Samsung Galaxy S4 found to use benchmark boosting app to increase performance scores


Samsung Galaxy S4 Verizon Wireless

In recent times, we have all lay witnessed to a sort of spec arms race, sparked by smartphone OEM’s doing whatever they can to trump the other, and gain valuable market share. As consumers, we’ve helped feed into this, demanding our devices add more cores with every iterations, and even higher ppi’s than the last. If you needed further proof that this has gotten out of hand, take a look at what was just uncovered.

Tipped by @AndreiF in a post over at Beyond3D, the boys at AnandTech did some digging around and found that Samsung has deliberately been attempting to make themselves look good in benchmarks results by clocking the CPU/GPU higher in the Samsung Galaxy S4 only during benchmark tests.

We’ve said it time, and time again that benchmarks should never be used to judge end-user speed and, in most cases, should only be taken with a grain of salt. While many of us already know that, there are still those that scream device superiority over another simply because their device scored slightly higher on Quadrant. You have a small ___, we get it.

I guess Samsung also knows this and in an effort to hype up both their flaghip device and their in-house Exynos 5 Octa processor, they’ve seemingly built a special benchmark boosting app that triggers whenever a user opens popular benchmark tests. The benchmarks Samsung’s benchmark boosting app has singled out include GLBenchmark 2.5.1 (but not GFXBench 2.7.0), Quadrant, Benchmark Pi, and good ‘ol Linpack. What’s even more interesting is the fact that other benchmarking apps can actually call on benchmark booster to max out the GPU/CPU on the device, thanks to the built-in intent.

Benchmark boosting app

What happens is whether you’re checking Facebook or playing a graphically intensive game, the Galaxy S4 wont let its PowerVR SGX 544MP3 GPU reach above 480MHz. Scoping out girls on Instagram? 480MHz. Playing Crazy Taxi? Still 480. But, fire up one of Samsung’s specifically targeted benchmarks and watch as the GPU suddenly stands to attention, maintaining a solid 532MHz clock speed.

Similar results were found for the CPU where — during normal use — the Samsung Galaxy S4 maintains a steady clock of 250MHz/500MHz in its low-powered Cortex A7 portion of the CPU. Once again, open a targeted benchmark and the bechmark boosting app tells the CPU to immediately switch over to its A15 cores clocked to 1.2GHz — even before any actual benchmark tests take place.

Where this only seemed to affect the Exynos 5 Octa on the GPU side of things, this maxed out CPU clocking occurred even in the Snapdragon 600 equipped model where all 4 cores were plugged in and ready to rock n’ roll.

Whether Samsung is simply being misleading, or downright deceptive is up for you to decide. Just might be think again if you were looking to purchase a Samsung Galaxy S4 based purely off benchmarks (and not real-world experience).

Chris Chavez
I've been obsessed with consumer technology for about as long as I can remember, be it video games, photography, or mobile devices. If you can plug it in, I have to own it. Preparing for the day when Android finally becomes self-aware and I get to welcome our new robot overlords.

Transit comes to Android, brings real-time public transportation directions

Previous article

Samsung Galaxy Note 3 leaked again, this time exposing its internals

Next article

You may also like


  1. lol, Samsung sure will try to stay on the top no matter what it takes. Sabotage competitors, false accusation, now benchmark performance booster. No wonder no matter how high their smartphone scores, it always lags here and there.

    1. Same with Apple, Tony LIE. They sabotaged, sued and also had skewed benchmarks from paid Apple review/tech sites. Their devices showed wear after a couple versions, just doing basic spec bumps, below the competition. Noticeable lag and app/media crashes for no reason. That’s the apple way.

      1. The difference here is that you have no proof that Apple sabotaged, or skewed benchmarks. And yes they sued, but they also WON which kinda shows it wasn’t just to be anti-competitive.

        As for your other assertions about Apple, those are all opinions about what makes a good device whereas what we’re talking about here is Samsung willfully cheating benchmarks to deceive consumers into thinking their devices are better than they are.

        When the S4 and HTC One were fresh on the market, most reviewers weren’t really able to pick a “winner” but ultimately many of them gave the win to Samsung PURELY because the benchmarks were higher. So yes, there were consequences here for both HTC and consumers.

        1. “…to deceive consumers into thinking their devices are better than they are.”

          Not necessarily, the S4 is still doing all the work to acquire that score. They’re just being a little underhanded by invoking the device’s true potential only when it comes to running a little test. The results of the benchmark may be misleading in relation to day-to-day use, but it’s technically not false information.

          1. some people never accept. This is pure cheating by Samsung. No matter who says what.

          2. Please don’t insult my intelligence by presuming that I’m unaware of Samsung’s inappropriate actions. I know they’re “cheating” and I don’t think they should be allowed to get away with it, but my comments above are addressing something else entirely.

          3. Wouldn’t you be pissed if you bought a 400 horsepower car that did 0-60 in 3 seconds but it turns out that it really does 0-60 in 5 seconds because it was modified to only do 3 second 0-60 when the manufacturer wanted it to look better?

            Equate this to that. It’s deceptive and Samsung should be ashamed of themselves.

          4. I think you may have misunderstood me, as I am well aware of the situation and don’t require equations to be aware of the deceptive nature of Samsung’s actions. Yes they should be ashamed of their underhanded practices, but that wasn’t the point of my previous post.

            I was questioning a specific selection of beeny’s comments wherein they insinuated that the S4 isn’t capable of the power outlined within the benchmark results. This isn’t true, the S4 is capable of that performance, it’s just not overclocked to that performance by default. This is where the true deception occurs, and that’s the reason Samsung should be ashamed, not because they’re saying their product is better than it actually is (they aren’t saying that at all) but because they’re failing to specify that the device is only being overclocked during benchmarking to showcase it’s maximum possible potential.

            My response works pretty much the same for your car example so you can equate this to that. And no, I wouldn’t be pissed, just mildly annoyed. I can think of very few situations where I would need a car to go that quickly in that short of an amount of time anyway.

          5. Alright, your last paragraph was all I needed. You’re the type who is fine -excuse me, mildly annoyed- with a corporation misleading you.

            I don’t think any normal person needs to go 0-60 in 3 seconds. But if it’s advertised that way, it damn well better perform that way when I want it to.

          6. I’m not really that type, no. If I wasn’t able to unlock the vehicle’s ability to go 0-60 in 3 seconds you’re damn right I’d be pissed. But if say I could tweak some sort of on-board computer setting, or adjust a certain engine component to unlock those speeds then I couldn’t care less that they car manufacturer is trying to pull the wool over my eyes, but it would (as I said before) annoy me slightly.
            The same can be said about the S4. If I wasn’t capable of unlocking the overclocked benchmark performance myself to be used as I pleased then I’d likely never buy another Samsung product, hell if I had the money I might even attempt a law suit against them on the grounds of false advertising. But as it stands, the benchmark performance can be accessed with overclocking applications and/or a custom rom. So I remain slightly annoyed by the fact that Samsung has the audacity to think they can pull the wool over my eyes.

            I apologize for yet another wall of text.

          7. Exactly! I apologize for being snarky but I’m just tired, as a consumer, of these mega corporations being able to deceive us, and we don’t really have a choice other than to sit there and take it. Obviously Samsung isn’t the only example. I mean, there IS something we could do, but we’re sort of… complacent. We could all stop buying the products, but that’s not going to happen, so the alternative is we just take it and say thank you…lol

          8. If I’m paying upwards of 600 dollars for the device (off contract of course) then I don’t have to thank them ;).
            You don’t have to apologize for being snarky when it’s my fault for not explaining myself properly in the first place. I’m just glad that were both on the same page now.

            It is an unfortunate part of consumerism that companies will stretch the truth to borderline unlawful levels just to hype their product, and even more-so that our capitalist society allows this to happen, but I can’t think of a situation where such things have had a visible negative impact on my enjoyment of the products I purchase so I try not to let these types of things bother me to much.

            If and when it becomes a real issue, there will certainly be enough people in the world willing to ban together to effect change for the better. Until that time, it would likely do us well not to sweat the small stuff, Samsung likely doesn’t care that I’m wagging my finger at them anyway. :P

  2. So, Samsung borrowed Turbo Boost from Intel for the sole purpose of benchmarking? What SHOULD be done, however, is someone who is good at that stuff, figure out a way to enable it for all programs not just benchmarking.

    1. I believe that this is called over-clocking and can be done with root apps. At least, that’s how I’m interpreting this.

      I’m seeing this as setting the governor to performance, which leaves the cores at max: always.

      1. Yeah, I figured it was over clocking, I just went with the Intel Turbo Boost reference. I have seen some modified Kernels in the Sprint XDA forums that have overclocking ability and underclocking as well.

      2. Many roms boast overclocking abilities. That would likely be the most direct method of unlocking the S4’s true potential. Though rooting, installing a rom, and THEN overclocking the unit to maximum performance shouldn’t be necessary to get the power shown in the benchmarks which is what has people so riled. Samsung should have either allowed users I direct method of adjusting the performance levels themselves or not input the benchmark auto-adjustment to begin with.

  3. This is a “dirty little secret”… but we’ve kind of suspected this since the Galaxy S2 *wink*

  4. Many benchmarks are also poorly written.

  5. This was in the s3 as well.. lol I love how this news surfaces once a year with a bunch of ppl trying to put down samsung.. I bet htc has the same exact thing… literally this is 2-3 yr old news

    1. Maybe Samsung should start listing it in their features chart:

      5-inch full HD display
      Micro SD card slot
      S Voice
      S Camera
      S Notes
      S Benchmark Boost

      1. Maybe they’ll start a late-night infomercial as well,lending it even more credibility………………

      2. Like you’ve never exaggerated?

        1. Like you can never tell difference between lies and exaggeration?

          1. Like you’ve never lied before?

      3. Amazing…. :)

    2. lol I love how you get used to liars.

    3. just for anyone who didnt believe me this was pulled out of vzw s3 build vrbmf1

      Hello World!


      DVFS booster permission

      Allow application to send hardware boost intent. This permission is only available to Samsung preload applications.

      it does seem to only boost touchwiz apps though :/

  6. whaaaa?!

  7. phones on PEDS????so basically its the AROD, BARRY BONS , of smartphones

  8. iOS benchmarks were notorious for the same thing. Either way, the day to day consumer doesn’t care about benchmarks nor probably knows they do benchmarks on mobile devices. Furthermore, basic app functionality and even media playback requires little in the way of computing power so some benchmarks are a little overboard anyways.

    1. Let me guess – you own a Samsung device.

    2. How much you got from Samsung. please give us a proof that iOS did this

  9. Well is the S4 has the lines of code, i’m sure other devices do as well.

  10. I honestly could care less about benchmarks… I chose the S4 for a removable battery, SD card, and, uh, oh yeah, i hate Sense.

    1. So what you’re saying is, that you _couldn’t_ care less.

      1. A commendable troll… well played sir, despite knowing my precise meaning.

        1. Some of us are rather tired of folks not understanding the mistake of saying “could care less”. Nothing to accuse anyone of being a troll for. ;)

          1. It has nothing to do with the subject matter, and instead correcting language versus contributing despite knowing my exact meaning. This is, by definition, a troll.

  11. They’ve been ripping off Apple’s R&D for years, why would we expect them not be a bunch of liars too?

  12. Bad Samsung. But, lets recap for those who just refuse to acknowledge this: benchmarks suck. They’re stupid, irrelevant, inconsistent, and easy to manipulate. Anyone that uses these to determine what device to buy next is sadly naive and scheduled to be a victim of flashy corporate advertisement any minute now.

    I see that they have benefits to developers, hardware engineers, or people that want to tweak their device as a hobby to squeeze every ounce out of the components. But comparing one device to another and concluding anything but that the graphs look pretty is lunacy.

    1. Well put mark.

  13. Similar to what Ferrari’s been known to do to their press cars… http://jalopnik.com/5760248/how-ferrari-spins

    For the record – I’d still take a Ferrari :)

  14. Most veteran Android users know that in general, benchmarks don’t mean a damn thing. They are not accurate in the least are are only meant for those who don’t really realize this as an appeasement to say ….”hey look, my phone is superior to yours” crowd. This is not new news, and is nothing more than a “marketing ploy” geared to those who mistakenly believe that benchmarks are the end all final word. I’ll guarantee that this is implemented by more OEMs than just Samsung and once the code has been found in other OEMs (and it will), there will be no write up on this or any other blog about it…. At the end of the day….who cares, as long as you are happy with your phone.

    1. “nothing more than a “marketing ploy” geared to those who mistakenly believe that benchmarks are the end all final word”

      Absolutely correct. The average dullard consumer sees “high numbers” and gets excited easily. It’s the same thing camera makers do with their “pixel wars”. Higher pixels means better photos right? ;)

    2. Which means Samsung is the first to be found cheating its customers. lets see which other companies doing…

  15. I guess I’m the small minority who runs custom kernals, so unless I’m stock this might be an issue.

  16. Samsung sends spies into LG buildings, they hired students to bash Htc on the Internet and now this, Samsung is losing their credibility.

  17. I hope this doesn’t surprise anyone, Samsung has a past of doing shady things.

  18. Whether this is old news or not,it still ranks right up there w/the shenanigans of the anti-virus companies & their Munchausen by Proxy-style practices.

    All advertising for any consumer goods should be taken w/a grain of salt,especially when they claim their product “tests/polls” better than the competition.

  19. it’s only kinda bad i think, because it still psychically has to use all it’s power, vs, what the graphic card companies did in the early 2k’s, and wouldn’t a custom, or rooted rom, be able to access the same kind of power?

    1. Yea you root your phone and install a kernel control app like S-Teaks or TricksterMOD.

  20. Quack3.exe anyone?


  21. Just one more reason Samsung leaves a bad taste in my mouth.

  22. Just run one of the new Roms in XDA, the benchmark app will not be in it and the phone still kicks ass in all benchmark tests. So how do you figure that. Don’t listen to what anyone says about the phone you like. Try it yourself and make your own mind up. Samsung Galaxy S4 with any new Rom will kick all others in benchmark scores, that must answer any question if this phone is fast or not. IT IS. I don’t even own one. I have the HTC ONE but will not blast the SGS4, it is very fast.

    1. Yes you’re right, the phone is fast on other ROM’s. The “why?” is easy. The stock ROM has an inhibitor and the “boost” app is really just removing that inhibition. Anyone remember the 486 days of the “turbo” button? It’s the same kind of misnomer.

  23. Can we get some copy editing? Please.

  24. is this like snapdragon for vellamo benchmark? Samsung should at least mention that 533 MHz clock speed is used for benchmarks.. to be honest , with todays high end smartphones’ specs and hardware, all of them can run with no issues except that benchmark results entice gadget geeks into buying whatever has the best results.. benchmark results should be a non-factor now but instead look for daily-experience results..
    why is there so many articles that seemingly put down Samsung… have they checked HTC one if it has a similar tweaks in the system? they use the same soc and have almost similar benchmark results anyway.. right? or is the snapdragon GS4 version gives way better results compared to the HTC one’s?

  25. This article is misleading.. It tries to lure negativity against Samsung.. Would it be better if the title is, ” Samsung GS4 will not reach 100% full capability until you benchmark it”.. It means that the phone is physically capable of that benchmark score (truly-really fast) but that power it used to get that high benchmark score is limited in day to day normal use.

    1. You are right as far as the CPU tests. But the Eynos GPU will never hit 532MHz, except during benchmarks.

      1. or custom ROM, or if you call the intent, or if Samsung removes the inhibitor code. Careful how you use the word “never”. ;-)

        1. You are right. I should have prefaced that with “running the current stock firmware.”

    2. nah, Samsung lures negativity to itself.

    3. No need to “lure negativity against Samsung”. Samsung have generously stalked, trapped, and domesticated negativity all on their own. Negativity is curled up on Samsung’s lap purring as we speak.

      1. I am not talking about Samsung, I am talking about the journalism practice they did in this article. They try to give other meaning on the headline but when you read it, the story does not really represent it.

        This could be any article for Apple, Microsoft, Sony, Google… etc, My comment is about how they wrote the story.. If you don’t see it, then lucky you!

        1. The title of this article is, “Samsung Galaxy S4 found to use benchmark boosting app to increase performance scores”
          They have shown proof in this article that Samsung does indeed boost the benchmark scores with a boosting app. Do they not?

          I think it’s funny how some people will defend corporations even though there is proof they are being misleading. Why does that bother me, because I want to get the full performance advertised on my S4!

          If I buy this phone because I think, “this is going to be great, I won’t see any lag, even when I’m using multiple apps at the same time” but that isn’t going to be the case if they limit the processors during normal operations. Who gives a crap if the phone is “Capable” of performing something I never get to experience!!

          1. No they don’t. The SOC isn’t getting overclocked. That’s the stats the phone is capable of. Look it up. The question is “why is Samsung hamstringing the phone?” not “Why is Samsung boosting benchmarks scores”.

  26. I feel that only web sites care about that. People care more for battery times.

    That’s probably why samsung did that. They use clocks that allow a decent battery time but boost on occasion to show it’s potential.

    Me personaly have my HTC One X constantly underclocked to boost battery life. Everything works fine and I get more juice to get to the end of my day.

    It’s been for a long time now that I feel mobile power is useless. Most apps (dayly apps) are designed to work on most devices. That means that they must work fine on low end phones. So, what do I need 4 cores for if I’m not going to compile code or analyze space noise?

    What I really don’t understand about java and android is why a simple game app like candy crash where not power 3d graphics are used, but just a simple jewels moving in a static board can drain the battery in 2 hours. And the same applys to facebook games..

    To me they seem very poorly programed or it’s the whole system at fault. anyway it leads to really poor user experience…

    1. I agree. Also, its apps like facebook that are major battery hogs. Just try disabling facebook on your One X and use it for a day or two and you’ll notice that the difference in battery life is massive. There are a lot of other apps out there, most culprits being games, which without being too heavy on graphics still consume a lot of power.

      1. I dont have fb in my phone. only basic google eco-system with waze and some other utility apps (authenticators, backup, etc).

        I use to like FB messenger and it’s bubbles, but CM informed constantly that app stop responding (and became a processor hug) so I uninstalled. Now I only have hangouts for online messaging that doesn’t agree with everybody)

        I would install more things but I hate adaware infested apps (even many payed ones!)

      2. RE: Facebook – If you have Android 4.3 you can now revoke individual app permissions. That does wonders for apps like FB that want to access battery draining location services all day long. It is hidden in the settings but you can access the feature with the apps in the market.

    2. While I agree, the problem is that people only care about battery after they make their purchase. BEFORE making their decision battery life is irrelevant and all they care about is processing power, pixel density, screen size, etc. We are still at the stage of the smartphone industry life cycle where specs are driving sales, almost entirely. As long as consumer’s open their wallets in the direction of the highest clock speed then OEM’s will make fudging those metrics as important as real improvements.

  27. Ha! Maybe HTC should do this so they can turn a profit…

  28. so basically they are cheating…. that its so bad, but guess what it still wont matter since people will still buy samsung phones

    1. So basically they are using the phone’s hardware to in a test to see how the hardware performs. Not sure how this is cheating.

      1. You don’t see a problem with deceptive advertising? Like the phone only overclocks when it detects bench marking tests being run? And when benchmark tests are not running, the users can only get “normal mode”.

        It’s like an ISP throttling your network speed, but when you visit speedtest.net or other speed testing sites, they untrottle the connection. That’s perfectly fine because it’s the same physical network connection, right?

        1. How exactly is it deceptive? It’s not like they are overclocking the CPU/GPU, these are the exact specs published for this SOC. The question you should be asking is “why is there an inhibitor?” not “why are they cheating?”.

        2. Your ISP example is terrible because unlike an ISP, you are actually capable of overclocking your phone to the same speed. The ISP on the other hand is out of your control.

  29. I remember when intel falsely played a video instead of a directx 11 game for Ivybridge launch.

  30. Maybe this is how that guy burned down his house. His phone got stuck in God Mode.

    1. 1. IDDQD
      2. House on fire
      3. Profit?

  31. “Other apps can call on benchmark booster”. There isn’t anything misleading or deceptive about this. It simply puts the hardware into a maximum state which is useful information. It is showing you what the phone is capable of even if the current rom/kernel has it more limited for regular use. This is nothing like the GPU wars over a decade ago when ATI made a driver which specifically changed texture quality, anti-aliasing, resolution and other settings of popular benchmark applications in order to achieve better scores. The hardware is still performing the tests, so I don’t see that as cheating at all.

    1. That’s the thing, the phone ISN’T capable of running at maximum because Samsung caps it off at 480MHz no matter if running games, or browsing the web.

      This wouldn’t be deceptive in the least if Samsung simply allowed the device to reach it’s benchmarked potential, like it does on Snapdragon. It’s only the Samsung made Exynos that caps the GPU which is… just kinda shady.

      1. It’s an existing intent that can be called by another application. Not only that but you can install any number of custom ROM/kernel that include CPU/GPU govenor adjustment capabilities. the PHONE is capable the ROM has limitations. Misleading? Maybe, but ATI and Nvidia have done similar and much worse things in their arms race.

      2. Also I’d consider it deceptive if the Exynos wasn’t published as having the exact “boosted” performance specs you list in your article. The real questions are “has anyone noticed low performance in games or everyday use?” and “Why would Samsung hamstring the device that has such great stats”. Stop trying to demonize and sensationalize this and do some honest reporting.


    2. Although in real world results you’d probably never notice the difference, I think it is a bit deceiving.
      It’s like buying a car because it’s advertised as 444-hp but then they tell you “well it’s capable of that but we’re only allowing you to use 250-hp”

      1. Not really. It’s more like advertising a car that can do a quarter mile race in 14 seconds, but when you get it you discover that it has an inhibitor that uses GPS to detect if you’re at a race track before giving you access to the NOS tank used to achieve such times. The car is still capable of the performance and you can modify the car to remove the GPS inhibitor or program it to detect more locations than what it has in it. THAT is a much better analogy.

  32. The phone is actually doing the work so its not really cheating/misleading and I would bet other manufacturers do similar things. Personally I would never buy a phone based off a benchmark.
    Samsung seems to be somewhat underhanded in promotion type events … but … they are still giving updates to my GS2. How many companies push updates after 2 years? Samsung has earned some loyalty from me from continued support.

  33. lol Sammy keeping it classy. No wonder the S4 benchmarks 17% faster than the HTC One… because they were doping.

    1. doping implies they are going beyond their limits. The truth is that the ROM inhibits performance and this app simply is removing that limit for the benchmark. There’s no cheating, the hardware really is that good.

  34. I am Julletia Fiyong a citizen of Mexico but now recide in Numbai India,,i have been in a relationship with fiance for seven years, he suddenly started behaving violently and when i tried correcting him, he beat me up and broke up with me, I did everything possible to bring him back but all was in vain, I wanted him back so much because of the love I have for him, I begged him with everything, I made promises but he refused. I explained my problem to someone online and she suggested that I should rather contact a spell caster that could help me cast a spell to bring him back but I am the type that never believed in spell, I had no choice than to try it, I mailed the spell caster, and he told me there was no problem that everything will be okay before five days, that my ex will return to me before five days, he cast the spell and surprisingly in the fourth day, it was around 4pm. My ex called me, I was so surprised, I answered the call and all he said was that he was so sorry for everything that happened, that he wanted me to return to him, that he loves me so much. I was so happy and went to him, that was how we started living together happily again. Since then, I have made promise that anybody I know that have a relationship problem, I would be of help to such person by referring him or her to the only real and powerful spell caster who helped me with my own problem and who is different from all the fake ones out there. Anybody could need the help of the spell caster, his email is [email protected] you can email him if you need his assistance in your relationship or anything: [email protected]

  35. I would not put this past them, Samsung have been caught astroturfing on anandtech, hardforum, overclock.net and other hardware forums and earlier this year it was revealed they were paying students to promote Samsung and bash competitors on other language forums.

Leave a reply

Your email address will not be published. Required fields are marked *

More in Handsets