Nexus 5 performance review: the tale of benchmarks

The Nexus 5 has arrived and it is a powerhouse of a phone that you can buy for nearly half the price of other top-shelf models. Just how powerful is it, though? It’s time to push its limits with benchmarks and see whether it can hold its own against rivals.

The Nexus 5 is powered by the latest Qualcomm Snapdragon 800 system chip with Adreno 330 graphics. This is the same chip used in all latest top of the line devices like the Samsung Galaxy Note 3, Sony Xperia Z1 and the LG G2. Even though the performance of these devices differs slightly (the different amount of RAM and other factors matter), keep in mind that all of them should be on about the same level. Then, there is the new iPhone 5s with Apple’s A7 system chip and the move to 64-bit. We’ll take a look at how all these different architectures compare.

1. CPU Explained
1.1. Nexus 5 scores into top 3 on Sunspider
1.2. Kraken is not a test for cheaters
2. GPU: Play on
2.1. Basemark: Trailing close behind
2.2. GFXBench: the iPhone is the limit
3. Conclusion: best value for the money

*Disclaimer: Almost all manufacturers (except for Apple and Google Nexus devices!) cheat certain benchmarks by forcing CPU and GPU clock speeds to their peak capacity. This is an unusual behaviour for the otherwise varying clock speeds of the CPU and GPU and does not reflect the actual behaviour of a smartphone. This unusual behaviour happens after the handset detects it’s running a specific benchmark applications. The Samsung Galaxy Note 3 is particularly guilty having the largest list of apps it cheats, but others like the Samsung Galaxy S4, HTC One and LG G2 are also cheating in some apps.

1. CPU Explained


The Snapdragon 800 (MSM8974 model) in the Nexus 5 is a quad-core chip and it succeeds the Snapdragon 600 used in devices like the Samsung Galaxy S4 and the HTC One. Naturally, it’s an improvement in almost every aspect. (Right now, HTC is the only big phone maker that does not yet have a device powered by this chip, but this could change early next year.)

Now, Snapdragon is the name of the whole system chip. Each of the processing cores is of the latest Krait 400 type, an improvement over the Krait 300 used in the Snapdragon 600 series. One of the most notable differences is that Krait 400 is manufactured using TSMC’s new 28nm HPm process, a more advanced manufacturing technique allowing for optimizations in power usage and an increase in clock speeds. The Nexus 5 can thus run at clock speeds of up to 2.3GHz. Earlier devices with Krait 300 processing cores and Snapdragon 600 could only reach a maximum of 1.9GHz.

It’s a 3-wide out-of-order chip supporting a fairly deep 11-stage pipeline. It is characterized by the comparative reduced latency to main memory and a quicker L2 cache. The whole architecture is a 32-bit one, something that is bound to change on Android soon, and that has already changed in Apple’s ecosystem as it introduced its 64-bit Apple A7 earlier this year and it powers the iPhone 5s, iPad Air and iPad mini with Retina. Faring it against the A7, you'd notice that Apple uses 4 times more Level 1 cache (a type of static and extremely fast memory where often accessed data is stored), but half the L2 cache.

In the case of the Nexus 5, the Snapdragon 800 system chip works alongside 2GB of 32-bit dual-channel 800MHz LP-DDR3 RAM, the same as on the LG G2. 

1.1. Nexus 5 scores into top 3: Sunspider


The first test we run to see how the Nexus 5 fares against its rivals is the Sunspider Javascript Benchmark that measures on-browser javascript performance. The Nexus 5 scores into the top 3, an achievement that is quite remarkable given that it is actually more affordable than other phones and still outperforms them.

Keep in mind that Sunspider is the target of specific optimizations by various chip makers (definitely Qualcomm), so Snapdragon devices are unsurprisingly among the best ones here.

Lower is better
Sony Xperia Z1
746.2
HTC One
977.9
Samsung Galaxy S4
1082.9
LG G2
932.8
Samsung Galaxy Note 3
599
Apple iPhone 5s
415.7
Google Nexus 5
723.9

Mozilla Kraken is another stressful javascript benchmark. The Nexus 5 slips to the fifth place here, trailing closely behind its LG G2 Snapdragon 800 forefather. One notable thing is that unlike the widely popular Sunspider, Kraken is not the subject of optimizations by chip and phone makers.

Lower is better
Sony Xperia Z1
7636
HTC One
14598
Samsung Galaxy S4
12227
LG G2
7812
Samsung Galaxy Note 3
7917.6
Apple iPhone 5s
5931.8
Google Nexus 5
8871.7

2. GPU: Play on


The Nexus 5 features the Adreno 330 graphical chip running at clock speeds of up to 450MHz. This is the exact same peak frequency of the Adreno 320 (coming packed into the Snapragon 600 chip), but with optimizations and more resources, the Adreno 330 emerges as a clear winner over its predecessor. In the near future, we expect to see the very same Adreno 330 pushed to 550MHz in the MSM8974AB chip and some manufacturers like Chinese Xiaomi have already confirmed working on devices with this improved GPU. No other major phone maker has officially announced plans for such a device so far, though. The benchmark tests below show just how much of a difference it makes.

2.1. Basemark: Trailing close behind

The first benchmark we’ll look at to measure graphics performance is Basemark X 1.0. It runs game simulations on-screen and at 1080p off-screen. The Nexus 5 has a 1080p display, but since the on-screen buttons take up some of the space, the actual on-screen resolution is in effect slightly smaller so the device pushes less megapixels in the test and it’s on screen results are slightly higher.

For some devices like the iPhone 5s, though, the difference between on-screen and off-screen results is staggering because of the huge difference between the comparatively low resolution of its 4-inch display and the high resolution for offscreen tests.

Looking at the actual results, you’d see that the iPhone 5s leads but the Galaxy Note 3 is a close runner-up. Taking into account that the Note 3 has been proved to cheat this particular benchmark, we would say the Nexus 5 scores brilliantly here - on par with the Xperia Z1 and even outperforming its LG G2 forefather.

It’s also worth mentioning that this is a very stressful graphics test and we expect it to be at least a year until 1080p devices start hitting the coveted 30fps smooth framerate mark here.

Higher is better
Sony Xperia Z1
14.215
HTC One
7.677
Samsung Galaxy S4
8.566
LG G2
12.772
Samsung Galaxy Note 3
16.417
Apple iPhone 5s
16.757
Google Nexus 5
14.024

Higher is better
Sony Xperia Z1
17.007
HTC One
8.236
Samsung Galaxy S4
9.148
LG G2
14.662
Samsung Galaxy Note 3
18.1
Apple iPhone 5s
27.685
Google Nexus 5
17.342

GFXBench, formerly known as GL Benchmark, is one of the most GPU-intensive tests out there and it gives a detailed breakdown of a video card’s performance. Fill rates (the number of pixels a video card renders and writes to video memory) are the first thing we evaluate and you’d see that Apple is the clear leader, smoking all other devices.

We recommend you take a look at our detailed iPhone 5s performance review where we explain the amazing PowerVR G6430 graphics chip that powers Apple’s latest flagship.

As to the Nexus 5, it again scores perfectly on par with the Note 3 and Xperia Z1, its peers by platform.

Higher is better
Sony Xperia Z1
1372.3
HTC One
883.7
Samsung Galaxy S4
820.4
LG G2
1240.9
Samsung Galaxy Note 3
1393
Apple iPhone 5s
3317.1
Google Nexus 5
1361

The final two tests are another stressful game simulation - T-Rex. The Galaxy Note 3 pulls ahead here, and - interestingly - without any cheating. The device almost crushes the 30fps threshold. The Nexus 5 has the same Adreno 330 graphics chip, but scores a slightly lower 23 fps.

The T-Rex HD component of GFXBench is a demanding, GPU-centric test that simulates a graphically-intensive gaming environment on the screen. The results achieved are measured in frames per second, with more frames being better.

Higher is better
Sony Xperia Z1
23
HTC One
14
Samsung Galaxy S4
16
LG G2
21
Samsung Galaxy Note 3
26
Apple iPhone 5s
35
Google Nexus 5
23

Higher is better
Sony Xperia Z1
21
HTC One
15
Samsung Galaxy S4
16
LG G2
20
Samsung Galaxy Note 3
26
Apple iPhone 5s
23
Google Nexus 5
21

All in all, the Nexus 5 performs admirably. A $350 device, it is perfectly on par with much more expensive top-shelf Android phones when it comes to performance. It is clear that Google is subsidizing this handset hugely and just looking at the charts above you should not have any doubt this is the best value for the money device currently on the market.

Still, it is obvious that Apple’s iPhone 5s remains the top performer with its A7 system chip. Its advantage is not huge, but It’s clearly there. Comparing it to other Android devices, though, the Nexus 5 offers just as much in both CPU and GPU muscle.

The Nexus lineup first arrived with the idea to showcase the best of the Android platform, and more recently - the best of Android at an extremely affordable price. The Nexus 5 is no exception to that rule when it comes to its computing power.

Sunspider Lower is better
Sony Xperia Z1 746.2
HTC One 977.9
Samsung Galaxy S4 1082.9
LG G2 932.8
Samsung Galaxy Note 3 599
Apple iPhone 5s 415.7
Google Nexus 5 723.9
Mozilla Kraken Lower is better
Sony Xperia Z1 7636
HTC One 14598
Samsung Galaxy S4 12227
LG G2 7812
Samsung Galaxy Note 3 7917.6
Apple iPhone 5s 5931.8
Google Nexus 5 8871.7
Basemark X off-screen Higher is better
Sony Xperia Z1 14.215
HTC One 7.677
Samsung Galaxy S4 8.566
LG G2 12.772
Samsung Galaxy Note 3 16.417
Apple iPhone 5s 16.757
Google Nexus 5 14.024
Basemark X on-screen Higher is better
Sony Xperia Z1 17.007
HTC One 8.236
Samsung Galaxy S4 9.148
LG G2 14.662
Samsung Galaxy Note 3 18.1
Apple iPhone 5s 27.685
Google Nexus 5 17.342
GFXBench Fill Rate off-screen Higher is better
Sony Xperia Z1 1372.3
HTC One 883.7
Samsung Galaxy S4 820.4
LG G2 1240.9
Samsung Galaxy Note 3 1393
Apple iPhone 5s 3317.1
Google Nexus 5 1361
GFXBench T-Rex HD on-screen Higher is better
Sony Xperia Z1 23
HTC One 14
Samsung Galaxy S4 16
LG G2 21
Samsung Galaxy Note 3 26
Apple iPhone 5s 35
Google Nexus 5 23
GFXBench T-Rex HD off-screen Higher is better
Sony Xperia Z1 21
HTC One 15
Samsung Galaxy S4 16
LG G2 20
Samsung Galaxy Note 3 26
Apple iPhone 5s 23
Google Nexus 5 21

Related phones

Nexus 5
  • Display 5.0 inches
    1920 x 1080 pixels
  • Camera 8 MP (Single camera)
    1.3 MP front
  • Hardware Qualcomm Snapdragon 800, 2GB RAM
  • Storage 32GB,
  • Battery 2300 mAh
  • OS Android 6.0 Marshmallow

FEATURED VIDEO

103 Comments

101. Eycewind

Posts: 1; Member since: Nov 20, 2013

People who develop these tests need to make them more meaningful. There should be a threshold against each test (a vertical line for each bar graph). All devices that pass that line should be considered eqully good. This threshold could be a represntation of limits of our abilities or based on some other rocket science.This will help end user pay more attention to factors like UI, build quality, cost etc. that play significant role in everyday experience than just these numbers.

95. poncan

Posts: 1; Member since: Nov 12, 2013

I lol'd! at this one, first sunspider and Mozilla are sorftware based performance bench, most of these are largely single-threaded benchs which are highly dependent on software instead of actual CPU performance. In terms of GPU bench Basemark and GFXbench guess what? the reason "resolution" droidz be like "we on HD!" and most on the main list uses FHD i bet 5s would crawl on that off screen or on screen, no explanation needed... Apple is great at software optimization especially for the A7 at 5s/5c but hardware comparison in most parts, android phones are much to be compared to 5s/5c even with top tier China droidz. Anyways these competitions are good for end users/customers :)

94. TomTheNerd

Posts: 67; Member since: Oct 11, 2013

What about this?http://www.phonearena.com/news/Basemark-X-benchmark-now-listing-the-Nexus-5-as-the-2nd-best-gaming-phone_id48870 This?http://www.phonearena.com/news/New-Nexus-5-benchmarks-leak-reveal-stunning-graphics-prowess_id47749 And I saw another thread where they both scored 23? Anyway, these benchmarks mean nothing. PhoneArena, please do a side-by-side comparison. I predict Apple wins for browser (Safari) and N5 wins for apps and draw for gaming.

80. cj100570

Posts: 204; Member since: May 12, 2009

Awesome, a bunch of synthetic benchmarks that tell me absolutely nothing about how this phone will perform in everyday use.

76. skyuppercut

Posts: 1; Member since: Nov 05, 2013

It seems that chrome on windows 8.1 scores 1232 on sunspider where as on IE 11 it scores 588 ms What does it even mean.? Is my lappy's browser faster than galaxy note 3.? or IE is faster than chrome.? lappy config : (2nd gen i5, 2.3 gigs , 6gb ram)

75. Evil_SaNz

Posts: 259; Member since: Oct 20, 2011

Here you go N5 vs ip5s neck to neck video in real life usage: boot, browser (n5 with chrome that sucks but still there) apps, game. 350$ vs 800$ and 1080p vs 720p https://www.youtube.com/watch?v=aAqJty3eGhE&feature=youtube_gdata_player Please PA do a side by side comparison when possible, benchmarks are becoming silly.

71. Evil_SaNz

Posts: 259; Member since: Oct 20, 2011

Real life test on youtube shows us that n5 and ip5s are similar while opening apps and games with ip5s 1/2 second faster on Temple Run and n5 booting faster and browsing faster with a bigger screen with bigger resolution. So basically we consider faster a phone because it had a good benchmark and not because it seemed faster compared to another in real usage. Soon we will say an icecream is good because it is called "Good icecream". Check the performance comparison on yt. Where is this incredible advantage?

70. bugsbunny00

Posts: 2267; Member since: Jun 07, 2013

IPhone 5S smokes them all

68. Lakuta1978

Posts: 64; Member since: Jun 30, 2012

iP5s is the king....^^

60. Naepxpulse

Posts: 4; Member since: Oct 31, 2013

Even though apple looks to be dominating most of these benchmarks, top shelf phones are so fast these days that it really comes down to preference. I have switched back and forth between android and IOS and I am convinced they are both great. It just depends on what you prefer. It's awesome that the Nexus 5 preforms so well considering its price point, though.

57. robinlim85

Posts: 86; Member since: May 03, 2013

its just me or what,the comment for iphone performance being "the different is not huge, but its there" i find it offending when it crushes its competitor in ALL benchmark, with margin of 2 fold in some benchmarks,thats just weird saying "the different is not huge"

50. scriptwriter

Posts: 396; Member since: Nov 13, 2012

they're just numbers to me. The fact that some are a fraction of a second faster than others just doesnt bother me. You can very rarely notice any difference

49. mcshank

Posts: 47; Member since: Apr 24, 2013

Congrats to the Iphone for winning with its 4'in screen and below 720 quality. And Congrats to the 1080p Note 3 that was second at 5.7'in. Nexus 5 is pretty cool but sadly verizon likes to mess stuff up so we wont see it :(

44. Taters

Posts: 6474; Member since: Jan 28, 2013

Lol saying Apple doesn't cheat benchmarks is hilarious. They spin everything they do basically....

40. kabhijeet.16

Posts: 895; Member since: Dec 05, 2012

Nexus 5 is a Winner... Great performance at such low price... Thumbs up...

32. Bootutu

Posts: 228; Member since: Jul 11, 2013

Wow PA, nitpicking the benchmarks still are we, WTH is fill rate? its just part of the benchmark in GF, why don't you just give the overall T-REX 2.7 and GF 2.5, where is geekbench? etc where the iphone 5s is smoked, and what's the point of showing on-screen performance, jeeze this place....

31. ahhxd717

Posts: 332; Member since: Dec 08, 2011

Damn, that A7 processor is freaking fast. I'm excited to see what the other soc manufacturers bring to the table to compete with it. I wish we could see Apple chip in an Android phone, I just prefer the platform. But that will NEVER happen haha

48. Taters

Posts: 6474; Member since: Jan 28, 2013

A7 with android would have much poorer benchmarks....... iOS 7 is what makes those benchmarks soar. A perfect example is stick figures will animate much more smoothly than anything graphically intensive. iOS 7 basically runs in stick figure mode at all times except for gaming.

64. ahhxd717

Posts: 332; Member since: Dec 08, 2011

Yeah, I understand that point, but it would still be cool to see how this chip would perform with Android since there aren't similarly architectures chips on the androidsphere yet. But I'm sure Samsung and Qualcomm have some tricks up their sleeves

72. darkkjedii

Posts: 31797; Member since: Feb 05, 2011

Awww the sweet smell of hate. Sssssssniiiiiiiiiiiiiiffffff

30. machello

Posts: 35; Member since: May 03, 2013

So, Iphone 5S is the best? I actualy do own one, but I still love my Htc One...... Whats good for you is the only thing that realy matters.

29. Tuxedo

Posts: 356; Member since: Mar 19, 2013

iPhone 5S is mighty impressive.

27. deventure

Posts: 26; Member since: Apr 15, 2012

Imagine the scores if the 800 only had to push an 1136x640 screen. Thats why the Iphone loses GFX Offscreen when it has to go full resolution.

28. Ninetysix

Posts: 2966; Member since: Oct 08, 2012

http://arstechnica.com/gadgets/2013/10/galaxy-note-3s-benchmarking-adjustments-inflate-scores-by-up-to-20/ This is why the Note 3 is higher vs the other phones that has the S800 processor.

34. Bootutu

Posts: 228; Member since: Jul 11, 2013

There is nothing like adjustment, the CPU is just doing its job and running at full speed under stress, same thing with the Apple A7 processor running full speed on benchmarks, educate yourself, just because Apple doesnt provide application to check what their system is doing, doesnt mean they are not. Anandtech- "The second impact on performance is something I only noticed while digging around under the hood of the A7. It seems like the implementation in the iPad Air can, for whatever reason, hold more instructions in flight (over 20% more) than the A7 in the iPhone 5s. It’s unclear to me whether the A7 in the iPad is configured any differently via firmware/microcode or if perhaps we’re looking at a slightly different revision of the core, but the delta was repeatable in my testing. The third, and likely biggest change impacting the iPad Air’s implementation of the A7 is the additional thermal headroom afforded by the larger chassis. I’m not going to go into details on exactly what this next test does (unfortunately we’re going to occlude some of the low level work that we do in light of all of the benchmark cheating going on), but we’re looking at a curve of performance vs. time for a particularly power heavy mix of code. We’re running the same exact code on both the iPad Air and iPhone 5s here, the only real difference is the size of the chassis:" "You can see the 5s throttles back its CPU frequency to about 1GHz after the 2 minute mark. The crazy thing is that until that point the 5s manages to run at full frequency without so much as a hiccup for two full minutes, running an incredibly power hungry task. Given that most iOS apps aren’t this power intensive for such a sustained period of time, iPhone 5s users should almost always see the A7 running at a full 1.3GHz. Pretty crazy. The iPad Air by comparison shows much more controlled behavior. Early on in the test we see a 7.7% performance advantage, which lines up perfectly with the iPad Air’s 7.7% CPU frequency advantage. By the end of the test the iPhone 5s has throttled to 900MHz, while the iPad Air drops to around 1.2GHz. At this point the iPad Air’s performance advantage grows to almost 40%." http://www.anandtech.com/show/7460/apple-ipad-air-review/3

39. kabhijeet.16

Posts: 895; Member since: Dec 05, 2012

+1 to you sir... Nailed it...

41. Ninetysix

Posts: 2966; Member since: Oct 08, 2012

Right broski. S800 in one device gets 20%-50% higher score vs another device with the same S800 = no adjustment. Why don't you educate yourself? :) The code that is Samsung is using is not hard enough proof?..lol Rename the benchmark name to Bootutuisadummybenchmark and the Note3's score all of a sudden matches the Nexus 5/L2 G2/etc.

45. Bootutu

Posts: 228; Member since: Jul 11, 2013

Note 3 is running at 2.3GHz, 3GB of faster RAM, LG G2 is running at 2.26GHz, what's your point? http://www.gsmarena.com/samsung_galaxy_note_3-5665.php http://www.gsmarena.com/lg_g2-5543.php

46. Ninetysix

Posts: 2966; Member since: Oct 08, 2012

Go on...Now educate yourself and read the ars article. Rename the benchmark app so it bypasses the boost code. What do you get after renaming with your 2.3GHz, 3GB of faster RAM?

56. PhoneArenaUser

Posts: 5498; Member since: Aug 05, 2011

"The code that is Samsung is using is not hard enough proof?" Do you even realize what this code does? I'm sure that you even not able to understand it!

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 https://www.parsintl.com/phonearena or use the Reprints & Permissions tool that appears at the bottom of each web page. Visit https://www.parsintl.com/ for samples and additional information.
FCC OKs Cingular's purchase of AT&T Wireless