Galaxy Note 7 benchmarks: Exynos vs. Snapdragon head-to-head

It's always been our opinion that benchmark apps for Android phones really only tell you how well a phone runs the benchmark apps. They give a score that, while meaningful in some ways, tend to be used as some sort of gospel about how one phone is clearly better than another when that's not really the case. And they can be confusing for some folks who aren't exactly sure what any of those scores mean. We usually forgo running any or talking about them at any length and instead focus on the user experience.

This time, we're going to venture into the benchmark waters because we have both the Qualcomm Snapdragon 820 version and the Exynos 8890 version of the Galaxy Note 7 here. Samsung promises that users won't see any real-world performance differences between the two — and we agree — but we wanted to see the synthetic differences from a handful of benchmark testing apps in Google Play. We're not going to read too much into any of this, and we're certainly not about to declare any winner. Both models offer the same experience while you're using them. For the folks who do care about those numbers but may not have an opportunity to get both models, here are the results.

About the tests

We tested a Samsung SM-N930F (UK model) with an Exynos 8890 CPU and a Samsung SM-N930T (U.S. T-Mobile model) with a Qualcomm Snapdragon 820. Both phones have the same basic configuration — a personal account, a work account, and a small handful of apps from the Play Store installed.

Nothing special was done before we tested. We used the phone normally, installed the apps, then ran the tests without restarting or clearing anything or killing any apps. We even tested in the evening after a day of normal use. None of the phone's features were disabled. Really, we treated them like we do every phone and just ran the apps we installed to benchmark them.

Geekbench 3

Geekbench 3 has several tests that aim to simulate real-world CPU performance, and can utilize all of the processor cores. The results are calculated and returned as an overall indicator of your phone's processor performance in both single core tests and multicore tests. Higher scores are better.

As expected, the Exynos clearly calculates and processes data "better" than the Snapdragon. Surprisingly, the single core scores are both much lower than leaked benchmark results from early versions. We'll attribute that to two things — fully set up devices with apps and account synchronization running, and not clearing app cache and killing off the background processes. But keep in mind that a lower multi-core score on the Snapdragon 820 is expected because it has four cores, half the number as the Exynos 8890.

Download Geekbench 3 (opens in new tab)

AnTuTu 6

AnTuTu runs a series of tests of four key areas of your phone's performance — the processor, 3D graphics ability, memory, and user interface. The results are calculated and delivered as four individual scores, with a higher score being better.

The Exynos again scores better in processor performance testing, but the Snapdragon marks higher in 3D and user interface features. The RAM testing is as expected — both phones have the same memory modules and controllers, and the results are identical within an acceptable margin of error.

Download AnTuTu Benchmark (opens in new tab)


GFXBench measures graphics performance with a focus on long-term performance stability, render quality and power consumption. The tests use high-quality and high polygon 3D graphics, and we tested the off-screen versions of the Car Chase, both Manhattan and Manhattan 3.1, and T-Rex at 1080p.The results are the average frames per second (fps) for each test.

We are unable to compare the results of the render quality tests because our Exynos version would only return an Out Of Memory error. The Snapdragon version scored 2510 for the standard render quality test, and 3632 for the HQ render quality tests. These numbers are the PNSR (peak-to-peak signal-to-noise) values.

Download GFXBench (opens in new tab)

Hipxel's Disk Speed and Performance

This benchmark measures the read and write speeds of the internal memory in MB/s. The labels can be confusing as both tests (internal and external) are measuring the built-in internal memory and not any SD card or other attached storage.

The internal tests are the read and write speeds of the phone's application data storage partition. This is where the private data is stored and where most apps are installed. The external tests measure the speeds of the internal SD card storage, where media and some application data is stored.

The app will measure SD card performance, but we choose to pass on that test as we don't have two brand-new SD cards of the same make and model to use for a comparison. The results would be similar, and won't show any measurable differences.

Download Disk Speed/Performance Test by Hipxel (opens in new tab)

What does this mean?

Not a lot, really.

Sure, the Exynos has a "better" CPU and the Snapdragon has a "better" GPU if you go strictly by the numbers in these benchmarking apps. Both phones have the same memory and the same storage, as well as the same parts and logic to control each, and the results reflect that.

But using them will always count more than any benchmark numbers. As mentioned, we were curious and had both versions so we're sharing with people who are curious as well. Don't use these as proof of anything, and we'll call them both winners — it's like youth soccer where everyone wins a trophy because it's all good.

Feel free to do what you will with the numbers in the comments.

Jerry Hildenbrand
Senior Editor — Google Ecosystem

Jerry is an amateur woodworker and struggling shade tree mechanic. There's nothing he can't take apart, but many things he can't reassemble. You'll find him writing and speaking his loud opinion on Android Central and occasionally on Twitter.

  • It's the battery life than seems to really put them apart. The difference in screen on time for the exynos version of the S7 was...disturbing.
  • I agree with the conclusion that performance is very similar, but like other commenters pointed out, the exynos was reportedly significantly more power efficient. I would love to see some measurements on battery life. Posted via the Android Central App
  • Yes, interesting read but if you guys still have both machines... it really would be more informative to test battery life between the two. Thanks.
  • Where are the charts or tests that show this?
  • GSM Arena's review of the s7 and s7 edge compared the battery life between the two processors and the exynos had the better results.
  • Could you please share a link? I can't find it and am interested in reading Gold S7 Edge.
  • Cpu performance Battery life
  • scroll down for battery difference between Note 7 Snapdragon and Exynos
  • scroll down for battery difference between Note 7 Snapdragon and Exynos. Looks like about 2 to 3 hours average
  • The 820 has slightly better performance but worse battery life. There's always a tradeoff.
  • Interesting, to your point Jer, completely academic, but interesting none the less. I'd be interested to see a battery comparison too, but that takes more time than I'm sure you've all had with the device!
  • I'm curious about battery life differences, looked like there was a big difference in the S7, curious if it's the same for the Note 7. Posted via the Android Central App
  • If you disable the Samsung activities and processes, there is almost zero difference between the two. These services also report power usage incorrectly and merely show as "Android System" or "Android" in the list of apps that have used power. Hopefully, Samsung has worked on both issues for the Note 7. Better battery life and knowing what is killing your battery is an important feature.
  • What Samsung activities should be disabled? I know this is a Note 7 article but I have Galaxy S7 edge and am getting horrendous battery life the last couple of months. Since the internals are basically the same, I figured the Samsung activities would be the same. Thx in advance.
  • To properly do it, you'll need to root your phone and rename the actual Samsung stand-alone apk files as .apk.old files. Remember, this can void your warranty and it tripped knox on mine. Without root you can go into the settings and disable everything Samsung you can. If you're not seeing just Android System or Android as the thing using the majority of your battery, this probably won't help.
  • Thanks Jerry. I don't have root on my phone but that's exactly what I'm seeing as eating up a large % of my battery. Will disabling most/all Samsung services have any negative effects on my phone?
  • I had GS7E and it battery life it was great.
  • Yes, it will. But how, and when is determined by what you use your phone for, and what apps you use. I disagree with the shut off everything samsung theory. If you do your research there is stuff you can turn off, and depending what it is, it may, or likely may not make a difference. Really, who uses ANT+.... but turning it all off. Go buy a nexus and then leave it powered off, then you will have the maximum power savings.
  • Info like this becomes more important as it gets harder to root and mod our phones on the major carriers.
  • Thsnks for the info Jerry. Personally, I wouldn't root my phone (I use nfc payments on my phone which would stop working if I did this). However -- As for disabling 'everything samsung' -- could this possibly affect any TouchWiz or other samsung-specific features (I have a GS7 edge)? I'd love to extend my battery life as far as possible, but not at the expense of losing the value add features of TouchWiz (which I really like). -- Posted via my Titanium Silver Galaxy S7 edge --
  • Jerry, I would be really interested and I'm sure a lot of other people would be too in an article that discussed your thoughts on why there is such a difference between how the two processors were performing and what Samsung could do to fix it. I know when I first used the GS7E, I don't ever want to say it was slow because it wasn't, I would say it hesitated more than it should have when doing things. Thanks Posted via the Android Central App
  • So the samsung processes kill battery on the SD820 models but not the exynos models?
  • There is something wrong with your single core geekbench numbers. They should be in the 2200-2400 range.
    Having both 8890 and 820. I can tell you the eyxnos is a far superior soc. Think of this: it has 4 large cores that score 2200 while the 820 has 2. Having 4 big cores and 4 small cores gives a huge advantage of performance/battery life balance. I would say that the 820 has better games optimized for adreno and that the qualcomm modem might be better.
  • Something went seriously wrong with those Geekbench tests.
    Every single SD820 device is getting a score of around 2200 single core and 5200 multicore.
    And I'm sure the Exynos single core score should be higher as well.
  • These results are lower because he ran the tests after using the phone and not killing the running apps before running the tests Posted via the Android Central App
  • They could be lower because the tests may have been run right after the other.
  • I thought benchmarks weren't to be taken seriously because they are not done under real world settings. With a ton of apps installed open and a browser or two open, I know TW has features that are always on unless you force close them yourself in the app drawer in settings because TW slows down performance and eats the most power. Plus all samsung devices seem to slow down after 6 months from every person I have talked to especially with the s5 and s6 devices. Even though Google is minimalizing Android now to run buttery smooth on Low end devices and Top Tier, it seems OEM's UI skins still takes away from battery and performance, but maybe the new TW will fair better but maybe not from past experiences.
  • The problem with comparing tests from one site to another is no two tests are run the same.
    IE : there is no set standard to how the tests are run.
    As for general running of the device this will vary according to many factors, the services running on the phone, the apps running in the back ground.
  • Before reading the article, I just wildly predicted that the Exynos would perform better.
  • What about the real world? Battery life, SOT, connecting to phone network, LTE & WiFi download speeds, call quality?
  • All things we're looking at, they just take a little more time to process :)
  • Good to know :-)
  • I think that the N7 have only been out in the wild for about a week?
  • Does this phone support Unified Storage, where external SD Storage is merged with Internal Storage to create a single storage volume? It's the one thing keeping me from pulling the trigger. I cannot go back to two separate storage volumes in my phone. It's kind of horrible. Also, it isn't worth testing SD Card R/W Speeds if you don't have one of the top end UHS mSD cards from Samsung. Anything else is going to be pitifully slow (1/4th the speeds or less), particularly for writing to the card.
  • It doesn't out of the box, but I'm sure the same work around to enable it on the GS7 will work here too. I don't know why you'd want that, since every application that's stored on the SD card is going to run much slower.
  • 1. I'm not buying to use work-arounds. I just want a phone to work. I literally can't accept having to put in any work to get things to function. 2. Works fine on my M8, with much cheaper SD Card (and slower SD Card) than what is available on the market today. The Cards Samsung is giving away has 95 MB/s write speeds. The read speeds are going to be much faster. The Card I'm using in my M8 has 28 MB/s max write speeds and it's almost 3 years old... Compare that to the write speeds of the Exynos version of the the Note 7 in this benchmark, for example. It's not going to be a problem... Assuming Samsung uses a high quality SD controller in their device (maybe they skimped here to save pennies). 2A. Running an application off a memory card requires predominately reading data, which is typically faster than writing. Performance will not be an issue, and any tiny performance losses are worth the convenience of set-and-forget from Day 1. Once it is loaded, it's performance is not impacted. It's in RAM. You're overstating the situation to make it look grim, and make the lack of this feature seem less bad. I will wait to see if Apple is offering a 128GB iPhone for the same price as the 64GB currently are sold. If they are, I'll likely stick with that. I'm not going back to managing two storage volumes. It's management overhead that I an no longer willing to deal with. I never buy devices based on core "features" that can be gained (or recovered) through work-arounds, rooting, flashing, and replacement/redundant apps. I never have, won't, and never will. The same way I won't buy a phone with a bad UX simply because some third party launcher exists. I adhere to the K.I.S.S. principle when it comes to these devices. Sometimes a small compromise has to be made, and for me that is worth it. When I put the card in, the phone should ask me if I want to do this. I click yes, then confirm it and move on with my life. That's what the HTC does. I'm not sure why I should be okay with any other UX on any other Android device. The standard is set (for me).
  • I think samsung has been using slower sd controller because they always advertise up to 256gb, While Lg says 2tb.
    I have done some testing with some lg phone and samsung phones and the lg phones are faster
  • No, Samsung is just advertising the highest capacity sd card that currently exists, which honestly makes more sense. LG is listing a theoretical max capacity. Posted via the Android Central App
  • I have tested the s7 edge note 4 and lg g3 and the g3 performed better with the same sd card
  • I am also still rocking the HTC one M8 with a 32gb class 10 micro sd card, not adoptable storage though. Still love the Smartphone, best I ever had, and feel no need to upgrade to be honest. Do you feel the same?
    Will replace the battery in a repair shop by the end of the month though. 2 years of heavy use degraded battery life.
  • You might want to stick with HTC?
  • Adoptable* storage and no, according to other websites, but you can still move some apps over. There could be a workaround sometime later.
  • It's not for me. If an SD card fails, you're screwed.
  • Appreciate the work, but the major concern here is battery life, not so much performance. The Exynos showed around 30% battery life on the S7 models compared to Qualcomm. That's like chopping 800 mAh off the battery. Also, it is introducing a lot of uncontrolled variables to compare two phones after a day's use by two separate people and then run benchmarks. Have the same person do the same things on both phones, or just use two phones set up in the exact same way with the same apps open. At least make it somewhat scientific. A lot of us Canadians aren't too happy about our phones lacking the Exynos SoC and a proper benchmark showing not much difference in battery life might be enough to sway us away from going with the $500 cheaper OP3 or the older S7/Edge.
  • I hear ya, but that's just not possible right now. We did the best we could with what Samsung gave us. Based on using them, there's really no difference regardless of what these numbers show. Or any numbers. You can throw out most of what you read online about performance of anything because the person saying it isn't you, and these gadgets don't always follow a predictable pattern, For example, the phone I have here with the very best connection to a network based on call quality and signal numbers is the Zenfone 2 — a phone most people say is bad at both. Or the phone I use that has the best battery life (HTC 10) is different from what a lot of other people would say. I have no idea why. I do the same things on every phone I have here. It just happens. My take away from using the various Exynos and Snapdragon devices from Samsung is that in 2016, one may be better than the other but both are more than good enough. If you don't obsess over the "winner" of these sorts of things, you'll be plenty happy with either. But again — I'm not you.
  • When it is otherwise identical phones, used under identical conditions in identical manner on the same wireless provider, and one variation shows 30% worse battery life, I'm inclined to believe there's a significant difference there. Sure it might be off by a bit, but 30%?! That's huge. Even if it's half that, it's significant. Stuff like this is the raison d'etre of sites like this that review and compare phones. There used to be no reason (other than the SPen) for a Canadian to choose the Note 7 over the S7 Edge. And now, we see that not only is the battery smaller, it's also less efficient. As a long time Note user, I'm not impressed with the decision. I should have jumped ship after the SD expansion was removed last year.
  • How does this affect everyday use for the average consumer who doesn't care about this?
  • If everyday Joe runs down the battery on his S7 Edge to 30%, he still has 30% left. If he had bought a SD820 Note it would be dead and needing a charge.
  • Lol nice explanation
  • Thanks. Jerry!
  • With Phil off to do new "things" it got me thinking that Jerry would be great doing something along the lines of mods and root section for the readers who are into this kind of thing. Jerry is obviously pretty educated when it comes to code and what not.
  • Is this benchmark everything' for you guys ?. Its so pointless Posted via the Android Central App
  • Battery life.The discrepancy between exynos and snapdragon is unfortunate,especially for those of us stuck with the snapdragon. Posted via the Android Central App
  • I'd be happy to be stuck with SD820 ;)
  • Benchmarks are useless for everything, Antutu gives high scores to s7 and yet it sucks in gaming, not all, but lot of games stutter pretty heavily, same for Note 5 after 6.0.1 update, Note 5 on 5.1.1 ran all my games and tekken perfect. Exynos is superior to Qualcomm in battery life, i can get anywhere from 5.5hr to 7hr+ sot with S7E, i wonder if Note 7 can do more than 5sot with Exynos.
  • I think a lot of the time performance depends on how we as individuals have our phones set up, on occasion we download apps without really knowing what resources it uses and what effect it will have on the device.
    Some time ago I downloaded a google app, and a while down the track after that I used a battery saver app,
    upon looking at the stats and resources used by all the different apps I had on my phone , I found that very app I downloaded sometime ago , started itself and ran in the background consuming a considerable amount of CPU and battery without me even using it ...... Or knowing it was running in the background.
    Sometimes we are faced with options ...... Do you want to allow monitoring by app X to improve performance of the app and allow updates, etc, etc ...... For every single extra process running in the background we don't see our know about that often we don't even use , it reduces the performance of the device .
    And we just put it down to a **** device when often the problems are of our own doing imo.
  • So really this is just the continuation of the saying "lies, damned lies, statistics" ... and benchmarks. ;)
  • The best and more detailed reviews are from AnandTech. They are impressively techy and geeky in their testing. But they only tested the 820 version, so far.
  • Jerry, could you report back after a while with real-world performance differences? In the case of the S7, the Exynos seemed to perform slightly better in real-world use, alongside having seemingly better battery life.
  • These number nothing more than a just fun thing to do Mr. Jerry. It`s all depending on how we`re using our phones.
  • I would like to know when the new NEXUS phone coming out.
  • Go to the Nexus forum and find out
  • I just hope the S and Note 8 go with Exynos in all markets. That's what they did with the S6, right? Not sure why they had to go back to Qualcomm for the 7. Ps I wrote a forum post a week back with a collection of benchmark and real world comparisons of Exynos vs Snapdragon on the 7:
  • My guess is CDMA
  • Wow, guess numbers really don't matter, my 6P is hitting 1413/4460 in GeekBench, which is great but I imagine my battery life would be brutal running those speeds all day, great thing about custom rom/kernel is that all that power is available in just a tap on the screen.