Futuremark welcomes back the Galaxy S4 and Note 3 in 3DMark's ranklist

Futuremark welcomes back the Galaxy S4 and Note 3 in 3DMark's ranklist

In the beginning of March, Samsungdecided to do the right thing and remove the code that made some ofits Galaxy devices cheat on benchmarks. This happened in the latestAndroid 4.4.2 update for the Galaxy S4 and Galaxy Note 3. Now thatSammy is competing fairly in the never-ending hardware arms race,Futuremark, maker of the cross-platform benchmark 3DMark, hasre-listed the Galaxy S4 and Note 3 in its ranklist.

Allegedly, Futuremark was the onlybenchmark producer that acted against the inflated scores which someAndroid devices were coming up with. When Samsung and HTC (whose HTCOne and HTC One Mini smartphones were caught cheating too) were de-listed, thecompany's president, Oliver Baltuch, explained its principles asfollows:

“People rely on Futuremark benchmarksto produce accurate and unbiased results. That’s why we have clearrules for hardware manufacturers and software developers that specifyhow a platform can interact with our benchmark software. In simpleterms, a device must run our benchmarks without modification as ifthey were any other application.”

Hopefully, device cheating atbenchmarks has ended for good.

via GSMArena



1. tech2

Posts: 3487; Member since: Oct 26, 2012

'In the beginning of March, Samsung decided to do the right thing and remove the code that made some of its Galaxy devices cheat on benchmarks' - Luis D. Why is it 'wrong' to utilize the full potential of the device ?! The point for benchmarking is to see the maximum hardware capability so mods, to see how far they can tweak their ROMs. Also, you do realize they are not only for benchmarks some of the inbuilt apps (mostly S apps) do run at that peak performance, right ?

2. hung2900

Posts: 966; Member since: Mar 02, 2012

It is wrong because OEMs want to achieve both longer battery life and processing power (through the perception of benchmark scores). It would be nothing wrong if the code is true for any application, or at least apps need it, not benchmarking apps only. P/s: SONY has cheated for a LONG time, but Anandtech forgot to test on it (don't know the reason why, due to uncommon Xoerias??). So Sony bad@ss fans, please shu-t your mouth up in articles like this.

3. Ninetysix

Posts: 2965; Member since: Oct 08, 2012

Check the list of apps that are boosted and tell us again that it is not cheating. None of them are s apps btw.

4. rd_nest

Posts: 1656; Member since: Jun 06, 2010

Here's something you (as in everyone) should consider about benchmarking. Recently Anandtech reviewed Galaxy Pro tabs. One statement was striking: Sunspider, Kraken, and Andebench had the cores hitting a maximum 1.1-1.3GHz Now, this is really strange considering they should go upto 1.9GHz. So, what we see here is that the benchmarks are not "pushing" the processor to max limits. So, how do you trust benchmarks like that? The benchmark is supposed to give us the peak performance levels of the device. If the benchmark is not able to push the processor to highest stage, what is the use of such benchmarks?

6. wargreymon

Posts: 764; Member since: Nov 05, 2013

Either the benchmark isn't taxing enough or the optimizations they have made are throttling the tablet. THIS is the reason benchmark apps exist, to analyze how well you optimized your product in terms of battery vs performance.

13. rd_nest

Posts: 1656; Member since: Jun 06, 2010

If the benchmark isn't 'taxing' enough, that's not a good benchmark test.

23. wargreymon

Posts: 764; Member since: Nov 05, 2013

I agree, but in this case only a few out of thousands of devices have this problem. Which points to problem being those devices and not the benchmark.

9. AppleHateBoy unregistered

"Why is it 'wrong' to utilize the full potential of the device ?!" - tech2 Battery Life. Nobody don't wants a phone that utilizes the max. performance and busts the battery in 40 minutes. There is a reason some apps utilize only a fraction of performance. The extra performance isn't always worth the power. But your near and dear benchmarks don't take this into account.

12. Arte-8800

Posts: 4562; Member since: Mar 13, 2014

DO YOU SEE people driving 3.4,5,6 litre engine complaining about fuel consumption or driving v6 v8 v10 v12 engines... diesel petrol are for cheap skates tramps who dont have the money to drive good performance car, just like in phones there is no option to have a intel i3 or A57 CPU CLOCKED AT 2.5GHZ LIKE S801, cause it will rinse the battery in half, if only a higher 4500 mah battery is installed then you can see feel the diffrence of power like when i switch from stamina mode to speed performance on my laptop intel i7 i can see a huge diffrence. so in smartphones we should have the same opyion im a power user who loves power and not pathetic slow s801 cpu

20. AppleHateBoy unregistered

"if only a higher 4500 mah battery is installed then you can see feel the diffrence of power" THIS shows that you are an idiot.

21. PapaSmurf

Posts: 10457; Member since: May 14, 2012

Lol ignore him. He doesn't know what he's talking about.

26. Arte-8800

Posts: 4562; Member since: Mar 13, 2014

says the other kid who cant give nor provide infrences and just talks nonsense bogus blabla and calls others dime and thinks he knows everything.. you cant even form an idear or a presentation and backpedal knowing you no jack and then lose the dispute

29. AppleHateBoy unregistered

You won't find old people here. Other than yourself.

24. Arte-8800

Posts: 4562; Member since: Mar 13, 2014


27. AppleHateBoy unregistered

I am 16. Don't know about xperiaDROID though.

25. Arte-8800

Posts: 4562; Member since: Mar 13, 2014

A15 is a cpu at 9watts the s801 is 4 watts thats why the s801 draws less power, the higher the watts the more the battery drain go back to skool and learn more little kid

28. AppleHateBoy unregistered

Are you trying to say that A15 is better because it consumes more power?

30. Arte-8800

Posts: 4562; Member since: Mar 13, 2014

Having a conversation with you is like talking to a 12 years kid! I forgot also that your a Snapdragon fan, and a delusional one that you like the CPU and you have no information on what the current watts and what architecture the s800 is, how pathetic ,,,,!

31. AppleHateBoy unregistered

I bet you think the entire smartphone industry is full of Snapdragon fans. Because all smartphone OEMs use Snapdragon chips for their best products.

10. Arte-8800

Posts: 4562; Member since: Mar 13, 2014

no samsung did not cheat. its called smart processing where the system detects that a highend intense game is playing the GPU raises at 670MHZ which makes the game lag free, the higher the GPU MHZ the more frame rates will be at 55fps

16. Ninetysix

Posts: 2965; Member since: Oct 08, 2012

It only does this in the apps listed in the whitelist. If the name of the application/game is not in there, it doesn't boost the cpu/gpu. You can also change the name of your benchmarking application and make it run at "normal" speed. Do you even logic???

14. luis.d

Posts: 354; Member since: Dec 04, 2013

It's wrong because running a benchmark test on speeds higher than those allotted to "everyday usage" applications and processes results in unrealistic impressions about the hardware's practical speed. This is the first time we hear that factory apps cause the CPU to enter "berserk mode". While that's technically possible, we're sure that Futuremark's specialists had a good enough reason to exclude Samsung and other "cheaters" from the race until they got off the juice. Remember, they have nothing to gain from not having such popular devices on their list. Cheers!

19. _Bone_

Posts: 2155; Member since: Oct 29, 2012

Your logic is reasonable yet still flawed. You see every phone "cheats" 95% of the time, cause no app is running at maximum clock all the time. This "cheating" is otherwise known as OPTIMIZING PERFORMANCE for battery life, as you don't need 2GHz+ for an alarm clock or most of the other tasks. Enter the entirely different world of benchmarking, it's not "optimizing for endurance", but "let's freaking see MAXIMUM PERFORMANCE", and if some software such as an S-App or a benchmark tool is allowed to operate at those clocks, then that's the MAXIMUM the HW can do. There is no cheating, results are real, and brainy modders can unleash that 100%. You test everyday performance running everyday tasks. You test MAXIMUM performance at MAXIMUM possible clocks. Why else would you run benchmarks if not to see how far the HW can go? Which is exactly what the Note 3 did, take the cap off.

22. luis.d

Posts: 354; Member since: Dec 04, 2013

Your post is competent, but consider this - if benchmarks were truly made to push hardware to its absolute limit, past the frequencies which are recommended by manufacturers and available to apps by default, then benchmark developers would code them to do that. Instead, benchmarks are made to work within the practical limits imposed by manufacturers, on equal footing with other apps. By making their devices run slightly faster than this practical limit only when a benchmark app is detected, smartphone vendors consciously cheat, because they know that doing this will increase their scores compared to the competition's. While there's nothing wrong with pushing the hardware closer to its true potential, frankly, this isn't what the vendors intended - they did it to inflate results.

15. chocowii

Posts: 478; Member since: Jan 30, 2014

They did cheat. They only enlisted benchmarking apps on codes to have higher scores than usual. Thus, apps that are not on the list doesnt achieve the full potential. Its not good to see a phone runs at the best ONLY on benchmark apps and not on other apps or games.

5. Jason2k13

Posts: 1469; Member since: Mar 28, 2013

I personally believe that this is not cheating, reason is 'benchmark' tools were made to make the hardware run to it's full potential. Now please tell me why other apps need all this power? Each apps are made to run differently, some are more graphics intensive/battery hogging than others. And these benchmark apps are doing what they were supposed to do. Are people seriously gonna think its called cheating because there angry bird games are not getting the same power as benchmark apps?

7. wargreymon

Posts: 764; Member since: Nov 05, 2013

If you want to see full potential, look at reference devices for the SoC.

8. itsdeepak4u2000

Posts: 3718; Member since: Nov 03, 2012

I think this is not cheating. If some device don't use it's full hardware for performance then it's simply a waste to have it inside.

11. Arte-8800

Posts: 4562; Member since: Mar 13, 2014

correct just like cars you can modify and get an extra 50BHP AND 50LB OF TORQUE

17. TylerGrunter

Posts: 1544; Member since: Feb 16, 2012

Pretty much none of the devices you use can use the full hardware for performance, as they are termally limited. This means they can use the full potencial only for short periods of time or they will burn the SoC, then they have to run at slower frequencies to cool down. As this happens pretty much in every chip on the market (all have the so called boost frequencies) I don't get your point. -_- ??

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.