuesday, September 23, 2014 - link
But in fact, from a developer point of view, and if we look at the low level stacks of the operating system, iOS is has much optimized as their SOC: the LLVM compiler rocks, the 64 bits ABI, the implementation of Swift, the work done on vectorization, the display optimization, the fluidity of the UI, animations and physics in the UI... in all these iOS shines...robbie rob - Monday, September 29, 2014 - link
Ive used every OS out there even had crackberry.. There is no advantage to the mass public to use Android. A smartphone is a internet appliance. The only thing that matters is the apps and both Android and iOS enjoy well over 70,000 apps a piece - and most of those apps are the same. The most popular apps on both are usually the same. A cell phone most important uses- texting, facebook, gps.. ? I suppose now you're going to tell me you can see the difference between you 13 MP picture and the iPhones 8 MP camera? What do you see when you take a picture on a 13 Mp camera picture and view it on your smartphone screen? or 60 inch HD TV? You see a 2 Mp picture because thats all your screen can display ! HAHAAH ! And you really thought you could see !! A 5 thousand dollar 4k SUPER HD TV - the best screen out there - can only display 8 Mp's .. This is why Apple and HTC have not played the marketing hype game of Samsung. A high quality sensor matter much more then Mp's .. Mp's are great for printing posters i suppose..robbie rob - Monday, September 29, 2014 - link
haha true.. Samsung needed 28 days to sell 10 million S5'srobbie rob - Monday, September 29, 2014 - link
well said.. if apple had used a 3100mAh battery they would've had beast battery life..lets not say battery life doesn't matter, but the truth is most of us plug in our phones much more then we admit. I work in healthcare and we have a healthy mix of Android and iPhones. I see alot of phones being charged through the day - both sides
darwinosx - Tuesday, September 23, 2014 - link
Hah!robbie rob - Monday, September 29, 2014 - link
You gotta keep in mind the iPhone 6 has one of the the smallest batterys of the pack.. A8 is very power efficient and the iPhone LCD has always been tweaked to use less power - which you expect as they are doing it with AMOLED's as well. Still.. if Apple had made the iPhone 6 just a hair thicker with a bigger battery - they could've easily had a 14 hour battery life..steven75 - Monday, September 22, 2014 - link
Yeaaaaaaah going to believe Anandtech over phoneareana.com.robbie rob - Monday, September 29, 2014 - link
Ive been reading at Anadtech for years.. and guess what ? Each product is reviewed without bias. And each product is reviewed not just for the bad - but the good things found in each device.When it comes to being able to come here and read about ANY product - I know I'm getting a real look at the device. If you don't like a company that's fine. This is no different then Ford versus Chevy except for some reason the Android boys get more offensive then anyone. They cry "Apple sheep" yet they are the ones acting FANATICAL and like little kids because post all the stupid crap - that won't change anyones mind at places like this and at blogs - ALL FOR A PRODUCT THEY DONT EVEN OWN ! Wow.. some of those Android guys really need a change in life style !
Rlin7 - Monday, September 22, 2014 - link
Your link does absolutely nothing to prove that Anandtech's result are rigged.darwinosx - Tuesday, September 23, 2014 - link
But your comments prove you are a moron. So there is that.minifi - Monday, September 22, 2014 - link
Why? Care to explain your accusations?Narg - Monday, September 22, 2014 - link
Excuse me for saying, but your comment looks rigged. :/tehdef - Monday, September 22, 2014 - link
Things AnAndTech doesn't do. Rig.They specifically call out groups for skewing and often blatantly cheating on benchmarks.
Please learn your place Valis.
Giraffy - Tuesday, September 23, 2014 - link
I thought it was selective to leave out the fact that NVIDIA cleaned house. Don't care about the fact that it's a tablet. Still worth acknowledging that what they bring to the table is epic and is poised to change the landscape of mobile gaming in ways that nobody else in the industry is even close to achieving.lucam - Tuesday, September 23, 2014 - link
I wonder why there is no Nvidia Shield Phone....kron123456789 - Tuesday, September 23, 2014 - link
Because Tegra K1 consumes too much power at full load.lucam - Tuesday, September 23, 2014 - link
That's why, isn't!!djgandy - Tuesday, September 23, 2014 - link
Yeah those damn selective consumers not buying the things. Must be Anands fault!darwinosx - Tuesday, September 23, 2014 - link
These are synthetic benchmarks and Apple slays them with real world app performance. Nvidia uses a crap os.FrenchMac - Tuesday, September 23, 2014 - link
But for the user to really wins, developer must support the new platform and it's particularity (OpenGL 4.0 in this case). Has there already is a lot of different GPU capacity on Android, developers are having hard time to optimize their games for all the hardware.DJDJPHOTO - Monday, September 22, 2014 - link
Then maybe the other top benchmark website in the US will help.http://www.tomsguide.com/us/iphone-6-benchmark-res...
Contrary to what android users think, all those really high specs on paper, are still lacking in overall performance.
It's because the operating system is clunky, and all those extra handwaving features and megapixels that 90% of the world will never use actually get in the way of a great user experience.
But, all those Asian companies will sell some phones because they will prey on the ignorant in America. "Our phone has a quad core processor, more is better". Wrong Samsung! They're just cheap Asian knockoffs of the iPhone.
Buy American people, and stop falling for the stupid "the numbers are higher so they must be better" BS. And people say Apple fans are fall for their marketing...geeez
DJDJPHOTO - Monday, September 22, 2014 - link
*alldmacfour - Monday, September 22, 2014 - link
Your comment is way off base.What's American anymore? Apple uses non-American components (this included Samsung) and android was developed by an American company. They assemble the iPhone in China along with just about anything else you're going to buy these days.
mrochester - Monday, September 22, 2014 - link
I agree with this. The real suckers are those who fall for specs.shm224 - Wednesday, September 24, 2014 - link
Ok. I just looked at Tom's benchmark briefly and noticed that the first benchmark (display brightness) is very misleading.According to Displaymate, Samsung Galaxy S5 has peak brightness of 696 nit vs Apple iPhone 6's 558 nit. Dr. Soneira remarks that "The Galaxy S5 has the Highest Brightness and Best Contrast Rating for High Ambient Light that we have ever measured."
Now, the S5 has three different display modes (professional, adaptive, cinema) and can be configured to respond to different ambience lights / picture levels at different nits. I'm assuming that these benchmarks are designed to gauge top, peak hardware performance. So why is Tom's G comparing Apple's top nit against the S5's lowest nit figure?
shm224 - Wednesday, September 24, 2014 - link
I'm not going to dissect the rest of Tom's G flawed review and I certainly don't want to flame any fanboi war here -- but if you want to demonstrate that the new wares from Apple are indeed better than Android's, you would have to bring legitimate, well thought-out reviews.Gasaraki88 - Monday, September 22, 2014 - link
Other site has iPhone6 battery as much lower.http://www.phonearena.com/news/All-bow-to-the-new-...
Mugur - Monday, September 29, 2014 - link
My 6" Xperia T2 Ultra Dual has 3000mA and I get 3.5 days of moderate use. And I use it with 2 SIMs. So I think that Sony has something good here...My previous phone was an HTC One M7 and I never got more than 1.5-2 days out of it (2300mA).
twobeer - Monday, September 22, 2014 - link
Of course it rigged and testcases choosen to highlight the software aspects of iOS rather than try to do real hardware benchmarking. Who do you think pays mr. Anands salary these days :)FrenchMac - Tuesday, September 23, 2014 - link
Hardware is used through the software. Users doesn't only the hardware part of their phone. They use their phone through two filters:the operating system (it must be optimized for the hardware) and the apps (games, video, music, others...)
If one of these two filters is not optimized for the hardware (it could be from a screen resolution standpoint of from a performance standpoint) the user won't get much.
That's the biggest force being the AppStore ecosystem: developers implements new iOS feature in droves to encougare users to stick to their apps. It's a really dynamic market.
Strallus - Saturday, September 27, 2014 - link
Well, Anand left the site, so... other stuff?http://www.anandtech.com/show/8456/the-road-ahead
vgermax - Monday, September 22, 2014 - link
Anand commented in passing in the iPhone 5s review. The physics portion of 3DMark is multi-threaded and capable of hitting multiple cores, this is where iPhone performance is penalized due to Ax remaining dual-core vs SnapDragon et al. going quad-core.http://www.anandtech.com/show/7335/the-iphone-5s-r...
nomster - Monday, September 22, 2014 - link
No offence but you look a bit rigged tooSamus - Monday, September 22, 2014 - link
LOL. I love how people can't accept the fact Apple can do 'something' right. They're not the most valuable tech company in the world because everything they make is 100% crap.They do get a great deal of shit right. But annoyingly, they get the most simply shit wrong. Like being 2-3 years late in making a phone with a usable on-screen keyboard.
darwinosx - Tuesday, September 23, 2014 - link
Apples keyboard is pretty great. Theirs was pretty great when all the Android keyboards were laggy shit which wasn't long ago.Strallus - Saturday, September 27, 2014 - link
Yeah, so ummm, I've used Android phones.I much prefer a keyboard with 0 noticeably latency than a keyboard that has predictive text and something gimicky like swype.
jacure123 - Tuesday, September 23, 2014 - link
I concur, at phonearena.com, the battery test were OK for 6 n 6plus. They couldn't beat the s5 and HTC m8, but here the 6 plus is beating the s5 n m8 by 3 hours.lolAndreRichards2010 - Tuesday, September 23, 2014 - link
Paranoid Android.darwinosx - Tuesday, September 23, 2014 - link
You are excused for being an idiot .robbie rob - Monday, September 29, 2014 - link
Always does when your phone is at the bottom of the benchmarks..robbie rob - Monday, September 29, 2014 - link
Very ignorant coment - considering many Android fans know ( and admit ) it was Samsung and HTC who got cheating around a year ago on benchmarks..Need a refresher? Samsung caught again:
http://www.androidbeat.com/2013/10/samsung-caught-...
Moctavian - Saturday, February 7, 2015 - link
Just because it isn't biased towards favour YOUR favourite phone, it doesn't mean it's rigged. Get over your jealousy.Alien959 - Monday, September 22, 2014 - link
Nice showing from apple, I can only imagine the battery life if apple used 2000+ mah battery.Narg - Monday, September 22, 2014 - link
I agree 100%. Thin is nice, but to make the 6 just thick enough to cover the camera bulge would have given it 30% more battery room, and still be thinner than the iPhone 5c. 3 weeks of standby on the 6 Plus!! I'd liked to have seen the results on that setup.Impulses - Monday, September 22, 2014 - link
Their chosen config is cheaper for them tho (better profit margin), and enough for most users (going by our current expectations anyway), and also make the devices obsolete sooner (promoting future sales).I've got very little interest in an iPhone, but I'd love to see Apple build one that absolutely destroys any and ALL Android phones in battery life, something where they can genuinely claim 2-3 days of battery life.
That would certainly light up a fire under everyone's butt and stop this obsession with thinner and thinner devices. Something 5-7mm thick isn't even particularly easier to hold than a "chunkier" 9mm phone which tapers at the edge or whatever.
Moving away from the first 12mm thick bricks (hello OG EVO) was great, but Apple and Samsung are just over doing it now. At least Samsung still packs in decent sized batteries, something Moto started with the Maxx but has now shied away from.
canadianchris - Monday, September 22, 2014 - link
Thanks for sharing some early results! I've been considering a move back to iOS, this is really helpful.DJDJPHOTO - Monday, September 22, 2014 - link
The IPhone 6 is the hands down the best phone I've ever used. My wife has a cell phone store, and I've used them all. Android for 8 years, too.The 6 is smooth as butter, and a crazy good camera to boot.
16vjohn - Monday, September 22, 2014 - link
Android 8 years... haha. Oh, I love the nut swingers. Android will be 6 years old tomorrow.Valis - Monday, September 22, 2014 - link
Phonearena puts the battery time under the S5, Z2 and the Z3, see for yourself.http://www.phonearena.com/reviews/Apple-iPhone-6-R...
JoshHo - Monday, September 22, 2014 - link
As far as I can tell the results of our testing are accurate. The web browsing test has been carefully designed to ensure that all SoCs have some level of sleep in order to avoid penalizing SoCs for being faster.bernstein - Monday, September 22, 2014 - link
will you be testing standby times too?jjj - Monday, September 22, 2014 - link
So it's not the same workload? Maybe detail the battery test some more?For the GPU the perf doesn't seem much. Snapdragon hasn't done a major update in quite a while and the A8 is on 20nm. With 810 arriving the Android bunch should leap ahead.
Anyway, as far as i can tell the GPU is some 19mm2 vs 21mm2 last year and the cores with the cache some 12.5mm2 while the SRAM seems to be some 4.5mm2 . One would assume it's the same 4MB but last year the SRAM area looks like it was 6mm2 so it would be just a 25% shrink for the SRAM.
The cores do seem rather huge for ARM and i do wonder how much that SRAM helps.
varase - Thursday, September 25, 2014 - link
You misunderstand - one way to save battery is to do your task faster and go into idle mode.Affectionate-Bed-980 - Monday, September 22, 2014 - link
But designing a test to have some level of sleep... is that more for an SoC comparison or is that more for real world usage? I think the important thing here is to consider how these results are representative of real world battery stats.Your test method should reflect a real world use scenario or something similar. Testing a certain because its good to normalize for other SoCs isn't alone a sufficient reason.
Impulses - Monday, September 22, 2014 - link
Your device doesn't spend a lot of time in sleep/idle tests thru the day?Affectionate-Bed-980 - Tuesday, September 23, 2014 - link
It does, but my point is the test should be designed not just to be fair for SoCs as Joshua has said, but to be representative of real world use. I think in many cases he's pushing too hard for apples to apples synthetic benches that real world use doesn't see.For example, most users I know use auto brightness. I'd argue that a test setup with autobrightness under CONTROLLED ambient lighting conditions makes more sense than pegging the brightness at 200 lux. Here's a good example. The Nexus 5 is known to have a autobrightness curve on the bright side. Who here actually gets 9 hours SOT on a Nexus 5 as shown by Anandtech? Not many. I'm guessing the autobrightness curve probably penalizes the Nexus 5's battery life. If you look at an apples to apples bench, you might expect the N5 to do well, but in real world use, most users find the N5 to be mediocre at best for battery life. Where's the discrepancy? Probably the autobrightness has something to do with it. That information is completely ignored in the battery life test.
I'd argue that in this case, you have a benchmark that isn't representative of what the phones actually behave like. If you end up being too focused about normalizing everything including SoCs, then you start creating unrealistic benchmarks for the sake of apples to apples.
lucam - Tuesday, September 23, 2014 - link
maybe it's the other way around? Maybe this is the only fai way to compare apple vs apple?Anand has been doing those test from a while, but when it comes to Apple it's never fair?
ufon68 - Tuesday, September 23, 2014 - link
Auto brightness is a software feature(for the curve). The manufacturer can change the curve with an update, thus invalidating all results previously recorded. On android, you can probably even tweak it yourself. You yourself have acknowledged the curves differ from a phone to phone, how does it then make sense to use auto-brightness, when you get different luminance on different phones ? That's what i would hardly call a fair comparison.What you call real world comparison here can rather be called "ignorant user comparison". There is no point in running tests for that.
The same with the SoC sleep. Would you penalize one phone in a battery life test for being quicker and loading 10 pages in a time it gets the other one to load 3 ? It makes a perfect sense for a device to load a page, put SoC to sleep as we expect the reader to read the page anyway - and only then load another one.
varase - Thursday, September 25, 2014 - link
That's probably it then; phonearena probably just let all candidates power through no matter how fast they were, never considering that the iPhone displayed the web page 300 times vs. 175 for some lesser phone. or whatever.I've been there, and phonearena appears to have their own agenda.
GC2:CS - Monday, September 22, 2014 - link
Yeah, but phone arena has some strange battery life testing in there... They runned it down in 5:22 which is less than a half compared to anandtech... What the heck they do with that phone :D.But for example the samsung galaxy s5 lasts significally more than the samsung galaxy note 3 and I call that strange as well.
We will need to wait for more battery life tests from other sites and somehow make a average from those as battery life testing is naturaly wildly inconsistent across the Internet.
anactoraaron - Monday, September 22, 2014 - link
The S5 lasting longer is the improvements made in samoled tech. Also makes sense how the note 3 looks bad on battery life - most webpages are mostly white with black lettering. That much white for an older samoled puts the note 3 in a disadvantage- it basically gives you the worst possible scenario for battery life.The note 4 should do much better with all the improvements in samoled tech shown in the S5's numbers.
melgross - Monday, September 22, 2014 - link
There hasn't been much of an improvement in Samsung's AMOLED screens as far as efficiency goes. AMOLED still has serious problems in that area. It's why AMOLED is about half as bright as the average LCD screen.Zink - Monday, September 22, 2014 - link
Displaymate disagrees, check out the Note 4 AMOLED tech. Given the current rates of progress it could very easily be LCD that has the efficiency problem next year. It has been confirmed the Apple Watch will be AMOLED, I would love to see Apple phones and laptops with AMOLED in the next few years.http://www.displaymate.com/Galaxy_Note4_ShootOut_1...
soccerballtux - Monday, September 22, 2014 - link
I question battery results as well, nobody no matter what is getting 8hours screen on time, much less while wifi browsing, out of their Nexus 5. I posit (well, bet) that these tests were run on 4.4.1 or 4.4-- battery lives dropped significantly, on the order of 35-55mins screen on time (thanks "Android System") when 4.4.2 came outSushisamurai - Monday, September 22, 2014 - link
The battery results should be accurate. It's not a question of their testing methodology, but a measure of their precision. Anandtech tests all their phones the same way (high precision), ergo, whatever differences that arise from one website to another, assuming they also standardize their testing methodology, would be from the testing method (function of accuracy). Therefore, any attributable difference from one website to another is due to how they test the phone, however, relative standings for the phones should still hold (e.g.: % difference in battery life should still hold per each phone per website, assuming proper standardization and control of variables)Zoomer - Monday, September 22, 2014 - link
It's irrelevant - they the numbers are used to compare against other phones running the same test.Affectionate-Bed-980 - Tuesday, September 23, 2014 - link
Its relative, but I doubt it scales that well. CPUs scale differently and also different generations of CPUs and network chips will act differently when you compress the time scale. Race to idle benefits the faster CPUs in the end.So its hard to take an Anandtech benchmark and say that a phone that lasts 10 hours will last twice as long as once that lasts 5 hours in their benchmark.
Anandtech's numbers are good for comparing against other phones, but in the end you have that difficulty in translating those numbers into real world usage figures. TBH this is where GSM Arena's endurance rating comes in. Its not the cleanest test method, but its the best we have right now and the only site that tests idle use. GSM Arena's test results are relatively consistent with what you hear out there. The N5 does well on AnandTech but almost every user agrees its mediocre at best in performance. Most users see S5s and M8s outperform their Nexus 5s in typical daily use scenarios in terms of battery, which is reflected very accurately in the GSM Arena benchmarks. The 1 or so hour difference in WiFi life from Anandtech's site between the N5 and other flagships isn't very realistic I don't think. I also find my iPhone 5 can easily outlast my Nexus 5 in screen on time, and the difference is pretty startling.
lucam - Tuesday, September 23, 2014 - link
Look guess what if I don't use my phone, it last 2 days. Is it a fair saying? That is my usage...compare with your phone if you can.uhuznaa - Monday, September 22, 2014 - link
You shouldn't treat these tests as something similar to real use. Loading one web page after another is VERY different from browsing (where you actually scroll on pages), since it just downloads and renders a page and then sits there sleeping until the next load. And the phones are pristine, with no services running in the background, no accounts being synced...But this is the only way to have a controlled setup that creates scores you can compare phones on.
Impulses - Monday, September 22, 2014 - link
I can get 4-5 hours under Wi-Fi in real world conditions with brightness on auto and all kinds of background stuff, I'm not surprised their controlled/standardized test can double that on lower brightness and a clean environment.As others said tho, the point isn't to show you real world results anyway, the point is to compare hardware on as level a playing field as possible in a repeatable and precise way.
Anandtech's numbers are only good for comparing against Anandtech's numbers, but that's true of many precise and repeatable benchmarks or tests on many site/publication regardless of the product we're talking about.
I don't compare Tyll's FR graphs for headphones from those from Headroom and I don't compare Top Gear's tests with Car & Driver's. I still value both, but cross comparison is pointless.
The Hardcard - Monday, September 22, 2014 - link
Do you have a scientific assessment of why the PhoneArena measurement is more valid and accurate? Or is it just, "I want the iPhone to be worse."We do have a difference. Not the same as rigged. Plus, even if one is rigged, how do you know which one was? I plan to wait for more info as to where things stand.
zunjae - Monday, September 22, 2014 - link
It depends on your usage. I have a Galaxy S5 and I may be use it different than you. Did you know the iphone can last 5+ days? Some android phones can only last 3 hoursminifi - Monday, September 22, 2014 - link
I hope you posted your comment (in a rephrased version and a link to this article) under the Phonearena article as well!? :)Narg - Monday, September 22, 2014 - link
As far as Phone Arena is concerned, one test is statistically invalid. Poor math, really poor math. Anandtech shines again on their great testing abilities.tehdef - Monday, September 22, 2014 - link
And? They don't disclose their testing methods and every device runs differently.Furthermore, it's been noted by people who have an iPhone 6 while at crowded cellular locations like Disneyland that through extensive usage they got 1 full day and had over 30% remaining when they plugged in at 11pm. The 6+ got through 2 full days with 13% left.
I don't trust PhoneArena. They simply don't disclose their testing methods. They don't talk about screen brightness, type of connection, etc. It's honestly ... bad journalism and reviewing on their behalf. It's why everyone trusts ArsTech and AnAndTech over every other blog.
blacks329 - Monday, September 22, 2014 - link
Another thing that I find none of these other site's battery tests seem to account for is actual screen brightness. Anandtech has always used 200 nits as it's baseline brightness for all phones in its testing. Some sites don't even mention the brightness they use for testing while others say 'we had the phone at about 50% brightness,' which is also kind of useless. The iPhone 6's get over 500 nits, where as some other phone might get 400 and some odd nits at max brightness and the 50% level of each is not comparable!If there's anyones battery life #'s I'm generally using as a good guage (because it's always a YMMV situation) it's AnandTech.
victorson - Sunday, September 28, 2014 - link
A PhoneArena writer here. We do disclose our battery testing method: all our tests are run with device displays set at 200 nits, and we run a script that simulates real world test (we have looked at average use stats and how much activities draw, and simulate this via a web script), looping it at a set interval. Test results are fairly consistent and in the iPhone 6 case we have run the test more than three times to ensure the results are indeed correct. As to the difference with AT's results - I'd just say different testing methods.Kureno - Monday, September 22, 2014 - link
Would like to see the results of those benchmarks once 3Dmark utilizes MetalNarg - Monday, September 22, 2014 - link
Two thumbs up on that comment! But, it'll take time for developers to move there.Infy2 - Monday, September 22, 2014 - link
I assume older iOs devices were running on iOs8 in these test? What about Android devices? Are they all using Chrome or whatever browser the phone maker has provided?anactoraaron - Monday, September 22, 2014 - link
"Normally an 1810 mAh battery with 3.82V nominal voltage would be quite a poor performer, but the iPhone 6 is a step above just about every other Android smartphone on the market."Yeah and just about every other android phone on the market would be 1080p and not some goofy scaled 1334x750. Screen resolution makes a difference: see LG G3.
How come these charts exclude the Note 3? Apple force your hand? Makes the most sense to compare these two devices...
anactoraaron - Monday, September 22, 2014 - link
Wait. There it isanactoraaron - Monday, September 22, 2014 - link
Its not on every chart. Guess that is what is meant as a preliminary reviewblackcrayon - Monday, September 22, 2014 - link
Yeah, it's Apple's fault for not putting in an unnecessarily high PPI and taking the performance and battery life hit for it. I'm not sure what you mean by "goofy scaled" either, it's the native resolution which is what you see. An *app* could be scaled but that's a different thing obviously.steven75 - Monday, September 22, 2014 - link
No amount or real-world use will ever convince spec-list shoppers.tehdef - Monday, September 22, 2014 - link
iPhone 6+ has 1080p. Your point is invalid.kron123456789 - Monday, September 22, 2014 - link
iPhone 6+ has ~2900mA*h battery.vFunct - Monday, September 22, 2014 - link
Screen resolution doesn't make a difference.You never noticed your screen resolution.
Ever.
dmunsie - Monday, September 22, 2014 - link
"not some goofy scaled 1334x750"The 1334x750 resolution is the native resolution of the panel and is the native resolution that iOS is rendering at on that device. Not sure what you mean by "goofy scaled".
In the case of the iPhone 6+, the display is 1920x1080 but iOS is rendering to an intermediate frame buffer that is 2208x1242 and then scaling down to 1080p at display time. This may qualify as "goofy scaled", but that wasn't what you were referring to in your post.
hanse - Monday, September 22, 2014 - link
The Bluetooth antenna in the iPhones are usually weak - Will you be testing that?steven75 - Monday, September 22, 2014 - link
[citation needed]darwiniandude - Monday, September 22, 2014 - link
I have my 5s by my bed. It plays fine to a bluetooth speaker in the bathroom while also connected to and being controlled from my Pebble over Bluetooth from the shower. This feels like a really long way away, I'm always surprised but it works. I also annoy my wife by playing my podcasts at her to the bluetooth in her car from my car when we're both driving in traffic together.I don't have much frame of reference for comparison, but I've been mightily impressed with Bluetooth range.
apunari - Monday, September 22, 2014 - link
How can a dual core cpu and 1gb ram device be as good or better than a 8core cpu and 3gb ram phone?shrim - Monday, September 22, 2014 - link
Samsung is great at marketing their phones through specs, not so great on choosing hardware that actually gives superior performance. Also, consider that iOS is optimized and designed specifically for the hardware used in the iPhone.Kureno - Monday, September 22, 2014 - link
Better optimized.Also: android runs app in dalvik, dalvik is a virtual machine and dalvik is based on java and java is slow and a resource eater.
widL - Monday, September 22, 2014 - link
It has to do both with hardware and software. On the hardware side, the thermal headroom and maximum power consumption is very limited in a phone. It also has to do with the CPU architecture.On the software side - Android simply is slower, and multi-core programming is very difficult, and in many cases theoretically impossible (not all problems can be parallelized).
That was just some of the reasons, explained very briefly.
DanNeely - Monday, September 22, 2014 - link
Very little non-graphics/scientific software scales well to many cores. The iPhone's poor showing in the physics benches is probably due to that test being able to scale to the extra cores common on most Android devices. Extra memory only matters for heavy multi-tasking. I suspect a test that loaded switched between enough browser tabs that the iPhone and budget android phones were forced to unload/reload due to memory pressure but android flagships weren't would generate very different results.uhuznaa - Monday, September 22, 2014 - link
That's like asking "How can a 4 cylinder engine be better than a 6 cylinder engine?"The count of cores has very little to do with how a SoC performs. Also note that this is NOT as in the Intel days when everybody just bought the very same chips. Apple's SoCs are utterly different from what others come up with and Apple is one of the very few companies that not only license ready ARM designs but have an actual license to design their own cores (and whatever goes between them).
DeciusStrabo - Monday, September 22, 2014 - link
Regardless if you produce 400 hp with 6 cylinders or with 8 cylinders - you still have 400 hp.Sushisamurai - Monday, September 22, 2014 - link
The difference would be displacement. If I had a 6 cylinder 2L displacement for 400HP versus a 8 cylinder 3L displacement for 400HP, i would choose the 6 cylinders for the efficiency.You could also extend the metaphor to the car is advertised as 400HP, but brake horsepower could actually be 250BHP vs 300BHP.
SirKnobsworth - Monday, September 22, 2014 - link
Apple makes up for the frequency difference with more IPC, and most smartphone workloads aren't threaded enough yet to notice a difference beyond dual core.Vjgoh - Monday, September 22, 2014 - link
If you go back to last year's iPhone 5s review, you'll find this tidbit at the beginning:"Then there’s the frequency discussion. Brian and I have long been hinting at the sort of ridiculous frequency/voltage combinations mobile SoC vendors have been shipping at for nothing more than marketing purposes. I remember ARM telling me the ideal target for a Cortex A15 core in a smartphone was 1.2GHz. Samsung’s Exynos 5410 stuck four Cortex A15s in a phone with a max clock of 1.6GHz. The 5420 increases that to 1.7GHz. The problem with frequency scaling alone is that it typically comes at the price of higher voltage. There’s a quadratic relationship between voltage and power consumption, so it’s quite possibly one of the worst ways to get more performance. Brian even tweeted an image showing the frequency/voltage curve for a high-end mobile SoC. Note the huge increase in voltage required to deliver what amounts to another 100MHz in frequency."
Mobile devices just aren't very good at parallelization yet. You're constrained by the amount of power you can draw and the amount of heat you can dissipate. Additionally, a lot of problems that we solve on our mobile devices (email, web browsing, etc.) just don't break up into parallel problems well. However, I expect this is why the Android phones all crush the iPhones on the physics tests--those problems CAN be solved by throwing more cores at them.
SirMaster - Monday, September 22, 2014 - link
Same way a 3GHz 4 core Intel CPU beats a 4GHz 8 core AMD.cheinonen - Monday, September 22, 2014 - link
Performance with multiple cores all depends on the task at hand. Think of building a house. There are a lot of tasks going on at once and so having more people/cores to work on that will get it done faster. If all you have to do is change a lightbulb, then having 8 people/cores will just result in one changing it while seven look around doing nothing. Having two really fast cores so you can do that task faster (while one handles background tasks) will work better.Software design is much further along than it was a decade ago when it comes to multi-core workloads, but there are still many tasks that just don't scale well to multiple cores. This is what hampered the PS3 as well, in that it had tons of cores but was harder to program for than a system with fewer, faster cores. There is no single right choice, as each has benefits and drawbacks, but it's why you'll see different systems win different tests as they're optimized towards different workloads.
tehdef - Monday, September 22, 2014 - link
Because iOS is native and Android is not.This is the most clear cut and precise answer you will receive.
tuxRoller - Monday, September 22, 2014 - link
Android IS native...most third party apps aren't.vFunct - Monday, September 22, 2014 - link
Very few software uses more than 1 or 2 cores.Basically those extra cores in an 8core system are wasted, designed to fool ignorant people into thinking they are better because it has bigger number.
The thing about engineering, is that a bigger number isn't better.
DeciusStrabo - Wednesday, September 24, 2014 - link
That's not correct. For example Android doesn't even allow you to run network stuff in the same thread as UI stuff. Most Apps are actually multi-threaded without you doing anything nowadays, because the libraries and APIs you call already do the whole scheduling stuff for you.Also, even if a App is single threaded you usually run several Apps on your phone at the same time, doing background tasks, you have tasks checking your connections, GPS, audio etc. All things that benefit from more threads.
Shatta_AD - Monday, September 22, 2014 - link
I'm surprised all these people claiming their phone(s) have 8-octo Cores doesn`t understand the concept behind the implementation of the BIG.little configuration. Most implementations actually doesn`t use all 8 cores. At any one time, EITHER the `BIG`(4) cores are processing heavy threads or the `little`(4) cores are processing light threads. So technically, all these marketing gimmicks of so called `Octocore` SoCs are in fact `Quadcore` systems.rafoix - Monday, September 22, 2014 - link
The same way that an Intel quad-core can outperform an AMD octa-core while the AMD also has a higher clock speed.A better design will give you better real world results.
Scannall - Monday, September 22, 2014 - link
Intel i5's with 4 cores beat out AMD 8 core CPU's all day long, with a much better IPC. 2 big cores make a lot more sense in a mobile device. It's had to get software that takes advantage of 4 cores in your desktop computer, let alone your phone.GC2:CS - Monday, September 22, 2014 - link
Apple A8 a power sipping beast.juicytuna - Monday, September 22, 2014 - link
"In in GPU benchmarks, we generally see a pretty solid lead over the competition for the iPhone 6/A8."Actually, it looks to me about on par with its contemporaries (SD 810, Exynos 5433)
Also your battery tests need revising as they don't reflect real world usage at all. I can tell you from experience the Moto G is not in the same league as the Note 3 in terms of battery life.
juicytuna - Monday, September 22, 2014 - link
SD 805*kenansadhu - Monday, September 22, 2014 - link
Hi there. I know this is Anandtech and you're few steps above the others, but may I kindly ask you to check the performance over time of this new iPhones? I heard that what's special about this new chip isn't the increase of performance, but the power/heat management (maybe I even heard about it from this site). Thank you!MacMan1984 - Monday, September 22, 2014 - link
Regarding battery life: don't you need to compare "apples with apples" (as it were)? You can't just list a device's name: both the size of it's battery and, more importantly, the weight of the device, plays into this. Yes, the Ascend Mate 2 has a longer battery life, but it also weighs in 202 g as opposed to the 129 g of the iPhone 6 and the 172 g of the 6+). So while you get a longer battery life with the Mate 2, you have to carry around the extra weight. This is an important trade-off and I think it should be in there with the other stats.Vjgoh - Monday, September 22, 2014 - link
Not really. If what you're concerned with is the absolute life of your phone throughout the day, and that's your main consideration when buying a phone, it's likely that you'll eschew other factors just to get the most life out of the phone that you can. A lot of people online seem to cite the swappable battery and SD card in the Samsung as reasons to buy that phone, regardless of the merits of other phones. If you want your phone to last all day, the power to weight ratio probably isn't going to impress you much.Zink - Monday, September 22, 2014 - link
I don't think I've every seen battery life compared like that. If no one wants to read it then it is probably a waste of time to create, I'd rather websites I read publish interesting things. No one is stopping you from publishing that article though.RebeccaS - Monday, September 22, 2014 - link
It's pretty much pointless comparing an Ascend Mate 2 with its 6.1" screen to a small 4.7" iPhone 6 anyway. They aim at different segments of the market. I wouldn't personally even consider a 5" or larger mobile phone, they are simply too big to be very useable, may as well get a tablet. Compare it to an Xperia Z3 Compact instead, as that's a similar size phone, aimed at the same market segment. Gives a much more interesting result. Unfortunately Anandtech doesn't appear to have tested the Z3 Compact yet, so how it would compare in those battery life tests is yet an unknown factor.AppleCrappleHater2 - Monday, September 22, 2014 - link
FACT: Apple has been forced to copy Android in style and size for years because people abandoned their tired, moribund and fossilized devices for superior and innovative Android devices.Steve Jobs said no one should want a 7" tablet until everyone went and bought Android devices forcing Apple to copycat with the iPad Mini. Apple didn't think anyone wanted a phone screen larger than a business card until they all bought Androids thus forcing the arrival this week of the iPhone Galaxy and iPhone Galaxy Note clone phones.
Swipe down notifications that don't interfere? Copied from Android and WebOS. Siri? Bought and ruined from a private developer; Google Now crushes it. 3rd-party keyboards? Welcome to 2010, iChumps! Widgets? Welcome to 2009 except you can't place them on your home screen. Live wallpapers and hidden icons? Maybe Apple will get around to copying those in iOS X in 2016. Who knows.
Apple lacks creativity and honest people acknowledge it. Steve Jobs gets credited as an innovator when all he was was a huckster who'd spot someone else's tech, polish it up nicely, then slap a gnawed fruit logo on the back, charge a premium price and wait for the rubes like Jim Smith to hand over their cash like the good iSheep they are.
But after that initial iteration, Apple is incapable of actually innovating something new. They literally cannot make a product until someone else shows them how and they copy it. They are also unable refine things because they believe to improve is to admit something was imperfect the first time. (This is why QuickTime 4 had a legendarily terrible UI that was never changed through QT7 a decade later.) All they can do is make things incrementally thinner or faster but it's just minor refinements since they can't invest their way out of a wet paper bag.
For all their squealing about Retina displays, they never even had a HD display until now; 8th time is the charm, though you need the iPhone Galaxy Note to get the 1080p that many Android users have had for at least a year and is now considered bare-minimum spec. At the rate Apple drags along, QHD screens should arrive in 2018. Maybe. A graphic went around after the reveal comparing the iPhone Galaxy to the Nexus 4 from 2012. Exactly.
TL;DR: Apple is finally (and desperately) copying where Android OS and phones have been for the past 2-4 years.
lagnaf - Monday, September 22, 2014 - link
And that has exactly what to do with these benchmark results? Why are you so invested in which corporate entity is aping the other corporate entity?Graag - Monday, September 22, 2014 - link
Ignore the troll. I wish this site had an ignore button so we wouldn't have to waste our time with these 16 year olds.uhuznaa - Monday, September 22, 2014 - link
God, you're really obsessed with Apple, aren't you?garbul - Monday, September 22, 2014 - link
May I suggest a brisk walk; or, perhaps, a lobotomy.blackcrayon - Monday, September 22, 2014 - link
Isn't it ironic that people post moronic walls of text in the very same articles that prove how stupid they are?TruthLoader - Monday, September 22, 2014 - link
You've been soiling yourself in public with your iTard blather all through these comments, but you totally beclowned yourself here, you sad pitiful foolish fucking cunt.Your ranting insult makes no sense at all, you are wasting space, oxygen, time, help make the world a better place, go and die.
dmacfour - Monday, September 22, 2014 - link
You sound like a neckbeard.TruthLoader - Monday, September 22, 2014 - link
Neckbeard:http://www.urbandictionary.com/define.php?term=nec...
apunari - Monday, September 22, 2014 - link
Do we really need qhd screens on a 5.5 inch mobile phone?tripleverbosity - Monday, September 22, 2014 - link
TL;DR I'm a butthurt Apple haterkirito - Monday, September 22, 2014 - link
Go back to your den and still dream about your device to be the best. your worst nightmare is here in front of you - apple pawned your super spec'ed sammy phone. the benches performed by anandtech is right in front of you.akdj - Monday, September 22, 2014 - link
Where in the world did that diatribe come from?Much more to a display than resolution, especially at a million pixels + @ <6"!
U sir have spent WAY to much time, alone with your spider named Andy and phone named Android. I use both and am quite lost with what you've added to a discussion based on this year's A8 performance. Not features.
iPhone, first out of the gate in 2007. Android 2008. First with 'retina' or HiDPI displays. Check. Now an OEM or two has decided we need qHD on our 5" phones when most are quite content with their 1080p 65" televisions in their living rooms?
I'm all for HiDPI, but you can go too far thus exhausting battery life.
Android has copied, Apple has copied, as has Windows. The latter two of high have matching and wide spread vertical and horizontal aggregation and integration with your 'other' everyday tools like the laptop or desktop. Who's carrying s Chromebook these days?
And if you believe the A8 second gen 64bit mobile SoC coupled with the new APIs, XCode and it's free update along with the marriage with OS X AND Windows 8.1 with continuity and handoff with OSx specifically, it's not JUST the phone! I'm slamming this response out on my year old lowly A7 Air with half the cores running at half the speed and using a third the RAM as my Note 3 while sporting a HELLUVA lot more pixels!
Wonder why it seems ...after a year of using both everyday, that regardless of the task, gaming, photo/video manipulation, or productivity...when you can find an app in parity, let's use Asphalt8. Doesn't matter which device I use. Note 3. Air. Second gen iPad mini, or my 5s...not a single one fails to SLAM the performance of my Note 3. Graphically, efficiency (this drains my note very quick!) wise and the Apple support network, vertical and horizontal support in my personal and professional life and NO Need to use KIES to organize, sync and support my media, I can count on update for a good two to three years well ahead of the next update necessary and unbelievable resale value!
Did I see ten mill sold in one opening weekend this morning? That's. Amazing!
Sorry, but your comment deserves to be smacked down and hauled out to the trash. Anand, who's now passed the torch to Brian have kept ALL of us informed for the better part of a decade now in the mobile sector all the while maintaining an 'objective' testing standard, call outs to companies 'cheating' their benches, and educating a LOT of us that don't have the time to relearn/graduate silicon, electrical and computer engineering degrees in our 40s with families.
Again, while doing all for 'nothing'. YOU, me and the rest of the world are allowed to learn for nothing and enjoy their reviews. It wasn't but six months ago five or take they were praising Samsung's latest SupAMOLED display technology now meeting the best performance of LCD and IPS/LED. Displaymate, another reputable site seems to agree.
Contributing is one thing. Even 'questioning' a result. But what you've said is almost as ignorant as one can get.
Seriously ...move along
J
RebeccaS - Monday, September 22, 2014 - link
Nope, no need for KIES on the iPhone, instead you need to rely on iTunes. Hardly better in my opinion.Superkonna - Tuesday, September 23, 2014 - link
You don't need iTunes at all. You don't even need a computer. This has been the case for a long time now.TheReckoner - Monday, September 22, 2014 - link
FACT: Google purchased Android and was forced to start over after the reveal of iPhone OS in 2007. Next.Hyper72 - Monday, September 22, 2014 - link
Yes, I actually saw the first Android prototype; it looked a lot like a Blackberry, even in UI. Credit to Google though for immediately recognizing the potential of Apple's new hardware and UI design where Nokia, RIM, MS and others scoffed at it. In 2008 they had it reworked.I had to wait for Android 1.5 beta (released 2009) to get a software keyboard needed for the EBook reading device we were designing that used Android.
Anyway, everybody borrows from each other all the time.
tuxRoller - Monday, September 22, 2014 - link
In the Plex has said that they had TWO phones in the works, one, the blackberry-like interface, and one, called Dream (iirc), that was only used a touchscreen for input.Nozuka - Monday, September 22, 2014 - link
Are any of those GFX Tests using "Metal"? Would be interesting to see what kind of effect this has.kron123456789 - Monday, September 22, 2014 - link
No, they don't. And more likely won't be.akmittal - Monday, September 22, 2014 - link
I always wonder how Apple SoC with dual core processor and 1GB can top quad core snapdragon quad core with 2 GB of RAM.juicytuna - Monday, September 22, 2014 - link
Dual 6 wide cores optimized for high per thread performance versus many 2/3 wide cores optimized for throughput.And amount of RAM has little relevance in the benchmarks posted above.
It's not rocket science.
blackcrayon - Monday, September 22, 2014 - link
2 comparatively fast cores can be faster than 4 comparatively slow cores. And the RAM doesn't matter as long as whatever job you're doing doesn't require more than you have.bobfromguam - Monday, September 22, 2014 - link
No test is perfect. However I really like this test as it shows what a realistic interaction of the phone would be. To say that the new iPhone is archaic or antiquated is amiss. The phone by most modern standards will be a good experience and apples continued improvements will be a boon to its users. KAI-ZENuncclewis - Monday, September 22, 2014 - link
I think something is wrong with the iphone 6+ benchmark. I downloaded the gfx app and ran the manhattan on screen on my iphone 6+ and I got a score of 31.2 fps. But the app was crashing a lot. I think the app needs to be updated to provide an accurate measurement!uncclewis - Monday, September 22, 2014 - link
Oh wait. I think that score is for the 1136x640 resolution. I see you increased it in the corporate settings. Nevermind!Sushisamurai - Monday, September 22, 2014 - link
FINALLY... i've been checking this for days like an OCDkcvpr - Monday, September 22, 2014 - link
I'm still surprised at how dense people are over hardware specs.
It's a comparable concept to cars. You could have 1,000hp in a car, but if it weighs 10,000lbs it's not going to perform as well as a car with a 1/4 of the weight and half the power.
Another example with cars, you could have two cars with exactly the same specs and weight. But one is 4 wheel drive and one is 2 wheel drive. The 2 wheel drive version will be faster, because of higher efficiency in the delivery of power. Which is comparable to this situation, in that apple designs their software and hardware together to boost efficiency. Most times android phones just don't have the luxury of doing this, so they have to take a premade software package and adapt it their hardware after the software has already been designed. Which leads to the need for higher specs to get the same level of performance.
Common sense (or maybe I should call it uncommon sense?).
tuxRoller - Monday, September 22, 2014 - link
Yeah, I'm not sure you're right at all. I know that this has been the "common sense" view, but I don't think the evidence supports that conclusion.In particular, when you normalize for screensize, battery, and process node you typically see similar results (for battery----third party apps vary in quality far too much to make useful comparisons).
In this case, apple has beaten everyone to a completely new node, so I'd certainly hope they have a lead over everyone else (save that one huawei device I'd never heard of).
IMHO, where "controlling" the hardware/software comes in most handy is performance of drivers. Apple+Imgtec have, hands down, the best drivers around.
frickingphil - Monday, September 22, 2014 - link
"The 2 wheel drive version will be faster, because of higher efficiency in the delivery of power."Not if the 2WD car can't get its power to the ground and it just sits there spinning tires all day.
Apple's integration would be like the manufacturer of the performance car cherry-picking the (or designing their own) grippiest tires for the car, implementing an amazing traction and launch control system, and their own designed dual-clutch gearbox.
The Android implementation would be a bigass twin-turbo LS motor making 1500hp on an average platform running regular tires with no traction control and a manual gearbox. It'll just spin through 4th gear.
It's vastly more powerful in terms of raw output, sure, but the chassis can't utilize the power for any useful work.
BTW when it comes to higher powered cars, it's more efficient to put power down through 4WD than 2WD...cuz you have twice the amount of rubber applying tractive force to the road, despite the inherent drivetrain efficiency losses. A more efficient drivetrain doesn't do jack squat if your tires are melting.
danbi - Tuesday, September 23, 2014 - link
Or, as they say 'All the steam went to the whistle'.centhar - Friday, September 26, 2014 - link
"Another example with cars, you could have two cars with exactly the same specs and weight. But one is 4 wheel drive and one is 2 wheel drive. The 2 wheel drive version will be faster, because of higher efficiency in the delivery of power."Not relevant to the discussion, but I can't let you slip this by. :P Two cars with the same HP at the crank, the 4wd will be faster because of the available traction at all 4 wheels. Ask anyone with a STi or an EVO.
Booty - Monday, September 22, 2014 - link
I realize as a Windows Phone user I'm in the minority, but it would still be nice to see a device from the Lumia line in these tables for reference.Brandon Chester - Monday, September 22, 2014 - link
You can check out the bench section of the site for data about more devices.Narg - Monday, September 22, 2014 - link
I'd say battery life is OK, but not stellar. If they'd not been so keen on thin, and just bumped up the thickness to match the camera bulge and put in a bigger battery, I think we would have seen super stellar battery. Then the iPhone would be a MUST have. I calculate 3 weeks standby on the 6 Plus. WOW!tehdef - Monday, September 22, 2014 - link
Rather have thinner than a battery that may last me 1.5 days instead of 1. I'm going to charge it at the end of the day either way. It honestly doesn't matter as long as it lasts the day of heavy use, which it is confirmed to be doing as long as you aren't blasting it with Bioshock for 4 hours straight.Marcos Stein - Monday, September 22, 2014 - link
Iphone 6 Plus Gfxbench is different from http://www.tomsguide.com/us/iphone-6-benchmark-res...Gfxbench website is down. Maybe updating?
kron123456789 - Monday, September 22, 2014 - link
gfxbench.com is working now. Did you notice that both iPhone 6 and 6+ are using iPhone 5's resolution(1136x640) in Onscreen tests? That's why they have over 30fps in Manhattan Onscreen and 18-19fps in Manhattan 1080p Offscreen.GC2:CS - Monday, September 22, 2014 - link
The app probably hasn't been updated to the new resolution yet.30 fps is still great considering the GPU is trying to upscale that from 640p.
kron123456789 - Monday, September 22, 2014 - link
It is definetely not updated to the new resolution. But, anandtech's results seems pretty valid.frickingphil - Monday, September 22, 2014 - link
Scaling these days seems to be pretty negligible in terms of performance; all digital monitors and TV's do scaling whenever your input resolution doesn't match the native display of the panel.It's a lot less work to render at a lower resolution and upscale to native vs. the other way around.
kirito - Monday, September 22, 2014 - link
Now, apple haters gonna hate again with this results. This would shut them off for a while. Waiting for the full review.vFunct - Monday, September 22, 2014 - link
Could use a comparison with the iPad for resolution effects.RickC - Monday, September 22, 2014 - link
And no Windows phones? no Lumia 1520, 925, 830? Talk market share all you want, these are valid competitors and aren't going away.vFunct - Monday, September 22, 2014 - link
No, no they're not valid competitors.Windows phones do not matter in the least. They are insignificant in every way.
They are largely useless because of their terrible design and slow responsiveness.
log0n - Monday, September 22, 2014 - link
Ok you have obviously never actually used a WP8/8.1 device. The speed difference alone was enough to make me give up on my Nexus 5 after 3 months & go back to my Lumia 920. It was much faster to load & use apps in general not to mention the complete & utter lack of any OS stuttering which I was connately hounded by on the Nexus 5 no matter what I did (enable ART, Cyanogen, AOSP).The consensus is Lumia's have the BEST cameras in a phone out there. They also happen to have the best Navigation suite (HERE Maps, Nav+, etc.) with the most accurate & up to date maps & in my experience the best usability (live tiles, Glance Screen), the best outdoor readability & Incredible build quality (Go youtube Lumia 920 Hammer Test).
darwiniandude - Monday, September 22, 2014 - link
Lumia's are great phones.They are fast, like iOS, unlike Android.
The cameras take awesome pictures.
The shot to shot time, focus speed, burst speed, are appallingly slow compared to even the iPhone 5s. This may or may not matter depending on what you take photos of. But with a Lumia you may miss the shot. (Eg the Bee flys from the flower before you can get a second shot in)
Also, low light performance at Night time is quite terrible on some models.
darwiniandude - Monday, September 22, 2014 - link
I would also like to see them in these tests. Is GFXbench, Geekbench, 3D Mark etc available on Windows Phone? (Not sure, just asking)tuxRoller - Monday, September 22, 2014 - link
Is there a video you can point me to that will prove your point? My experience is that win8 performs very similarly to android on the n5 (in particular). The input latency tests I've seen also tell the story that win8 and android have similar results.log0n - Tuesday, September 23, 2014 - link
No I don't have any videos to point to but just go to a store & try one for your self you'll see the speed difference (at least compared to the Nexus 5 class hardware)tuxRoller - Tuesday, September 23, 2014 - link
I have, that's why I said "my experience". It's entirely possible all the win8 devices i tried were crap, so I asked if you have something I can refer to verify your claimiLovefloss - Monday, September 22, 2014 - link
This is just the preliminary results. Once the full review come in, then we can judge. This preview doesn't even have the Note 3 in all the tests.kirito - Monday, September 22, 2014 - link
Motorola or LG should ask Nvidia for assistance. Nvidia is becoming a soc powerhouse in these tests.varinsgolf - Monday, September 22, 2014 - link
How come there's no comparison to Samsung's Note 3? If the iphone6 plus is being reviewed, it should compare with other phablets as well.tekeffect - Monday, September 22, 2014 - link
I got the 6. I had a palm pre and then all the htc evos and ones. I've never liked Apple much but liked apple care. When the screen got big I decided a year of Apple was worth a shot. This phone is great. They probably should have made it slightly thicker and put in a massive battery. But it's a great phone. Apple haters should give it a shot.lilo777 - Monday, September 22, 2014 - link
Note 4 is expected to be a great phone too (slightly thicker with bigger battery and a lot more goodies). So why bother with iPhone?PeteH - Monday, September 22, 2014 - link
Why not try both and pick the one you like?blackcrayon - Tuesday, September 23, 2014 - link
Sounds like you just said they were both great phones...bignbullish - Monday, September 22, 2014 - link
Why doesn't any of the NOKIA LUMIAs figure in your comparisons?Sushisamurai - Monday, September 22, 2014 - link
guys... seriously, just check the Bench part of this website, your Nokia Lumnia's are in there even if no full review of them has been done. Seriously, just check the bench site - it's all there.log0n - Monday, September 22, 2014 - link
Yes but the fact that there are no WP8/8.1 devices included in these tables is kinda bias by neglect.For a lot of people comparable performance is rightly or wrongly the biggest determine factor when buying a device. So when a entire class of devices unmentioned not only do the people looking for an honest comparison of ALL there options lose out but it also works to downplay the the platform to those reading it. Out of sight out of mind.
tuxRoller - Monday, September 22, 2014 - link
Win8 phones don't tend to have as good a battery life (web browsing) as android, for some reason. Though basemark tells a different story.Drumsticks - Monday, September 22, 2014 - link
I'd love to see normalized battery life for smartphones, similar to how you do with laptops. Is that possible?Bragabondio - Monday, September 22, 2014 - link
Guys for those of you saying that the article seems to make Iphone 6 better than it actually is and particularly in light of the recent site's founder move to Apple (unclear if Anand still owns Anandtech but likely as he did not said that he has sold it). Lets be fair - the test presented are typical of what the site was presenting (in terms of tests used) before Anand left. Yes, I personally felt that the Apple coverage under Anand was biased but not in the sense that any of the tests were manipulated but more in the sense of what tests were used.I always felt they were done more from the point of view of an Apple fan be-lovingly playing with its shiny new toy while trying to be objective regarding to the actual data recorded during the tests. For example, the coverage may include how fast you can open a web site but does not tell you how much the relatively smaller (4 inch until now) screen sacrifices in terms of content vs. a 5 inch (Nexus 5) or 5.7 inch (Note 3/4) phone. The default browser was always used and many times such test become more of a browser test under different platform than a device test.
Where I see Iphone 6 lagging is in terms of memory. In the full review I hope we can see tests of how many safari tabs one can open before the OS starts reloading them every time (in other words how good is multitasking under iOS 8). The same test should be replicated under Android L using Nexus 5.
Also it would be great to include in the full review comparison of Android L vs. iOS 8 of things like a) sharing content to Dropbox, Google drive, different e-mail ( i.e. default and Gmail).
b) uploading and downloading files from NAS (you can use Note 4 as a comparator)
c) comparison of what can be done in Anroid L (but not in iOS 8) and vice versa.
Marcos Stein - Monday, September 22, 2014 - link
Multitasking is better than on Android.I can run 8 or more soft synths at the same time on my ipad air connected to a controller without problems.
Apple only must add multiwindow that is in the works. Maybe ios8.1.
tuxRoller - Monday, September 22, 2014 - link
How many of those "soft synths" can you run in android?I'm assuming you have tested the same app on both devices?
davidcmal - Monday, September 22, 2014 - link
So one thing I noticed the reviewer did not mention is information about each phone. Are these with the current OS/browser level available? Or the version the phone shipped/reviewed with? I just ran Kraken/SunSpider on my N5 (ART RT). My results for SunSpider 1.0.2 were 784.3ms total and 6694.1ms Total for Kraken. As Anandtech knows for performance ART>Dalvik maybe some sort of information about that in such a heated post. Or maybe there test was with ART and I have a great N5?tuxRoller - Monday, September 22, 2014 - link
I'm not sure art would affect chrome (native) very much.davidcmal - Tuesday, September 23, 2014 - link
I am not sure, but that is where it would be good if we knew if the device scores are from initial review. Many years ago when I would go to THG they would show benchmarks for graphics card drivers. Not sure if they still do as it was early 2000s, but it seems like Anandtech might want to look at a similar article with different variations of the same device.tuxRoller - Wednesday, September 24, 2014 - link
Sure, the version of chrome would make a difference. So, that's a fair point.This is yet another reason why I wish AT would use a third party browser (my preference would be ff, but any would do) so we can actually get useful data about the hardware instead of having to factor in oem optimizations.
Rapha.194 - Monday, September 22, 2014 - link
What matters are the battery tests using 3G / 4G!PeteH - Monday, September 22, 2014 - link
Maybe for you, but not for me. 90% of the time I'm using my phone (either at work or at home) I'm on WiFi.Rapha.194 - Monday, September 22, 2014 - link
The Site PhoneArena reviews said the battery is badPeteH - Monday, September 22, 2014 - link
The Site AnandTech reviews said the battery is good.Affectionate-Bed-980 - Monday, September 22, 2014 - link
Can I just say that while I generally respect Anandtech's articles for detailed analysis of technical details, can we stop trying to release these piecemeal articles? If you truly want to write an article from start to finish with the typical analysis that Anandtech provides, then do it. If it takes 2 weeks, then wait 2 weeks. Please don't do these preview, then follow up then slight follow up articles.I can point to a bunch of examples (ahem GS4 Part 2 review?) where reviewers promise to follow up and never do. There's plenty of examples from the computer hardware world too where you guys promised a DDR3 roundup article, an i7 overclocking article, and a lot of other stuff following a single review article so we kept on waiting and waiting. Either do it now or never.
victorson - Monday, September 22, 2014 - link
This. I myself am checking the website a couple of times a day (work-related) to see whether the iPhone 6 review is ready, but this half-baked piece just looks amateurish and out of place. Anand has set the bar high, and you guys with Andrei Frumusanu (I loved his last review, hope I got the name right) certainly have what it takes to keep it up.Rapha.194 - Monday, September 22, 2014 - link
The Site PhoneArena reviews said the battery is baduhuznaa - Monday, September 22, 2014 - link
The site PhoneArena doesn't say how their test works though.Busterjonez - Monday, September 22, 2014 - link
There is something absolutely crazy going on with the web browsing battery life test.On no planet can the Nexus 4 ever achieve 6+ hours of screen on time. (Source: I own one)
There is zero chance that my LG G2 can ever get 10+ hours of screen time.
I get that the test is supposed to give a relative idea of battery strength. However, a better representation of an actual use case would be greatly appreciated!
Rapha.194 - Monday, September 22, 2014 - link
The Apple website shows that the battery is better iphone 6 1 hours "at most" compared to iPhone 5sHow can these tests take more than 2 hours apart?
Sorry. but I believe more in the battery test PhoneArena
DERSS - Monday, September 22, 2014 - link
Test on Phone Arena is crooked: the faster your phone, the faster it burns battery in their test.ninjaroll - Monday, September 22, 2014 - link
Jesus, I just went to PhoneArena out of curiosity, and can say I won't be going there ever again. Nothing but cry babies in the comments.lilo777 - Monday, September 22, 2014 - link
Comment sections on PhoneArena are horrible but the site does publish a lot of mobile-related news (probably more than any other site)shm224 - Monday, September 22, 2014 - link
Can anyone explain to me why SunSpider "a relatively useful proxy for CPU performance"?shm224 - Monday, September 22, 2014 - link
Seems like those benchmarks gauge browser performance, not CPU.Microsoft for instance claims IE11 (139) performs better than Firefox 25 (205), Chrome 30 (195), Opera (188) on a Dell Optiplex with a 3.0 GHz Core 2 Duo Intel processor, 4GB RAM, Intel Integrated Video. That's about 40% difference in performance.
I also just ran Octane test on my rMacbook Pro (mid 2012) -- Chrome 37 (24530) outperforms Safari (20853), Firefox (20718) by about 20%.
lilo777 - Monday, September 22, 2014 - link
Probably because this test performs CPU intensive operations but there are many caveats. As I understand, operations in this test can't be multi-threaded (hence CPUs with smaller number of large cores do better in it). Secondly it is very browser dependent so the same CPU (phone) can produce vastly different results using different browsers. Because of that it is mostly useful in just two scenarios:* running the same browser on two different Android phones
* running the same browser (same iOS version) on two different iPhones
Cross-platform comparisons are not very useful (at least not as CPU comparisons)
shm224 - Monday, September 22, 2014 - link
Sure, I can see that it'd be useful look at it within one product line/browser, but, as you indicated, it seems nonsensical to use this composite "web/javascript benchmarks" to gauge "CPU performance."I just did a quick test of my own on my rMacBook 15" (mid-2012, 2.3Ghz Core i7)
Octane SunSpider 1.0.2 Kraken 1.1
Safari 8 20853 189.1 2176.3
Firefox 32 20718 195.2 1156
Chrome 37 24530 190.3 1515.4
delta(min,max) % 18.40 3.23 88.26
So they are obviously not browser neutral, even within one same platform, so apparently they are not just measuring "CPU intensive operations." It's one thing to say the iPhone 6's browser or javascript engine is faster, but to say the A8 is faster than those used by competitors is laughable.
blackcrayon - Tuesday, September 23, 2014 - link
Well, your sunspider scores seem to be pretty consistent...shm224 - Tuesday, September 23, 2014 - link
Sure, my limited sampling is by no means representative, but according to Microsoft's recent announcement, IE 11 (139) outperforms Firefox 25 (205), Chrome 30 (195), Opera (188) on a Dell Optiplex with a 3.0 GHz Core 2 Duo Intel processor, 4GB RAM, Intel Integrated Video in SunSpider scores. That's about 30% difference in SunSpider performance.So these benchmarks are highly dependent on webbrowser/javascript engine implementation (duh!) -- and they make no false pretense about measuring CPU performance as AnandTech does here.
ams23 - Monday, September 22, 2014 - link
Hey Josh, a forum member noticed that the 3DMark Unlimited "Physics" graph appears to be off for the NVIDIA Shield Tablet (it appears to be using the Shield Portable data point instead): http://forums.anandtech.com/showpost.php?p=3673681...On a side note, have you taken a look at the actual gfxbench website for data on iPhone 6 and 6 Plus? It appears that these phones are consistently rendering at the lower iPhone 5S resolution of 1136x640 in the T-Rex and Manhattan On-screen tests (and most other reviews reported those numbers too): http://gfxbench.com/compare.jsp?benchmark=gfx30&am...
Ryan Smith - Monday, September 22, 2014 - link
The Physics graph has been fixed. We had copied over the value for the wrong SHIELD. Thanks for pointing it out.Ex0dus1985 - Monday, September 22, 2014 - link
I have repeatedly run both the Manhattan onscreen and T-Rex onscreen tests on my iPhone 6 Plus and have received much better results than in this test. 37.1 fps and 52 fps respectively. I'd suggest to the author they re-run the benchmarks again.kron123456789 - Tuesday, September 23, 2014 - link
Did you notice, what resolution is used? I know there was 1136x640 resolution. If it was native resolution, there wasn't be much difference between offscreen and onscreen results.tuxRoller - Monday, September 22, 2014 - link
Is there a reason you aren't using the basemark battery test? I'd asked about this before and I was told that you used it often (and basemark clearly works on ios).Shatta_AD - Monday, September 22, 2014 - link
I'm surprised all these people claiming their phone(s) have 8-octo Cores doesn`t understand the concept behind the implementation of the BIG.little configuration. Most implementations actually doesn`t use all 8 cores. At any one time, EITHER the `BIG`(4) cores are processing heavy threads or the `little`(4) cores are processing light threads. So technically, all these marketing gimmicks of so called `Octocore` SoCs are in fact `Quadcore` systems.tuxRoller - Monday, September 22, 2014 - link
It really depends on the implementation. I'd imagine all of the recent versions of big.LITTLE can make use of gts.Laststop311 - Monday, September 22, 2014 - link
It's using the 6 core power vr rogue gpu according to shots of the die. i think its called the power vr rogue 6600Laststop311 - Monday, September 22, 2014 - link
its the gx6650 the a7 used a 4 cluster rogue gpu apples added 2 more clusters to gain the 50% jump in performance, it's the exact same architecture and same gpu just with 2 more gpu clusterskron123456789 - Tuesday, September 23, 2014 - link
It is not so "exact same" architecture. It's PowerVR Series 6 in A7 and PowerVR Series 6XT in A8.Sofia - Monday, September 22, 2014 - link
iPhone 6 and iPhone 6 plus are obviously stronger than their competitors in general. This overview brings Apple new smartphones.appiphany.com.au
Sofia - Monday, September 22, 2014 - link
iPhone 6 and iPhone 6 plus are obviously stronger than their competitors in general. This overview brings Apple new smartphones.http://www.appiphany.com.au
euklid81 - Tuesday, September 23, 2014 - link
So you pick and choose the benchmarks that show iphone6 ahead to say it beats other Android phones? Where's antutu and geekbench, oh wait those show phones such as 6 month galaxy s5 faster than iphone 6 plus lol.jacure123 - Tuesday, September 23, 2014 - link
They are so Apple bias, that's why the Indian guy that started the site is now working for Apple.lol.Tim Cook was like ' good job Asian man for making us look good on your site".
lilo777 - Tuesday, September 23, 2014 - link
I hope I am wrong but it is possible that the author's choice of benchmarks was heavily influenced by the fact that Anandtech Inc owner now works for Apple.jacure123 - Tuesday, September 23, 2014 - link
Bingo!busky2k - Tuesday, September 23, 2014 - link
Arstechnica have just released their own battery testing. The author notes day to day use was similar to a 5S and the android flagships actually outlasted the 6+ in their own metric test. Interesting.http://arstechnica.com/apple/2014/09/iphone-6-and-...
golem - Tuesday, September 23, 2014 - link
The author of the Arstechinca author notes this about their battery testing, "But the short answer is I know our current test is limited and we're working on better ones, I just keep1 getting2 sidetracked3 by other stuff4."kgelner - Tuesday, September 23, 2014 - link
They also note the battery life is better than the sheer numbers would indicate:"Chalk it up to something our Wi-Fi test can't really measure—Apple is pretty good at tamping down idle power usage, at least when your signal is good. I've had an iPhone 6 Plus sitting on my desk all day, buzzing at me only sporadically as new Twitter and Messages notifications come in. It's been sitting at a 90 percent charge for at least the last two hours"
Because real people don't browse for 10 hours straight, they are doing stuff in-between. The less power consumption happens during those periods, the longer the real-world battery life is.
thetanman17 - Tuesday, September 23, 2014 - link
This.DeciusStrabo - Wednesday, September 24, 2014 - link
So? This way even my Moto X (2013) doesn't go below 70 % after one full day of "use".JDG1980 - Tuesday, September 23, 2014 - link
You should include a benchmark chart showing how many browser tabs can be opened without thrashing and reloading when switching between them. Maybe this would finally convince Apple to put a modern amount of RAM in their new devices.jacure123 - Tuesday, September 23, 2014 - link
Hey, don't tell AppleAnandtech that.lolTorrijos - Tuesday, September 23, 2014 - link
Maybe you should devise that test and post your results?1) Design a fake website with controlled webpage sizes
2) Load your webpages in tabs, checking everytime if some of your others tabs had to reload
3) Register the number of max tabs open before reloading become necessary
4) Repeat test multiple times to try to average result
5) Repeat test on different devices and platform
6) Prepare illustrations of your results
7) Write article analysing your results
8) Publish
or
1b) Admit that you don't have the patience, time or technical skills to do any of those things, reaching the conclusion that you have no standing to demand for an engineering spec whose implication you cannot judge.
of course there is always
1c) Keep trolling... (optionnal: put a tin-foil hat before adding something about Anadtech being completely controlled by Apple)
ciparis - Tuesday, September 23, 2014 - link
I'd love to see them address that. It's a big failing, and it has left me scratching my head about how Craig, Jony, Phil & co. use their phones, if they haven't been frustrated by this already to the point where adding RAM was a no-brainer. It was a big missed opportunity to further improve the experience of using these devices.danbi - Tuesday, September 23, 2014 - link
Let's just pretend that more RAM does not use more battery power and other resources, such as .. more RAM.When was the last time your iPhone could not perform what you asked it to, because it ran out of memory?
Rapha.194 - Wednesday, September 24, 2014 - link
when you open many tabs in Safari, it starts to update the pages when you alternate between open tabs[email protected] - Tuesday, September 23, 2014 - link
Since a huge proportion of 6/6+ buyers will be coming from the iPhone 5, it would be useful if that phone's performance was always included in the graphs.JonCBK - Tuesday, September 23, 2014 - link
Great work, but could you spare the one extra line per graph and include the iPhone 5? In the form of the 5c it is still for sale (and in fact this year as a 5c will probably outsell a number of the models you list above). The 5 and 5c is frequently used and together have outsold even the Galaxy S5, I believe. So it is valuable to see that info. Also your readers are frequently going to be looking at this data to compare the merits of upgrading after two years. (Obviously very common time frame for phones.) So it would be super useful to see this.And if possible, could you do this going forward for all your Apple reviews? Looking back two generations really is part of the story you are trying to tell, especially since two year ownership before upgrade is very common for phones and might be more common for tablets. It is just an extra line in your tables (and you even did it for the battery test) and lets face it that Apple products have a different position in the market from many of the other units you review.
toukale - Tuesday, September 23, 2014 - link
Man, android fans are livid with this preliminary results. Here is the thing android fans, the fact is the current mobile world have apple's hand all over it, whether you want to admit it or not. All current smartphones are using templates from 2007, nothing and I mean nothing has change from that fundamental truth. The current Arm chips that's in everyone smartphones have apple written all over it, it was founded by apple and ARM holdings and served as the basis for the Apple Newton in 1994. Apple sold a lot of their stake when the company ran into financial trouble in 1997. Apple still own a big stake into ARM even today, which might explain why they were able to get 18-24 months jump on everyone for the 64bit ARM chip. This is why most android fans are mad, they just don't want to give apple any sort of credit no matter what. Look, unless android change the current mobile market as far as how we are currently using our smartphones, Apple will always have those things as the pioneer of the modern smartphone. I will not hold my breath for it either since no one in the android camp have the clout for that kind of industry change movement. Have you ever notice what android fans count as revolutionary, NFC, bigger screen, 3gb ram, really! How about market defining things, like you know, forced the industry from bb and their clones blackjack to touch interface.haruhiko - Tuesday, September 23, 2014 - link
Well said, sir. The whole modern smartphone revolution is based on Apple's iPhone. But competitors are now trying to rewrite history, as if Apple didn't start the revolution.KoolAidMan1 - Thursday, September 25, 2014 - link
They don't know their history.They do know how to be obnoxious.
marko3283 - Tuesday, September 23, 2014 - link
Sorry for my english I wanted to ask you a question that I can not figure out between 6 and iphone 6 iphone plus I noticed a difference in the performance of the GPU being equal resolution. I wanted to know if they have a difference in the frequency of the GPU or know what it depends on the differenceblackcrayon - Tuesday, September 23, 2014 - link
There is at least one difference, the 6 plus as being physically larger for cooling and able to rely on a larger battery doesn't need to throttle down its clock speed as far, as quickly as the 6.marko3283 - Tuesday, September 23, 2014 - link
So you think that the GPU iphone 6 + has a higher clock speed of the GPU iphone 6?blackcrayon - Tuesday, September 30, 2014 - link
I don't think so, I just think the 6 throttles down to a lower clock speed slightly more quickly than the 6+.Celcon - Tuesday, September 23, 2014 - link
I'm curious to see how all the other SOCs would perform if they were placed inside an odd console with a fan such as the shield.danbi - Tuesday, September 23, 2014 - link
Or liquid metal (not the crystalline material) cooling, such as what Apple has patented few years ago.Ethos Evoss - Tuesday, September 23, 2014 - link
Huawei Honor 6 BETTER than iphone !casperes1996 - Wednesday, September 24, 2014 - link
Surely the reason why the physics test sees little improvement is that it's limited mostly by memory and nothing else. Of course it should still see some improvement, but perhaps it's not limited by either CPU n'or GPU.And are those resolution numbers right? 2208x1242? That's a 1,15x scaling factor for the 1920x1080 display.
Rapha.194 - Wednesday, September 24, 2014 - link
Iphone 6 vai ser melhor somente 1 hora a mais em rede 3G/4GNão esperem mais que isso.
Rapha.194 - Wednesday, September 24, 2014 - link
Battery Life Iphone 6 in 3G / 4G network is only 1 hour more compared to the iPhone 5sDo not expect more than that.
lucam - Friday, September 26, 2014 - link
279 comments for the preliminary results. I wonder how many will be for the final review!!Anand, when is gonna be the final review then? :)
SilthDraeth - Friday, September 26, 2014 - link
Anand isn't around anymore. He's working for Apple.lucam - Friday, September 26, 2014 - link
Hahaha...even I have to disagree with your comment, you made me laugh...lol! :)WoodyPWX - Saturday, September 27, 2014 - link
no, he was right: http://www.theverge.com/2014/8/31/6091393/anandtec...lucam - Monday, September 29, 2014 - link
I didn't know that!...bloody hell!yon1000 - Friday, September 26, 2014 - link
I have a problem with the Battery test.I own an iPhone 5 and i love it, but the battery is shit, I cant belive that this device does better than iPhone 5S 5C and Galaxy S5.And how come the MOTO G CAME ALMOST ON TOP, this phone has mediocre battery.
xnay - Sunday, September 28, 2014 - link
This is interesting why many other sites write about poor battery life and you say otherwise.Sorry, but there isn't any tech out there that allows good battery life from a 1810 mAh battery, period. Maybe just take the phone and test it in real life for a few days.
(http://www.phonearena.com/reviews/Apple-iPhone-6-R...
Jimrod - Monday, September 29, 2014 - link
Battery life is good on my iPhone 6. Don't really care about all these random tests coming out - in the "real world" this is what I get - http://i940.photobucket.com/albums/ad245/Jimrod/BF...Mugur - Tuesday, September 30, 2014 - link
You call that good? :-)http://i1050.photobucket.com/albums/s405/mugurmire...
Over 3 days with 21% left and no power savings active...
Jimrod - Tuesday, September 30, 2014 - link
But you didn't mention what device that was... :) My iPad 2 has pretty awesome battery life, I sometimes leave it weeks between use, but I wouldn't compare it to the small iPhone...Mugur - Tuesday, September 30, 2014 - link
It was in a previous post. Sony Xperia T2 Ultra. 6" screen, 3000mA battery.robbie rob - Monday, September 29, 2014 - link
Thanks for the review.Gary Payton - Wednesday, October 1, 2014 - link
yeah indeed good over other smartphone, ans with specification too likewise talk about weight of the Apple's new smartphone iPhone 6 and iPhone 6 plus, via.. howtoisolvemacyjones - Tuesday, October 14, 2014 - link
Apple again rocks with iphon 6 and iphone 6 plus http://www.appxperts.com.au/leonhk1 - Wednesday, October 29, 2014 - link
Hi,I have stock of Brand New Apple iPhone 6 - 64GB Unlocked phones for sale at $650 only, sealed in box with 1 year warranty. Available in Gold , silver and space grey colors
Interested buyer should E-mail me at: [email protected]