Samsung Galaxy Note 3 benchmarks: is Samsung really cheating?

Samsung Galaxy Note 3 benchmarks: is Samsung really cheating?
Synthetic benchmarks, as unreliable as they can be sometimes, are handy little tools we often use to test a given smartphone's performance. Their results allow us to stack up most handsets against their contenders and see which one of them is, theoretically, the fastest. That is why we test each smartphone we review, from those cheapo entry-level handsets to the top-of-the-line flagships.

And naturally, now that the Samsung Galaxy Note 3 has materialized in retail form, it is time for us to give it the benchmark treatment. It comes with Qualcomm's best SoC so far, namely the Snapdragon 800 packing four Krait 400 cores that can sprint at up to 2.3GHz, a 450MHz Adreno 330 GPU, and a whopping 3GB of RAM has been thrown in for good measure. Needless to say, our expectations are set pretty high.

"But wait, I've read that the Note 3 uses some booster thingy to cheat in benchmarks ." Well, we've heard this one too and it is actually kind of true. This recently discovered "benchmark booster" on the Samsung Galaxy Note 3 is a piece of software made to detect when a benchmark app is running. As soon as one is launched, it locks all of the CPU's cores at their maximum frequency of 2.3GHz in attempt to temporarily increase the phone's performance, thus artificially boosting its benchmark score. Of course, we had to test this, so we loaded an app on our Note 3 showing the CPU's clock speed. Sure enough, it spiked and stuck to the chip's maximum clock rate as soon as we opened AnTuTu, Quadrant, or Vellamo. Moreover, the processor's speed stood at 2.3GHz during most of the time as these benchmarks ran. Interestingly, GFX Bench wasn't affected, so perhaps it isn't being targeted by this so-called booster. 

Just to make sure this kind of behavior wasn't normal, we also installed AnTuTu X Benchmark Editor, which is an edition of the popular AnTuTu benchmark designed "to prevent fraud and cheating". And nope, there was no sudden CPU clock increase when we opened the app, nor was its speed locked at 2.3GHz as the benchmark ran. Instead, there was a lot more fluctuation of the processor's frequency going on as this alternative AnTuTu benchmark was doing its thing. 

Yet surprisingly, test after test, both the official and alternative AnTuTu apps came up with identical scores, generally between 29 000 and 31 000 points, but most often within several hundred points of each another. Therefore, while there really seems to be some kind of a cheating attempt going on, we have reasons to believe that AnTuTu might not be affected drastically by the CPU clock boost. Other benchmark apps, however, might be, so keep that in mind. 

Despite all that, we know it's solid figures you, guys, want from us. Well, here we go.

General benchmarks

We'll start with a couple of benchmark apps measuring the overall performance of each device – Quadrant, AnTuTu, and Vellamo Metal. The first two test the raw computation power of a smartphone's CPU, taking into account all of its cores, of course. The performance of its GPU, RAM, and storage are also measured. The latter does all that as well, except that it doesn't benchmark the handset's graphics.

Higher is better
Samsung Galaxy Note 3
Sony Xperia Z1
Samsung Galaxy S4
As the figures show, the Samsung Galaxy Note 3 is one of the top performers overall, but it isn't an undisputed leader. The LG G2 and the Sony Xperia Z1 are well capable of reaching similar results. This comes as no surprise, actually. After all, they are powered by the same Snapdragon 800 SoC. 

Graphics benchmarks

Next up we have several GFX Bench tests. This benchmark app is focused solely on graphics performance, which is of utmost importance to those who play a lot of heavy games on their devices. It can really push a smartphone's GPU to its limits, all the while demonstrating the kind of eye-candy modern smartphones are capable of delivering.

Higher is better
Samsung Galaxy Note 3
Sony Xperia Z1
Samsung Galaxy S4
This time we have Samsung Galaxy Note 3 leading the pack. Even though the Xperia Z1 and the LG G2 come with an identical GPU, Samsung's phablet somehow manages to squeeze a few more frames per second out of it. The gap isn't vast, but it is clearly there.

Browser benchmarks

Yup, the following tests show how well a phone's web browser performs. Vellamo measures extensively the phone's ability to handle HTML 5 content, while Kraken, developed by the guys at Mozilla, tests JavaScript performance. Note that the latter runs in the web browser itself so it eliminates the possibility of benchmark results being affected by the aforementioned booster. This time we have the Samsung Galaxy Note 3 beaten by both the LG G2 and the Xperia Z1, albeit not by much.

Higher is better
Samsung Galaxy Note 3
Sony Xperia Z1
Samsung Galaxy S4

So all in all, the Samsung Galaxy Note 3 is, as expected, one of the fastest Android phones out there and its benchmark results do support this claim. It does not sit alone at the top, however, for its rivals can easily match, or even surpass its scores. As for those accusations against Samsung for artificially boosting the Note 3's scores, they are indeed supported by a lot of evidence. However, we can't say that the phablet's results are drastically improved by whatever software tool Samsung is supposedly using to cheat with. Ultimately, we all know that a smartphone's real life performance is far more important than its benchmark scores – that's what we should take into account before labeling any smartphone as "fastest". 

Related phones

Galaxy Note 3
  • Display 5.7" 1080 x 1920 pixels
  • Camera 13 MP / 2 MP front
  • Processor Qualcomm Snapdragon 800, Quad-core, 2300 MHz
  • Storage 64 GB + microSDHC
  • Battery 3200 mAh(21h 3G talk time)



1. Shatter

Posts: 2036; Member since: May 29, 2013

If Samsungs cheating LG must be doing tons of steroids + cheating.

39. rd_nest

Posts: 1656; Member since: Jun 06, 2010

^^ My reply not meant for you. But PA in general. Phonearena should first explain why are the using 4rd class benchmarks like quadrant, antutu? Why use vellamo when you know that this was developed by Qualcomm? I challenge them to explain what these numbers mean. My question to them is to explain what does 25,000 on Antutu means in terms of benchmark? Everyone can tell what exactly it means by when someone says on-screen fill rate in GFXbench. Why keep using these GARBAGE benchmarks which gives you an arbitrary number and explains nothing about it's significance? Even if someone "optimizes" for Sunspider, it still makes some sense, but some of these benchmarks are plain useless. BTW, if a benchmark doesn't not push the CPU and GPU to 100% utilization, I consider that useless. It's like performing benchmark at 50% efficiency. What's the point of a benchmarking tool if it doesn't saturate the hardware??

54. nnaatthhaannx2

Posts: 820; Member since: Oct 19, 2011

"My reply [is] not meant for you.." Then why'd you click the reply button?

80. heyhi

Posts: 137; Member since: Feb 19, 2013

cuz i think he wanted to catch PA's attention, posting his message at the beggining is easier to everyone read it usually when there are a lot of messages, some users dont read till the bottom or next page(s).

90. nnaatthhaannx2

Posts: 820; Member since: Oct 19, 2011

It was more a rhetorical question, but whatever.

40. SuperAndroidEvo

Posts: 4888; Member since: Apr 15, 2011

Hey it's not cheating if you don't get caught..... Ohh wait, they did get caught.... right?!? lol

70. sharks

Posts: 236; Member since: Feb 16, 2013

Nope. The G2's Snapdragon 800 SoC is actually slightly over-clocked. This means that its hardware is perfectly capable of delivering better performance relative to the stock SoC on the Note 3 and Z1.

89. Mittal

Posts: 494; Member since: Dec 14, 2011

Obviously you think LG is some kind of an inferior company and that they are nothing but desperate copycats/cheats whose products nobody wants... Any facts to support your feelings or are you just blabbering nonsense ?

2. _Bone_

Posts: 2155; Member since: Oct 29, 2012

If you understand what benchmarking means, then no, they are not cheating - benchmarks merely measure the HARDWARE MAXIMUM without telling much about daily usage. You see 90% of your daily apps don't use peak frequencies in order to save battery, it's called OPTIMIZATION, and it WORKS, as your Note is a battery champion. What Sammy and others do, is letting benchmarks do what they are supposed to: have access to the hardware maximum and show practically useless numbers, but show them as far as the CPU and GPU can go, and that's what the Note III allows.. The "cheating" is optimizing every other app, but I don't feel cheated when I get two full days of usage. I feel blessed.

6. AstronautJones

Posts: 305; Member since: Aug 01, 2012

True. These phones with the same processor and OS should all be close enough it doesnt really matter anyway, right? More about the end user experience with custom software, ie Touchwiz. Matter of preference IMHO

7. HansGoneInsane unregistered

Samsung cheats. Other manufacturers cheat too. Apple doesn't and still sets the benchmarks. Buy iPhones, be happy.

15. WHoyton1

Posts: 1635; Member since: Feb 21, 2013

baaaaaaa go graze

16. PapaSmurf

Posts: 10457; Member since: May 14, 2012

I wouldn't buy an iPhone simply because I care about battery life. Note 3 = two full days on a single charge. Thanks but no thanks. :)

44. bigstrudel

Posts: 605; Member since: Aug 20, 2012

Tablet. Enjoy stretched pockets or your man-purse.

47. Pancholo

Posts: 380; Member since: Feb 27, 2012

You probably need to stop wearing tight, skinny jeans if that's your only reasoning. My pockets are A-OK.

51. GeekyGene

Posts: 88; Member since: Aug 25, 2013

Man purse lol. That actually made me spit out laughing XD thanks.

66. PapaSmurf

Posts: 10457; Member since: May 14, 2012

Fits perfectly in my pants thank you very much. Once a hater always a hater.

48. papss unregistered

Yeah my battery life has been great on my note 3. Note two days but a good day and half.

19. kamejoko

Posts: 252; Member since: Nov 10, 2011

How do you know apple doesn't cheat? If apple cheated, you don't have any tool to know it and can not access to source ios to see it.

27. Shatter

Posts: 2036; Member since: May 29, 2013

The only way to know will be if someone gets android 100% ported to the iphone, people have tried it is extremely hard but possible.

20. tedkord

Posts: 17408; Member since: Jun 17, 2009

We have no idea if Apple "cheats", since we can't examine the source to see what they do. Regardless, I am much happier with my GS3 than I ever could be with a limited device like the iPhone.

91. ncarlosmiguel

Posts: 206; Member since: May 14, 2013

I'm also annoyed on how my GS3 fits in my pocket but I wouldn't trade it for an iphone. Maybe I'll just buy moto x or droid mini for my next phone for pocket and one hand use issues. Android all the way.

43. dratomic

Posts: 483; Member since: Oct 09, 2013

in real life scenarios more people like freedom (file manager and ...) which windows os gives them on their mobiles. and this was available by symbian then and android now. I personally would never think of buying an iphone when there is no system wide file manager and file transfer by bluetooth or wifi direct or... no matter how great user experience is. I don't count it in the same group with android cause it needs a pc (or mac) to be fully operational. so iphone (and ios) would never make me happy. :)

56. papss unregistered

I agree with most of what you said but one part... i don't agree that most people want file manager..hell most consumers have no idea what file manager is.

94. bluekoi

Posts: 1; Member since: Nov 29, 2015

Windows is crap, and their phones are worse

8. SleepingOz unregistered

You've said it all but then, fanboys and haters are too biased to acknowledge this.

13. Pancholo

Posts: 380; Member since: Feb 27, 2012

Fangirls gonna hate. After all, they are merely making the benchmark's job easier, right? If Apple did it, it'd be OK. I don't care who does it; it'd make sense anyway. Every other product out there shows you (generally) their maximum output when it comes to performance. Oh well - Can't stop people from taking a grain of salt and turning it into their biggest problem in life.

24. Jobayer

Posts: 167; Member since: Feb 22, 2013

benchmarking present participle of bench·mark (Verb) Verb Evaluate or check (something) by comparison with a standard google it

42. blazee

Posts: 414; Member since: Jan 02, 2012

Those are fake numbers bc their hardware doesn't operate at that level on ANYTHING else but WHITELISTED BENCHMARK tests. So yea you will have a great device that is a killer at only POPULAR benchmarks lol but nothing else. And not all manufacturers do it. Motorola, nexus drives and apple doesn't. F#ck it I love android but I'm not gonna let some manufacturers spit at me and tell me its rain.

45. Pancholo

Posts: 380; Member since: Feb 27, 2012

Just play the same graphically-intensive game across devices and come back with a report of your findings.

Latest Stories

This copy is for your personal, non-commercial use only. You can order presentation-ready copies for distribution to your colleagues, clients or customers at or use the Reprints & Permissions tool that appears at the bottom of each web page. Visit for samples and additional information.