Samsung rigs benchmark scores to make Galaxy Note 3 seem faster

Well, well, well. Samsung thought it'd get away with a little bit of cheating, but it didn't take long to figure out that Samsung tried and failed to pull the wool right over our eyes. It turns out that the new Galaxy Note 3 is designed specifically with benchmarks in mind so that scores are inflated to make the device look faster than it actually is.

Discovered by Ars Technica, the Galaxy Note 3 kicks into high-gear whenever a benchmark app is launched and in fact will not let the CPU idle whatsoever. This high performance mode allows the phablet to run up to 20 percent faster, but just while the benchmark app is running. Once the test is complete and the app is closed, Samsung's device will return to its regular state and return to normal speeds. The clear benefit to doing this is to make the Galaxy Note 3 look faster than what it is to reviewers, technology enthusiasts and, indirectly, customers.

The Note 3 has a 2.3GHz Snapdragon 800 processor and so does the LG Optimus G2. Ars Technica knew something was up when Samsung's handset was scoring far better than LG's, yet the two feature the exact same processor. Monitoring CPU usage led to the realization that the Note 3 got an unfair boost in apps like Geekbench. In fact, using a renamed version called Stealthbench deactivated the power boost.

Samsung, you've been caught red-handed.

Source: Ars Technica | Image via Ars Technica

Report a problem with article
Previous Story

NVIDIA launching GeForce GTX Battlebox brand for high-end desktop PCs

Next Story

Valve trademarks Half-Life 3 name; development team revealed

233 Comments

Commenting is disabled on this article.

exynos, your defense of the point is ludicrous. you make some valid arguments, but you are being highly polarized in your view point. it's as if you are the person that had the brilliant idea of putting in the code on the phone to do the benchmark fudging.

if you say a benchmark is a tool to compare two equipments, then you want the two equipments to be comparable. no arguments. stop making excuses or looking to justify anything. period.

you would be pretty ****ed off if porsche for a benchmark run gave you the 400hp but on your everyday drive to work only gave you 300hp. as a paying client you have everyright to not have been misled.

(by the way car manufacturers fuel benchmarks are one of your "everyone does it" and still not morally right from everyone's point of view)

bjc4ever said,
Lol, how much time have you spent arguing the same point over and over again? Don't you have anything better to do?

I'm just making sure someone gets the whole point as someone clearly haven't done yet here.

I looked at the test in more detail and here are the facts. In all the benchmarking tests on the phones, in the case of the Galaxy Note III, all 4 cores were running at 2.3Ghz which is the max. However, even though they were all running at a full 2.3Ghz, they were not all pushed to 100%.

In the case of the LG G2, all 4 cores were not pushed to the full 2.3Ghz. It looks likes like, one core ran at full speed while the other cores were idle. ON another test, the same LG phone ran the cores at the full 2.3Ghz, but not at 100%.

In fact if you look at the facts of this article and the pictures http://www.slashgear.com/galax...igans-accusations-01299884/ none of the CPU cores on any of the test ever ran at 100% even if an when they cores were running at full clock.

What has been shown as fact is, on the Galaxy Note III, the phone calls up a java file which simply tells the phone to run all 4 cores at 2.3Ghz. Even though they are running at 2.3Ghz each, they are not all running at 100%. In fact in the test the combine percentage was roughly 4% of the total available power in one of the tests.

So my question is, how is this misleading? The facts shows the cores didn't even overclock, they were simply running at full clock speed.

Any of you who have a Windows PC can just open the task manager to see all your CPU cores. If you fire up an application like Photoshop, 3 or my 4 cores ramped up to full speed than they dropped. If you use your PC to convert a video file from one type to another, I have have render 4 files at one time as the software will force 1 core to each file. In this case, in the rendering, all 4 cores are at full 2.4Ghz and they are all at 100%. This is real world usage based on an application.

All Samsung did was tell the phone, when this benchmark tool is ran, run all 4 cores at a full 2.3Ghz. The fact that a typical user will never do this is because none of the apps they use are designed to run the cores at full speed. If an app ran the cores at full speed all the time simply because they are there, it would be a complete waste of battery and power. Benchmark tools typical run the hardware at the fastest possible speed for the task. Samsung simply made sure that under such conditions the hardware was running at full speed. The fact I will never reach full speed does matter.

If I buy a Ferrari F40, then I know the full speed of the care is 240mph. That doesn't mean when I get in may car and drive to the grocery store, I am going to do 240mph every time. What it does mean is that if the situation is right on an open highway with no cops or a track, hitting 240 mph won't be a problem.

The Note III calling up the Java app, is equal to me simply pressing the accelerator hardware and shifting to the highest gear when needed. How is that cheating when I am simply pushing my car to the speed it is capable of?

HipHopSinceFriday said,
I looked at the test in more detail and here are the facts. In all the benchmarking tests on the phones, in the case of the Galaxy Note III, all 4 cores were running at 2.3Ghz which is the max. However, even though they were all running at a full 2.3Ghz, they were not all pushed to 100%.

In the case of the LG G2, all 4 cores were not pushed to the full 2.3Ghz. It looks likes like, one core ran at full speed while the other cores were idle. ON another test, the same LG phone ran the cores at the full 2.3Ghz, but not at 100%.

In fact if you look at the facts of this article and the pictures http://www.slashgear.com/galax...igans-accusations-01299884/ none of the CPU cores on any of the test ever ran at 100% even if an when they cores were running at full clock.

What has been shown as fact is, on the Galaxy Note III, the phone calls up a java file which simply tells the phone to run all 4 cores at 2.3Ghz. Even though they are running at 2.3Ghz each, they are not all running at 100%. In fact in the test the combine percentage was roughly 4% of the total available power in one of the tests.

So my question is, how is this misleading? The facts shows the cores didn't even overclock, they were simply running at full clock speed.

Any of you who have a Windows PC can just open the task manager to see all your CPU cores. If you fire up an application like Photoshop, 3 or my 4 cores ramped up to full speed than they dropped. If you use your PC to convert a video file from one type to another, I have have render 4 files at one time as the software will force 1 core to each file. In this case, in the rendering, all 4 cores are at full 2.4Ghz and they are all at 100%. This is real world usage based on an application.

All Samsung did was tell the phone, when this benchmark tool is ran, run all 4 cores at a full 2.3Ghz. The fact that a typical user will never do this is because none of the apps they use are designed to run the cores at full speed. If an app ran the cores at full speed all the time simply because they are there, it would be a complete waste of battery and power. Benchmark tools typical run the hardware at the fastest possible speed for the task. Samsung simply made sure that under such conditions the hardware was running at full speed. The fact I will never reach full speed does matter.

If I buy a Ferrari F40, then I know the full speed of the care is 240mph. That doesn't mean when I get in may car and drive to the grocery store, I am going to do 240mph every time. What it does mean is that if the situation is right on an open highway with no cops or a track, hitting 240 mph won't be a problem.

The Note III calling up the Java app, is equal to me simply pressing the accelerator hardware and shifting to the highest gear when needed. How is that cheating when I am simply pushing my car to the speed it is capable of?


Bingo, you have understood it. +1 to you.

The fact that alot of the crybabies / Samsung haters in here doesn't even understands this makes the whoile thing a comedy at best.

they changed the name of the benchmark application and the result was different.
this is the fact.

and the fact that others do it doesn't make it ok: _No One Should_ it is deceitful to us as customers.

seamus2a2 said,
they changed the name of the benchmark application and the result was different.
this is the fact.

and the fact that others do it doesn't make it ok: _No One Should_ it is deceitful to us as customers.

All benchmark scores are deceitful. Fact, you will NEVER achieve those number in normal usage so it doesn't even matter.

seamus2a2 said,
they changed the name of the benchmark application and the result was different.
this is the fact.

and the fact that others do it doesn't make it ok: _No One Should_ it is deceitful to us as customers.


Because the CPU and GPU will throttle the speeds on them if they don't allows the benchmark apps to run at full speeds, witch makes the whole benchmarking thing a waste. And that's why the script takes cares of that so the benchmark apps overrun the speed throttling.

When you run a benchmark app on a Windows computer, do you think the speed on the CPU and GPU are throttling then?

Do you think my 3.7 GHz Turbo Speed CPU will throttle down to 1.8 GHz under a benchmarking run?

Read the facts - http://www.slashgear.com/galax...igans-accusations-01299884/

Read the facts - http://www.slashgear.com/galax...igans-accusations-01299884/

I think if people were not biased and read these 2 articles with a open mind, they would see Samsung did nothing. They didn't cheat. They did what everyone else did, and the phone does what it was designed to do. When apps that are designed to push the hardware harder are loaded, the hardware runs faster which is what it is suppose to do.

In the GS4 article, they stated as a fact, the GPU is designed to run up to 533Mhz and when a full screen app is loaded like games or benchmarking software, the GPU ran at 533Mhz like it was suppose too.

That is not cheating and this phone isn't cheating either. The fact is where the above devices had similar hardware, the Galaxy Note III simply performed better under the same conditions. Get over it!

HipHopSinceFriday said,
Read the facts - http://www.slashgear.com/galax...igans-accusations-01299884/

Read the facts - http://www.slashgear.com/galax...igans-accusations-01299884/

I think if people were not biased and read these 2 articles with a open mind, they would see Samsung did nothing. They didn't cheat. They did what everyone else did, and the phone does what it was designed to do. When apps that are designed to push the hardware harder are loaded, the hardware runs faster which is what it is suppose to do.

In the GS4 article, they stated as a fact, the GPU is designed to run up to 533Mhz and when a full screen app is loaded like games or benchmarking software, the GPU ran at 533Mhz like it was suppose too.

That is not cheating and this phone isn't cheating either. The fact is where the above devices had similar hardware, the Galaxy Note III simply performed better under the same conditions. Get over it!

Samsung denies benchmark boosting, yet a renamed benchmark achieves less performance.

Why lie if they've done nothing wrong?

you guys are crazy when you don't see the problem with this--or maybe you just aren't aware of what is going on.

When benchmarking, the idea is to be as systematic as possible. you should in fact benchmark all sorts of possible scenarios.

The most usefull benchmark is the "reality" benchmark, which is how anything will perform under real circumstances.

when your test case manipulates how it will behave based on the application being executed, and if this application is the benchmark, then this is for obvious reasons is unwanted because if the other test cases don't do the same, you cannot compare results.

it's as simple as that. if you try to argue otherwise you are suffering from some sort of serious brain or judgement deficiencies and maybe should stop to think about what you are saying.

if you still insist think about it:

that fancy galaxy note will have a useless program running wasting your resources executing code just to make sure that if a benchmark app is running all priority is given to it.

Question - Is it not a fact that benchmarking software is designed to push a device/computer hardware to excessive limits where a normal user isn't going to reach anyways?

So in this case, Samsung added a little boost by allowing the CPU to run faster during the tests.

Which is equal to the following fact. A computer modder will usually install software that will allow them to push hardware (overclock) to speeds no typical user will use anyway.

Fact - even if you consider it cheating it isn't. The fact is it shows that the hardware is more powerful than what the typical benchmarks shows and that it has been capped for good reason. Like battery-life, or to prevent excessive over-heating or the like.

I don't see the problem.
The articles heading is bogus. Samsung didn't rig the score, they rigged the phone to get a better score. That is a huge difference. he writer did this just to start a flame war with over-sensationalized titles to draw the lovers and the haters. Becoming like Engadget.

Edited by HipHopSinceFriday, Oct 2 2013, 3:52pm :

who even cares about benchmarks? I have never bought a phone by looking at numbers. If I like it, i buy it. Benchmarks dont mean anything.

alwaysonacoffebreak said,
So when Samsung does it it's not okay but when Intel has done this for years no one even cares?

Funny.

I didn't know Intel made smartphones...

stevan said,

I didn't know Intel made smartphones...


This isn't about smartphones, it's about how benchmarks apps are made and how they are used.

Exynos said,

This isn't about smartphones, it's about how benchmarks apps are made and how they are used.
What all the Apple fans and Samsung haters are all up in arms about is this fact:

In every model of the iPhone, the CPU was significantly under-clocked, likely to save battery power. That is why in the first 3 iPhone models, they used the exact same CPU, and simply clocked it up. The original CPU Apple used in the first 3 models weree capable of running at over 1GHz. So why did Apple only run them at 400Mhz, 612 and 800? IMO, it was because they were to lazy/cheap to simply make the battery bigger.

Look at today's hardware. What is different? Nothing. Its faster, its smaller and they all suck more juice but simply do it more efficiently. So how do you overcome more speed? You make the battery bigger. If you took a battery out of an iPhone 3G and placed it in today's iPhone 5S, it would drain in likely 1/2 the time as the battery does now.

The way I see this article is simple. Samsung allowed for its phone to run 20% faster in benchmark score so that there top device beats the competition. As we can see, the hardware is capable of more but was likely limited to same on power consumption or similar. It proves the Exynos/Snapdragon is a very powerful platform even if we as typical consumers will never use all of the power.

HipHopSinceFriday said,
What all the Apple fans and Samsung haters are all up in arms about is this fact:

In every model of the iPhone, the CPU was significantly under-clocked, likely to save battery power. That is why in the first 3 iPhone models, they used the exact same CPU, and simply clocked it up. The original CPU Apple used in the first 3 models weree capable of running at over 1GHz. So why did Apple only run them at 400Mhz, 612 and 800? IMO, it was because they were to lazy/cheap to simply make the battery bigger.

Look at today's hardware. What is different? Nothing. Its faster, its smaller and they all suck more juice but simply do it more efficiently. So how do you overcome more speed? You make the battery bigger. If you took a battery out of an iPhone 3G and placed it in today's iPhone 5S, it would drain in likely 1/2 the time as the battery does now.

The way I see this article is simple. Samsung allowed for its phone to run 20% faster in benchmark score so that there top device beats the competition. As we can see, the hardware is capable of more but was likely limited to same on power consumption or similar. It proves the Exynos/Snapdragon is a very powerful platform even if we as typical consumers will never use all of the power.

So they should call it the exynos benchmark rather than the Note 3 benchmark, if it is underclocked in the Note 3.

I presume you're saying that the iPhone 3GS used the same processor as the iPhone and iPhone 3G, but was simply clocked faster? If so, you never saw Apple showing benchmarks saying that the original iPhone was as fast as the 3GS did you?


So hardware is running faster ONLY when benchmarks app is running. How on earth is this NOT cheating? Some people will do anything to deny...

stevan said,
So hardware is running faster ONLY when benchmarks app is running. How on earth is this NOT cheating? Some people will do anything to deny...

Have you ever used a computer before where you have been running a benchmark app?

Clearly not.

Exynos said,

Have you ever used a computer before where you have been running a benchmark app?

Clearly not.

Like I said....

Some people will do anything to deny...

I think the majority of us here on Neowin have used benchmarking apps on our computers. The difference between a PC and a Smartphone is obvious. You don't put a bigger cooler on the CPU of your smartphone after the benchmark app found it can run at 120%.

stevan said,

Like I said....

I think the majority of us here on Neowin have used benchmarking apps on our computers. The difference between a PC and a Smartphone is obvious. You don't put a bigger cooler on the CPU of your smartphone after the benchmark app found it can run at 120%.


A smartphone like the Galaxy Note 3 / Samsung Galaxy S4 is a computer to. It's just that it can be used as a phone to. Or to say it more precise, it's a mini-computer. And as long it can be used as a computer, it should be allowed to run benchmark apps on it and it should be allowed to run any benchmark apps on it to determine how powerfull those devices actually are.

Exynos said,

A smartphone like the Galaxy Note 3 / Samsung Galaxy S4 is a computer to. It's just that it can be used as a phone to. Or to say it more precise, it's a mini-computer. And as long it can be used as a computer, it should be allowed to run benchmark apps on it and it should be allowed to run any benchmark apps on it to determine how powerfull those devices actually are.

You totally missed the point. No one is saying it can't run benchmark apps. It's just that the processor on it is clocked at 100% for every single app out there UNTIL it runs a benchmark app. Then it basically overclocks itself.

You used the example of desktop computers and benchmarking. So tell me this, when you benchmark a PC, you always put the frequency of the CPU in the results, right? That way you can differentiate between STOCK CPU and an OVERCLOCKED CPU.

So basically what Samsung did is run the benchmark on an overclocked CPU masked as a STOCK CPU. It's that simple. Not something you would sue the company for but the cheat is definitely there.

stevan said,

Then it basically overclocks itself.


What?

I think you have missed the whole point of this topic completely.

I can't do other than just LOL.

Exynos said,

What?

I think you have missed the whole point of this topic completely.

I can't do other than just LOL.

What?

This high performance mode allows the phablet to run up to 20 percent faster, but just while the benchmark app is running

stevan said,

What?


20% faster than the normal speed of what the normal apps are running. This is still within the default speed of the CPU and GPU.

Running Google Drive (to take an example) on your phone doesn't even need 20% of the full potential the CPU and GPU have on todays smartphones.

Again, read the topic and get a clue, KTNKBAI.

Exynos said,

20% faster than the normal speed that the normal apps are running. This is still within the default speed of the CPU and GPU.

Again, read the topic and get a clue, KTNKBAI.

I'm sorry man but you're hilarious.

Here's a proof to show just how miserable you are:

The Note 3 has a 2.3GHz Snapdragon 800 processor and so does the LG Optimus G2. Ars Technica knew something was up when Samsung's handset was scoring far better than LG's, yet the two feature the exact same processor. Monitoring CPU usage led to the realization that the Note 3 got an unfair boost in apps like Geekbench. In fact, using a renamed version called Stealthbench deactivated the power boost.

stevan said,

I'm sorry man but you're hilarious.

Here's a proof to show just how miserable you are:


It's getting even more hilarous that you still don't get it that the benchmark apps are allowed to run at the default CPU's and GPU's speed, witch still are the max default speed on them.

But still, even with the boost disabled, the Galaxy Note 3 is alot faster than the LG G2 to. OMGWTF SAMSUG USE MORE HAXES WAAAAAAH WAAAAH.

The point is that Samsung have optimized their Samsung ROM's to run better, in the same way as Samsung have optimized their Samsung ROM's to have longer batterylife on a TouchWiz ROM than you can get on a stock Google Edition ROM.

Go look on GSMArena if you don't believe me.

Exynos said,

But still, even with the boost disabled, the Galaxy Note 3 is alot faster than the LG G2 to. OMGWTF SAMSUG USE MORE HAXES WAAAAAAH WAAAAH.

Actually it isn't, it's somewhat faster but not a long. Not as much as the benchmark cheat made it seem. Look at the chart, then calculate the speed increases in % and you'll see.

stevan said,

Actually it isn't, it's somewhat faster but not a long. Not as much as the benchmark cheat made it seem. Look at the chart, then calculate the speed increases in % and you'll see.


It's still quite faster according to the picture in the article. This is the point.

I'm also gonna write down the same thing as i wrote down in the same post as you quoted from me as you most likely ignored it or didn't see it, but here it is: The point is that Samsung have optimized their Samsung ROM's to run better, in the same way as Samsung have optimized their Samsung ROM's to have longer batterylife on a TouchWiz ROM than you can get on a stock Google Edition ROM.

Go look on GSMArena if you don't believe me.

Exynos said,

It's still quite faster according to the picture in the article. This is the point.

I'm also gonna write down the same thing as i wrote down in the same post as you quoted from me as you most likely ignored it or didn't see it, but here it is: The point is that Samsung have optimized their Samsung ROM's to run better, in the same way as Samsung have optimized their Samsung ROM's to have longer batterylife on a TouchWiz ROM than you can get on a stock Google Edition ROM.

Go look on GSMArena if you don't believe me.

You mind replying me with some proof that Apple does the same thing?

They optimized their code to kick the processor to 100% when benchmark app is running. The fact that the benchmark app gets speeds from the cpu and gpu and those speeds are not available to many other applications can be called cheating.

I'm not blaming you for defending Samsung, it's just so obvious they cheated and so many different tech sites are writing about it.

stevan said,

You mind replying me with some proof that Apple does the same thing?

They optimized their code to kick the processor to 100% when benchmark app is running. The fact that the benchmark app gets speeds from the cpu and gpu and those speeds are not available to many other applications can be called cheating.

I'm not blaming you for defending Samsung, it's just so obvious they cheated and so many different tech sites are writing about it.


Because it's obvious that everyone does this, then YOU can rather find evidences that they don't do it?

Again, why shouldn't Apple do it when everyone else is doing it?

Exynos said,

Because it's obvious that everyone does this, then YOU can rather find evidences that they don't do it?

Again, why shouldn't Apple do it when everyone else is doing it?

Apple wasn't caught doing it. Apple also doesn't brag about benchmarks, they compare their hardware in keynotes to their previous gen and that's pretty much it. Others run the hardware for them.

You're defending this pretty bad even though a lot of users are disagreeing with you and a LOT of tech sites are saying opposite of you as well. Must make you right?

Samsung got caught cheating, giving the benchmark app access to resources other apps couldn't get. It's fair and simple.

stevan said,

Apple wasn't caught doing it. Apple also doesn't brag about benchmarks, they compare their hardware in keynotes to their previous gen and that's pretty much it. Others run the hardware for them.

You're defending this pretty bad even though a lot of users are disagreeing with you and a LOT of tech sites are saying opposite of you as well. Must make you right?

Samsung got caught cheating, giving the benchmark app access to resources other apps couldn't get. It's fair and simple.


Because no one have been able to look into their code in iOS 7 yet.

Does that makes sense to you?

No, Samsung gave benchmark apps access to use the full potential out of the hardware at stock speeds where other apps can do the same if they actually require such power. But yeah, most of the apps out to Android today doesn't even needs 40-50% of the full potential of the Galaxy Note 3 to run perfectly, so again, i don't see why they should run at 100% to just waste battery?

Edited by Exynos, Oct 2 2013, 1:31pm :

i dont see the problem? you're benchmarking what it can do , so what is the problem
im pretty sure my graphics drivers have optimizations for certain benchmarks

DKAngel said,
i dont see the problem? you're benchmarking what it can do , so what is the problem
im pretty sure my graphics drivers have optimizations for certain benchmarks

The problem is that they should disclose what you'll be benchmarking exactly (maximum hardware performance or actual device performance), because else any benchmark comparison is apples to oranges and defeats the whole purpose of benchmarking.

ichi said,

The problem is that they should disclose what you'll be benchmarking exactly (maximum hardware performance or actual device performance), because else any benchmark comparison is apples to oranges and defeats the whole purpose of benchmarking.

this is the real issue...if samsung says hey this phone is this fast then fine...but if it says hey this phone is that fast but not for you only for the benchmark then it's an issue

Some will just never understand the concept of how this is inaccurate. We all had this discussion in another thread with the G4 and some people didn't get it then either.

rippleman said,
Some will just never understand the concept of how this is inaccurate. We all had this discussion in another thread with the G4 and some people didn't get it then either.

So maybe you should all send every other benchmark app maker an e-mail and cry to them to not use the full potential of the computer because it doesn't shows what most of the users will get in performance as normal?

Halleluja, those arguments that the benchmark apps on a smartphone is not supposed to use it's full potential just because most users will never use an app that use that power is just retarded beyond anything, because that's not the point of a benchmarking tool. Because a benchmark tool is not there to give you results on how normal users or descent users use their computers or phones in the daily, it's to give us the results on how powerfull the hardware in those things is at full speed on the default speeds. Aka what they can do at it's full potential.

Get it?

just like last time, you simply just don't get it. Imagine if all industries tried rating things this way... /s

Exynos said,

So maybe you should all send every other benchmark app maker an e-mail and cry to them to not use the full potential of the computer because it doesn't shows what most of the users will get in performance as normal?

Halleluja, those arguments that the benchmark apps on a smartphone is not supposed to use it's full potential just because most users will never use an app that use that power is just retarded beyond anything. Because a benchmark tool is not to give you results on how normal users use their phone, it's to give us the results on how powerfull the hardware is at full speed on the default speeds.

Get it?

rippleman said,
just like last time, you simply just don't get it. Imagine if all industries tried rating things this way... /s


Every other benchmark tools that is out to computers does it. Why shouldn't Samsung do it then?

Or to explain it better. NVIDIA or AMD haven't created their graphic driver so every apps should use the GPU at full power either. They are designed to use the GPU after how much power it needs. And this is for normal usage. But when a benchmark tools is running, then the developers of that benchmarking tool will force the GPU or CPU or whatever to run at full power, simply because that's how it supposed to be.

if apple tried this, all you would be all over them for this exact same reason... people want REAL WORLD results they can use, not some hypothetical situation. (by the way, my car can get 1000/mpg... if i am going all downhill for the first 900 miles LOL)

Exynos said,

Every other benchmark tools that is out to computers does it. Why shouldn't Samsung do it then?

Or to explain it better. NVIDIA or AMD haven't created their graphic driver so every apps should use the GPU at full power either. They are designed to use the GPU after how much power it needs. And this is for normal usage. But when a benchmark tools is running, then the developers of that benchmarking tool will force the GPU or CPU or whatever to run at full power, simply because that's how it supposed to be.

but here is the key for why their testing is flawed... NO user can access it because its not actually reachable by any other means OTHER THEN the benchmark app itself.

Exynos said,

give you results on how normal users or descent users use their computers or phones in the daily

rippleman said,
if apple tried this, all you would be all over them for this exact same reason... people want REAL WORLD results they can use, not some hypothetical situation. (by the way, my car can get 1000/mpg... if i am going all downhill for the first 900 miles LOL)


No, Apple lets the benchmark apps for iOS use the full potential of their hardware to let everyone see how good their new devices perform. Why wouldn't they?

rippleman said,
but here is the key for why their testing is flawed... NO user can access it because its not actually reachable by any other means OTHER THEN the benchmark app itself.

And here ladies and gentlemans, is the reason why you should just ignore rippleman, because he CLEARLY doesn't know anything at all about this. Because in the article about the Galaxy S4 on the same thing, it was proven that many of the other Samsung apps on the Galaxy S4 was using the full potential of the CPU and GPU to, so yes, other apps are allowed to use the full potential as long it's an app that needs it and not just because it can.

i am glad i do not have blind faith in Samsung as much as you... it doesn't bother me that you do, I just disagree when people lie to get that blind faith... believe what you may.... you would anyways...

Exynos said,

And here ladies and gentlemans, is the reason why you should just ignore rippleman, because he CLEARLY doesn't know anything at all about this. Because in the article about the Galaxy S4 on the same thing, it was proven that many of the other Samsung apps on the Galaxy S4 was using the full potential of the CPU and GPU to, so yes, other apps are allowed to use the full potential as long it's an app that needs it and not just because it can.

rippleman said,
i am glad i do not have blind faith in Samsung as much as you... it doesn't bother me that you do, I just disagree when people lie to get that blind faith... believe what you may.... you would anyways...


Wow, such an amazing counterargument to what i was saying, lol. Oh wait, there isn't anything to counterargument for, because there isn't any. I just told the truth.

it can not be explained to you any more simply than it already has been. As stated before, you just simply don't understand what we are all saying, or more likely, you don't want to see your faith have faults in it. (with either G4 or Note 3 topic). Will I buy a Samsung mega next week? Yes.

Exynos said,

Wow, such an amazing counterargument to what i was saying, lol. Oh wait, there isn't anything to counterargument for, because there isn't any. I just told the truth.

rippleman said,
it can not be explained to you any more simply than it already has been. As stated before, you just simply don't understand what we are all saying, or more likely, you don't want to see your faith have faults in it. (with either G4 or Note 3 topic). Will I buy a Samsung mega next week? Yes.


Says the guy who can't even explain what benchmark tools are mainly been developed to do, then i'm pretty sure that i and others here wont take you seriously here. Simple as that.

Benchmarking tools have one purpose, and that is to let us see what the full potential of the hardwares is capable of doing.

Those who doesn't think that's the way to go, then you are not on the same benchmarking planet as the rest of us have been for over a decade by now.

i would guess most on neowin would know the purpose of a benchmark program, most also understand that giving a benchmark number that is unachievable under any circumstance is called cheating.

Question: can they also claim that they get 2 weeks battery use out of the Note 3? Would you consider that misleading or accurate to publish?

Exynos said,

Says the guy who can't even explain what benchmark tools are mainly been developed to do, then i'm pretty sure that i and others here wont take you seriously here. Simple as that.

Benchmarking tools have one purpose, and that is to let us see what the full potential of the hardwares is capable of doing.

Those who doesn't think that's the way to go, then you are not on the same benchmarking planet as the rest of us have been for over a decade by now.

rippleman said,
i would guess most on neowin would know the purpose of a benchmark program, most also understand that giving a benchmark number that is unachievable under any circumstance is called cheating.

Question: can they also claim that they get 2 weeks battery use out of the Note 3? Would you consider that misleading or accurate to publish?


If you want to find out the normal usage performance, you can simply press CTRL + ALT + DEL to go to the performance viewer there who shows you the CPU usage of every CPU Cores and the RAM usage and so on. If you want to find out the full potential of you hardware, you have to run a benchmark tool to find that out.

It's so simple. Why cry over something that is avaliable for both things?

And do you actually know the difference from normal usage to benchmarking?

Clearly you don't.

I am surprised that someone involved with tech can just simply NOT GET this, frustrates me to no end... I made a picture to show you to give you ONE single last way to see what everyone is saying...

link: http://i39.tinypic.com/am71jm.gif

If you don't get it after that, I quit trying to explain it.


Exynos said,

If you want to find out the normal usage performance, you can simply press CTRL + ALT + DEL to go to the performance viewer there who shows you the CPU usage of every CPU Cores and the RAM usage and so on. If you want to find out the full potential of you hardware, you have to run a benchmark tool to find that out.

It's so simple. Why cry over something that is avaliable for both things?

And do you actually know the difference from normal usage to benchmarking?

Clearly you don't.

rippleman said,
I am surprised that someone involved with tech can just simply NOT GET this, frustrates me to no end... I made a picture to show you to give you ONE single last way to see what everyone is saying...

link: http://i39.tinypic.com/am71jm.gif

If you don't get it after that, I quit trying to explain it.


You can try to explain whatever you like, but as long as you haven't explained to me what the main purpose of a benchmark tool is, then you can rather be there wasting everyones time with your wall of nonsense.

Benchmarking in all cases I have seen is the showcasing of the product under BEST CASE scenarios. Under Samsung's cheating way, you/me/anyone can NEVER achieve BEST CASE scenario. Its simply unattainable.

Exynos said,

You can try to explain whatever you like, but as long as you haven't explained to me what the main purpose of a benchmark tool is, then you can rather be there wasting everyones time with your wall of nonsense.

rippleman said,
Benchmarking in all cases I have seen is the showcasing of the product under BEST CASE scenarios. Under Samsung's cheating way, you/me/anyone can NEVER achieve BEST CASE scenario. Its simply unattainable.


Stop being butthurt that Samsung does this like everyone else. Even HTC and LG are doing the same as Samsung, because they know it's the right way to go for finding out how your device performs: http://www.slashgear.com/galax...igans-accusations-01299884/

again you fail to see that your "device" will NEVER perform this way. And i am defiantly not butthurt. As I said earlier, will be getting a Mega as early as tomorrow. I just find it sad that Samsung has to lie to get their fans and the "Android MORE COREZ" crowd to buy their phones over the competition.

Exynos said,

Stop being butthurt that Samsung does this like everyone else. Even HTC and LG are doing the same as Samsung, because they know it's the right way to go for finding out how your device performs: http://www.slashgear.com/galax...igans-accusations-01299884/

rippleman said,
again you fail to see that your "device" will NEVER perform this way. And i am defiantly not butthurt. As I said earlier, will be getting a Mega as early as tomorrow. I just find it sad that Samsung has to lie to get their fans and the "Android MORE COREZ" crowd to buy their phones over the competition.


I can get my device to perform this if i want to without spending any energy / time on it. Just because some apps wont run at full speed, because they don't have to run at full speed since they can't be any better than they already are, then i see no reasons why they should run at full speed just because some benchmarking apps have to.

I don't care what other apps does or use out of the CPU and GPU, because i do only care about what my device is capable of under benchmarks.

Again, you are comparing normal usages to benchmarking, witch makes the whole thing stupid.

And like i also have been asking for. What are the main purpose of a benchmarking app?

Is it to determine how you use your smartphone / computer under normal conditions or how your smartphone / computer perform at full power?

stevan said,

So Androids then....?


Apple does it to. Oh, and every other benchmarking tools out there to computers does it to like they always have been doing.

Nothing new here, so just move on.

Exynos said,

Apple does it to. Oh, and every other benchmarking tools out there to computers does it.

Nothing new, so just move on.

Apple does it? When and where?

stevan said,

Apple does it? When and where?


Do you think Apples dual core-CPU's are getting those benchmark scores without running it's hardware on full power under those benchmark tools?

You are deep in a dreamworld if you think they doesn't.

Exynos said,

Do you think Apples dual core-CPU's are getting those benchmark scores without running it's hardware on full power under those benchmark tools?

You are deep in a dreamworld if you think they doesn't.

You still haven't provided a proof? In my dreamworld we have proof and facts.

Oh and a sidenote, the most hilarious thing about your argument is that all of a sudden, this phablet/smartphone is now a PC since benchmarking tools are involved.

My words about some people being so delusional that they will deny and defend anything is right on.

stevan said,

You still haven't provided a proof? In my dreamworld we have proof and facts.

Oh and a sidenote, the most hilarious thing about your argument is that all of a sudden, this phablet/smartphone is now a PC since benchmarking tools are involved.

My words about some people being so delusional that they will deny and defend anything is right on.


Do you have any proofs that they don't?

It's it quite obvious that they will do it when everyone else does it?

Exynos said,

Do you think Apples dual core-CPU's are getting those benchmark scores without running it's hardware on full power under those benchmark tools?

yes they are, actual real reachable unreserved speeds. They would be even higher if they "cheated" on the numbers and ONLY overclocked it for benchmarks

rippleman said,

This could be where you been missing the point... no... you can't.


Yes i can. It take me 2-3 clicks for me to do it. If not, you are a noob with Android.

I can get every apps to run at 1.9 GHz (witch is the default speed on the Galaxy S4) by changing the governor to 'performance' on the CPU. If not, the apps would just throttle the CPU at the speed the app it self wants under the 'ondemand' governor.

Edited by Exynos, Oct 2 2013, 3:41am :

Exynos said,

Yes i can. It take me 2-3 clicks for me to do it. If not, you are a noob with Android.

So you missed the point where it says that if games, which usually max out the cpu and gpu, ran at these frequencies they would possibly fry the processor.

They are giving a benchmarking app more power to run than games but somehow nothing fishy there....

LOL

this is like telling a religious person their faith is wrong, they would rather die then have the truth be accepted. The phones are great phones, but to defend it blindly till death is just silly. No matter what we say, no matter what is shown, no matter how clear the numbers are in front of them, if they just cant percieve it, they just cant see it. (we are all like that with some things)... I always think about what people say and try to see if i am just blind to it myself in certain subjects but one can never know.

stevan said,

So you missed the point where it says that if games, which usually max out the cpu and gpu, ran at these frequencies they would possibly fry the processor.

They are giving a benchmarking app more power to run than games but somehow nothing fishy there....

LOL

stevan said,

So you missed the point where it says that if games, which usually max out the cpu and gpu, ran at these frequencies they would possibly fry the processor.

They are giving a benchmarking app more power to run than games but somehow nothing fishy there....

LOL


Because a benchmarking tool only lasts for some few minutes while games lasts for many hours and even days.

There is a difference.

But i still want to know the full potential of the hardware. So that's why i'm benchmarking the phone then.

rippleman said,
this is like telling a religious person their faith is wrong, they would rather die then have the truth be accepted. The phones are great phones, but to defend it blindly till death is just silly. No matter what we say, no matter what is shown, no matter how clear the numbers are in front of them, if they just cant percieve it, they just cant see it. (we are all like that with some things)... I always think about what people say and try to see if i am just blind to it myself in certain subjects but one can never know.


Haha, try backpedaling more. It's funny and it makes me to consume more popcorn.

Exynos said,

Because a benchmarking tool only lasts for some few minutes while games lasts for many hours and even days.

There is a difference.

But i still want to know the full potential of the hardware. So that's why i'm benchmarking the phone then.

Thank you so much! I was waiting for this reply. Now here is the reply from Samsung when they got caught doing it with the S4.

The maximum GPU frequencies for the Galaxy S4 have been varied to provide optimal user experience for our customers, and were not intended to improve certain benchmark results. We remain committed to providing our customers with the best possible user experience.

Samsung claims that they designed it in this manner for the benefit of the user. If that had any truth, then you would have seen the names of a number of games in that list where they force the clocks. But you dont see that. Instead you see the names of benchmark apps. Why?

stevan said,

Thank you so much! I was waiting for this reply. Now here is the reply from Samsung when they got caught doing it with the S4.

Samsung claims that they designed it in this manner for the benefit of the user. If that had any truth, then you would have seen the names of a number of games in that list where they force the clocks. But you dont see that. Instead you see the names of benchmark apps. Why?


That's what Samsung does yes. Not what the developers does with their benchmarking tools on the phones.

Exynos said,

And here ladies and gentlemans, is the reason why you should just ignore rippleman, because he CLEARLY doesn't know anything at all about this. Because in the article about the Galaxy S4 on the same thing, it was proven that many of the other Samsung apps on the Galaxy S4 was using the full potential of the CPU and GPU to, so yes, other apps are allowed to use the full potential as long it's an app that needs it and not just because it can.

If it's supposed to do that as you said, how come renaming the benchmark app doesn't give the same result even though it demands the same resource?

Stoic Little said,

If it's supposed to do that as you said, how come renaming the benchmark app doesn't give the same result even though it demands the same resource?


Because Samsung is letting the benchmark apps to use the full power of the hardware that way instead of letting the CPU and GPU throttle control it so you don't get the results you should get for testing the hardware to it limits.

Exynos said,

Because Samsung is letting the benchmark apps to use the full power of the hardware that way instead of letting the CPU and GPU throttle control it so you don't get the results you should get for testing the hardware to it limits.

So only benchmark apps can achieve those results. This is what everyone has been saying is wrong, yet you appear to accept this.

Benchmarks should utilise the full power of the CPU/GPU that are available to EVERY APP, if required. There's no reason why a benchmark should run faster than a game, or a video encoder or something. That clearly just inflates the results. That shows the phone as having unachievable performance other than in the benchmark apps themselves.

It's wrong, whichever way you are Samsung try to spin it, and no, not every other company does it.

Hardcore Til I Die said,

So only benchmark apps can achieve those results. This is what everyone has been saying is wrong, yet you appear to accept this.

Benchmarks should utilise the full power of the CPU/GPU that are available to EVERY APP, if required. There's no reason why a benchmark should run faster than a game, or a video encoder or something. That clearly just inflates the results. That shows the phone as having unachievable performance other than in the benchmark apps themselves.

It's wrong, whichever way you are Samsung try to spin it, and no, not every other company does it.


Other Samsung apps are allowed to run at full speed + some other apps that need such power as it was explained in the same topic about the Galaxy S4 is also allowed to use 100% of the power the hardware can deliver.

Why do other apps that doesn't need the 100% power output needs to be able to run at 100% and rather waste battery and don't become any more effective than just using 40-50% of the full potential of the hardware?

I'am allowed to run Winamp 5, Firefox or Google Drive at 100% load on my normal computer?

No, and why should i when the apps never will even need 20% of the full potential of the hardware in my computer?

Just because benchmark apps needs to use 100% out of the hardware it doesn't mean everything else have to do it.

So yes, everyone who makes benchmarks apps makes sure to use 100% out of the hardware potential.

stevan said,

You still haven't provided a proof? In my dreamworld we have proof and facts.

Oh and a sidenote, the most hilarious thing about your argument is that all of a sudden, this phablet/smartphone is now a PC since benchmarking tools are involved.

My words about some people being so delusional that they will deny and defend anything is right on.

Question, did you even read the article he linked? http://www.slashgear.com/galax...igans-accusations-01299884/

The writer in this article said that benchmarking tools push the hardware using all 4 cores while most applications barely used 2 cores. We already know that even on PC's, no matter how many cores your PC has, most apps that are used everyday only push 1 or 2. The only apps that use more power are media creation tools like those from Adobe and high end games.

I do not understand what you are debating. Do you? The point is simple, Samsung didn't cheat. They allowed benchmarking hardware to use more power of the already existing hardware. The hardware is factually capable of doing these numbers which mean nothing anyways. If the hardware wasn't capable and Samsung actually inflated the numbers and we ran the same tests and got lesser numbers, that would be different.

What you fail to get is, the hardware did these numbers because it was allowed to run faster than normal. So what?! Benchmarking software already push hardware to potential beyond normal usage.

The article states that HTC and even LG did so as well. So why all this hate for Samsung which is another fact you are missing.

Oh and here is nothing thing. Apple announced during the iPhone 5S release that it was MORE than double the speed of the previous model, I repeat...they stated it was more than double, yet benchmarks show the 5S was not. It was roughly 20-30% than the previous model, not 50% faster or above. Would it be safe to say that Apple artificially inflated their numbers? Hmmm. Biased much?

last post to you about this: your apps on your samsung phone are indeed getting 100%. What everyone is trying to say is that ONLY the benchmarks are allowed 120% of your power

Exynos said,

Other Samsung apps are allowed to run at full speed + some other apps that need such power as it was explained in the same topic about the Galaxy S4 is also allowed to use 100% of the power the hardware can deliver.

Why do other apps that doesn't need the 100% power output needs to be able to run at 100% and rather waste battery and don't become any more effective than just using 40-50% of the full potential of the hardware?

I'am allowed to run Winamp 5, Firefox or Google Drive at 100% load on my normal computer?

No, and why should i when the apps never will even need 20% of the full potential of the hardware in my computer?

Just because benchmark apps needs to use 100% out of the hardware it doesn't mean everything else have to do it.

So yes, everyone who makes benchmarks apps makes sure to use 100% out of the hardware potential.

rippleman said,
last post to you about this: your apps on your samsung phone are indeed getting 100%. What everyone is trying to say is that ONLY the benchmarks are allowed 120% of your power


Running the CPU at 2.3 Ghz under any benchmark apps is 100%. Running at 2.76 Ghz is running it at 120%.

Try harder next time.

Exynos said,

Other Samsung apps are allowed to run at full speed + some other apps that need such power as it was explained in the same topic about the Galaxy S4 is also allowed to use 100% of the power the hardware can deliver.

Why do other apps that doesn't need the 100% power output needs to be able to run at 100% and rather waste battery and don't become any more effective than just using 40-50% of the full potential of the hardware?

I'am allowed to run Winamp 5, Firefox or Google Drive at 100% load on my normal computer?

No, and why should i when the apps never will even need 20% of the full potential of the hardware in my computer?

Just because benchmark apps needs to use 100% out of the hardware it doesn't mean everything else have to do it.

So yes, everyone who makes benchmarks apps makes sure to use 100% out of the hardware potential.

I agree with everything you've said in this post and, if you're correct, there's nothing wrong with that.

However, some people and indeed this article seem to suggest that ONLY benchmark apps are able to use the processor to its full potential.

In other words, even apps that require the horsepower like games and video encoders get less performance than these benchmark apps, because Samsung has hard coded in this 'extreme performance' mode to only occur in benchmark apps. If that is correct, then what Samsung is doing is quite clearly wrong.

If you are correct and other apps are able to use the full 100% performance - if required - then they are not doing anything wrong.

I've just noticed that renaming a benchmark app causes a decrease in performance. In other words, when the EXACT SAME APP REQUIRING THE EXACT SAME LEVEL OF PERFORMANCE is renamed, the APP DOES NOT GET THE SAME LEVEL OF PERFORMANCE AS BENCHMARKS.

So Samsung are dirty rotten cheats and exynos is an apologist through and through.

KTXBAI

Hardcore Til I Die said,
I've just noticed that renaming a benchmark app causes a decrease in performance. In other words, when the EXACT SAME APP REQUIRING THE EXACT SAME LEVEL OF PERFORMANCE is renamed, the APP DOES NOT GET THE SAME LEVEL OF PERFORMANCE AS BENCHMARKS.

So Samsung are dirty rotten cheats and exynos is an apologist through and through.

KTXBAI


Like it have been said multiple times already, it's just a script to make sure the apps / benchmark apps that needs the full power output who use the original app name will be able to go around the speed throttling that is made to save battery.

Just because someone can rename the apps doesn't change the fact that Samsung and many others made the script simply to target the specific apps to go around the speed throttling.

What's so hard with this to understand, really?

Exynos said,

Like it have been said multiple times already, it's just a script to make sure the apps / benchmark apps that needs the full power output who use the original app name will be able to go around the speed throttling that is made to save battery.

Just because someone can rename the apps doesn't change the fact that Samsung and many others made the script simply to target the specific apps to go around the speed throttling.

What's so hard with this to understand, really?

Nothing hard to understand about that. It's still inflating the results though, because other apps can't do the same thing. The benchmarks are therefore measured using performance that other apps CANNOT OBTAIN.

Cheating. Straight up.

Hardcore Til I Die said,

Nothing hard to understand about that. It's still inflating the results though, because other apps can't do the same thing. The benchmarks are therefore measured using performance that other apps CANNOT OBTAIN.

Cheating. Straight up.


No it's not, the apps are still not allowed to use more power than the hardware lets you use as default.

If the CPU are developed to run at 2.3 GHz on each core or if the GPU are made to run at 450 MHz as default speed, then what the freaking hell is wrong by letting apps use that then?

I don't get it. Either you have a total lack of understanding how technology works or you are plain stupid. Sorry for saying that. This have gone way to far and i still don't know why you keep saying the benchmark apps aren't allowed to use the power the hardware are made to run at.

If i buy a Ferrari F355, it say i can take it to 295 km/h. That just let me knows how fast the car is. Does that mean i have to run it at that speed just because Ferrari it self benchmarked it to that speed?

Exynos said,

No it's not, the apps are still not allowed to use more power than the hardware lets you use as default.

If the CPU are developed to run at 2.3 GHz on each core or if the GPU are made to run at 450 MHz as default speed, then what the freaking hell is wrong by letting apps use that then?

I don't get it. Either you have a total lack of understanding how technology works or you are plain stupid. Sorry for saying that. This have gone way to far and i still don't know why you keep saying the benchmark apps aren't allowed to use the power the hardware are made to run at.

If i buy a Ferrari F355, it say i can take it to 295 km/h. That just let me knows how fast the car is. Does that mean i have to run it at that speed just because Ferrari it self benchmarked it to that speed?

Benchmark apps are allowed to use the full power for extended periods of time though. Regular apps can't. Therefore. They. Have. Inflated. Performance.

Hardcore Til I Die said,

Benchmark apps are allowed to use the full power for extended periods of time though. Regular apps can't. Therefore. They. Have. Inflated. Performance.


And when did benchmarking tools / apps get a regular app?

And i'm not sure, but have you realized yet that we are talking about benchmark apps witch are supposed to run at full power (100%) all the time and not regular apps who runs at 10-30% of the hardware potential here?

Because the benchmarking apps are supposed to run at full power (100%) all the time, then Samsung and the other brands have made a script that will set the phone in the 'Performance' governor under those tests so the tests actually can give us the results of a test that have used 100% of our hardware and not 40-50% of it.

Wake up maybe?

Exynos said,

And when did benchmarking tools / apps get a regular app?

And i'm not sure, but have you realized yet that we are talking about benchmark apps witch are supposed to run at full power (100%) all the time and not regular apps who runs at 10-30% of the hardware potential here?

Because the benchmarking apps are supposed to run at full power (100%) all the time, then Samsung and the other brands have made a script that will set the phone in the 'Performance' governor under those tests so the tests actually can give us the results of a test that have used 100% of our hardware and not 40-50% of it.

Wake up maybe?

I'm not talking about Facebook though am I... I'm saying that processor intensive tasks such as video encoding or graphic intensive games ALSO cannot use the processor in the way that the benchmarks do even though they would benefit from it!

There would be no point in running the CPU at 100% for Facebook, but there most certainly would be for video encoding.

You're on a one man crusade here and you're losing badly.

Hardcore Til I Die said,

I'm not talking about Facebook though am I... I'm saying that processor intensive tasks such as video encoding or graphic intensive games ALSO cannot use the processor in the way that the benchmarks do even though they would benefit from it!

There would be no point in running the CPU at 100% for Facebook, but there most certainly would be for video encoding.

You're on a one man crusade here and you're losing badly.


First of all, if the app actually needs the full power of the hardware, it will use it, because it will be coded to force it around any speed throttling systems then.

It's not Samsung's fault that the current apps today aren't made to actually utilize 100% of the current hardware that is very powerfull today in smartphones.

That's like saying omg, why doesn't the first Far Cry game utilize 100% of my hardware today on my computer when it did that on a computer i had many many years ago?

Think for a second here and realize that apps have to be optimized and fixed to actually be able to use all of the power the current smartphones have today.

So instead of circlejerking for full bananas, you can rather stop with your nonsense-crusading and make sure you don't lose badly your self.

Exynos said,

First of all, if the app actually needs the full power of the hardware, it will use it, because it will be coded to force it around any speed throttling systems then.

It's not Samsung's fault that the current apps today aren't made to actually utilize 100% of the current hardware that is very powerfull today in smartphones.

That's like saying omg, why doesn't the first Far Cry game utilize 100% of my hardware today on my computer when it did that on a computer i had many many years ago?

Think for a second here and realize that apps have to be optimized and fixed to actually be able to use all of the power the current smartphones have today.

So instead of circlejerking for full bananas, you can rather stop with your nonsense-crusading and make sure you don't lose badly your self.

THE EXACT SAME BENCHMARK APP just renamed doesn't get the same performance as the benchmark app with a name that Samsung recognises.

This shows that apps CANNOT achieve the same level of performance.

It's that simple.

You have no counter argument.

Hardcore Til I Die said,

THE EXACT SAME BENCHMARK APP just renamed doesn't get the same performance as the benchmark app with a name that Samsung recognises.

This shows that apps CANNOT achieve the same level of performance.

It's that simple.

You have no counter argument.


Because of the speed throttling. Therefor Samsung and the others have to go around the speed throttling by making a script that forces the CPU and GPU to run at full speed all the time while running those benchmark apps that are named in their script.

Samsung can't forbid someone from renaming an app so it doesn't get recognized by the script. That's not even the point. The point is that Samsung have added the benchmark apps names into the script to go around the speed throttling system. If someone rename that benchmarking app, then they makes so the benchmarking app wont be fully utilized by the hardware.

There is a reason why they have made a script for it. To let named apps go around things to let the apps works like they are made for doing.

One more stupid argument from you and i will report you for trolling. You are warned.

Exynos said,

Because of the speed throttling. Therefor Samsung and the others have to go around the speed throttling by making a script that forces the CPU and GPU to run at full speed all the time while running those benchmark apps that are named in their script.

Samsung can't forbid someone from renaming an app so it doesn't get recognized by the script. That's not even the point. The point is that Samsung have added the benchmark apps names into the script to go around the speed throttling system. If someone rename that benchmarking app, then they makes so the benchmarking app wont be fully utilized by the hardware.

There is a reason why they have made a script for it. To let named apps go around things to let the apps works like they are made for doing.

One more stupid argument from you and i will report you for trolling. You are warned.

You said that other apps can utilise the full capacity of the hardware, yet you're talking about speed throttling and how Samsung has to hard code benchmark apps into the hardware to get around the speed throttling.

Can't you see that if Samsung throttles the speed of non-benchmark apps, they can't attain the same performance as benchmark apps?

Your position doesn't make any sense.

Hardcore Til I Die said,

You said that other apps can utilise the full capacity of the hardware, yet you're talking about speed throttling and how Samsung has to hard code benchmark apps into the hardware to get around the speed throttling.

Can't you see that if Samsung throttles the speed of non-benchmark apps, they can't attain the same performance as benchmark apps?

Your position doesn't make any sense.


Yep, if they are DEVELOPED to do it. But extremely few apps are developed to actually use the full power on todays smartphones because they are extremely powerfull.

And if an app that was made in 2010 was made to run at full speed on phones that times will not automaticly just run at full speed on todays smartphones, because the CPU's and the CPU architecture and the way they run are totally different today than it was in 2010. And not to mention that Android today are changed and optimized alot from the Android versions that was in 2010.

If you can't see or understand the technical stuffs behind this, then i'm afraid you're waste of text to argue with.

Exynos said,

Yep, if they are DEVELOPED to do it. But extremely few apps are developed to actually use the full power on todays smartphones because they are extremely powerfull.

And if an app that was made in 2010 was made to run at full speed on phones that times will not automaticly just run at full speed on todays smartphones, because the CPU's and the CPU architecture and the way they run are totally different today than it was in 2010. And not to mention that Android today are changed and optimized alot from the Android versions that was in 2010.

If you can't see or understand the technical stuffs behind this, then i'm afraid you're waste of text to argue with.

Are you suggesting that benchmarks aren't written to take full advantage of the hardware?

Hardcore Til I Die said,

Are you suggesting that benchmarks aren't written to take full advantage of the hardware?


Are you blind?

Didn't i say very few apps are using the full power of the current smartphones today?

And guess what?

The few benchmark apps that are out now are those few apps, because they have been optimized, fixed and utilized to run on todays smartphones.

Exynos said,

Are you blind?

Didn't i say very few apps are using the full power of the current smartphones today?

And guess what?

The few benchmark apps that are out now are those few apps, because they have been optimized, fixed and utilized to run on todays smartphones.

No not blind. Just trying to make sense of your tangled web of crap.

So what you're saying is, normal apps can take full advantage of the processor if they are coded to do so. Benchmarks are coded to take full advantage of the processor. Yet as soon as you rename a benchmark app (an app which is STILL coded to take full advantage of the processor), it doesn't get anywhere near the performance.

Riiiiiight.

We're done here.

To be honest I don't really zero in on the product which scores the highest benchmark figure when I'm in the market for a new device. I just go for the one that suits me the best.

I don't really see what they have gained by 'rigging' the benchmark as you guys say.

I guess people can also expect the battery to last a 2 weeks as well... (under no use and everything turned off) Remember, its what it CAN do.... /s

Nik Louch said,
Yawn, this again? Argued it hard last time against zealots. It's what the hardware CAN DO.

Most people who buy something and knows a thing or two, the benchmarks not even the 2nd or third thing they consider first. Others dont even know what benchmarks are and are oblivious to them. So in the grand scheme of things, this doesnt really matter a whole lot. Only really to the tech crowd who needs a new hot topic of the day.

Samsung should of followed the same guidelines as the others to give a consistent reading. Just skewed at this point. Whether it was right/wrong and if Samsung was cheating (which is really debatable at this point)....who cares. Lets see LGs benchmark results using the same "cheat" Samsung did. Would shut this argument down either way if that happened.

techbeck said,
Most people who buy something and knows a thing or two, the benchmarks not even the 2nd or third thing they consider first. Others dont even know what benchmarks are and are oblivious to them. So in the grand scheme of things, this doesnt really matter a whole lot. Only really to the tech crowd who needs a new hot topic of the day.

Samsung should of followed the same guidelines as the others to give a consistent reading. Just skewed at this point. Whether it was right/wrong and if Samsung was cheating (which is really debatable at this point)....who cares. Lets see LGs benchmark results using the same "cheat" Samsung did. Would shut this argument down either way if that happened.

Specific benchmark scores are irrelevant, but the word-of-mouth about which is the "best" phone when considering buying a new one isn't.

In this case though the Note3 still achieves a pretty good score without the "optimizations" and comes ahead of the LG.

How is it cheating? The CPU IS RUNNING AT THOSE SPEEDS AND ACHIEVING THE SCORES REPORTED.

Isn't that exactly what a benchmarking application is supposed to do? Push the limits of the hardware to see what its capable of?

Lord Method Man said,
How is it cheating? The CPU IS RUNNING AT THOSE SPEEDS AND ACHIEVING THE SCORES REPORTED.

Isn't that exactly what a benchmarking application is supposed to do? Push the limits of the hardware to see what its capable of?

It's not cheating, but it's misleading because customers will never get that performance out of the device.

I mean, who cares if the note3 could score 100 in a benchmark when it'll always effectively behave as if it had scored 80.

Just imagine how high the benchmarks would have been if it was running Windows 8. They had to ban Windows 8 from benchmarking tools because it is that good.

Agree with most of the comments here, yeah it's misleading, but not exactly illegal/wrong, the CPU is legitimately able to perform the tasks at that speed. It WOULD be a scandal if they falsified readings but they didn't.

Another way of looking at the car analogy, 0-100 in 7secs and a fuel economy of 12 lp/100k is advertised. Sure it can go 0-100 but it will be averaging 30-40 lp/100k at that speed. Same with the chip, if you use it at max power economy will drop. The device can still perform to expectations.

All in all though, who cares about benchmarks? Power is so acceptable these days pretty much any flagship phone will outperform expectations, even most mid range phones will chew through most applications. My old S2 performs fine in everything I throw at it and it would be considered low mid range today. My new Oppo Find 5 doesn't even blink. I've never considered my phone choice through benchmarks. I see them as a bit of a dick waving contest tbh, especially when the "stock" phone will perform significantly less than the same phone running a custom ROM and kernel. Too many variables.

So by allowing the hardware to run at full speed and removing the idling during the test, they're cheating? Aren't benchmarks supposed to show what hardware is capable of when run at maximum?

This is the second time Samsung has used this tactic, they did it with the S4. Glad to see the blogs are picking up on this.

This used to happen quite frequently between Nvidia, ATI & even Intel, especially with Futuremark products.

I love how everyone here speaking negatively is ignoring the fact that the note 3 outperformed the G2 with AND without the boosting.

banggugyangu said,
I love how everyone here speaking negatively is ignoring the fact that the note 3 outperformed the G2 with AND without the boosting.

Benchmarks results not being representative of the performance users will get don't mean that the device isn't actually quite good. It just means that they don't represent what most people would have expected them to represent.

has this ever happened and NOT come into light within days of the device's release? Answoer: No.... This practice, while a marketing fallacy, is going to happen. It is also going to be circumvented by people just like the ones that did this original article. It's been happening for decades in every other market, why people are shocked by it happening in the cell phone market now eludes me.

The fact of the matter is, this is still the superior device. People will still buy it. This practice will most likely STILL happen in the future (and by other manufacturers).

banggugyangu said,
has this ever happened and NOT come into light within days of the device's release? Answoer: No.... This practice, while a marketing fallacy, is going to happen. It is also going to be circumvented by people just like the ones that did this original article. It's been happening for decades in every other market, why people are shocked by it happening in the cell phone market now eludes me.

The fact of the matter is, this is still the superior device. People will still buy it. This practice will most likely STILL happen in the future (and by other manufacturers).


Samsung isn't alone: http://www.slashgear.com/galax...igans-accusations-01299884/

Guess why, because everyone does it as you say and like they have done for decades and because that's how it's supposed to be when running benchmark tools :-).

banggugyangu said,
I love how everyone here speaking negatively is ignoring the fact that the note 3 outperformed the G2 with AND without the boosting.
In one very short sentence you simply summed up the whole truth. Even without the boost it still won. The only difference is it won my a nose, and not by football field. lol

Here is what is even funnier. All Samsung did was make sure all 4 cores ran at 2.3Ghz for each test. If you look at the graphs the CPU were running at lest than 20% even at 2.3Ghz. Which means if they really wanted to cheat, they could have not only made sure the cores ran at 2.3Ghz each, but all a full 100% too. That way the phone would have more than double the other score.

This isn't cheating, this is winning with style!

exotoxic said,
I though the idea of benchmarks was to measure what the hardware is capable of??

It's supposed to be but when money is to be made, most consumer companies resort to tactics in the end.

Yes it is, but what Samsung did is force the hardware to behave the way it wouldn't behave otherwise. You'll never see the 20% boost in real life usage, Samsung is hard-coding the cheat to only overclock the CPU during benchmark usage.

Aergan said,

It's supposed to be but when money is to be made, most consumer companies resort to tactics in the end.

Like what? Making sure their hardware looks better? In this case it is better.

Mean while

Apple executive Phil Schiller -- senior vice president of Worldwide Marketing and the most prolific tweeter amongst Apple's senior staff -- linked to the Ars article in a tweet, saying only "shenanigans".

StandingInAlley said,
Mean while

Apple executive Phil Schiller -- senior vice president of Worldwide Marketing and the most prolific tweeter amongst Apple's senior staff -- linked to the Ars article in a tweet, saying only "shenanigans".

Who cares. The Galaxy Note III can record full 4K video at 30FPS and play it back at the same rate without a 64bit processor and the iPhone's hardware which likely can as well, isn't allowed too amongst other things its not allowed to do.

Strangely enough, Microsoft using Javascript cheats in IE to do the same thing never made FPN here. Nor did the BSOD's caused by IE10, or Gates' admission that Microsoft failed by entering the smartphone business too late.

Neowin: Where skewed technology news doesn't sound better. Especially when it's this false. Reality is, if it can do that in a benchmark, the hardware is capable of it, they just dial the speed back when it isn't needed for the purpose of battery life. If it ran at full tilt all the time and ate through it's battery you can bet it would gleefully be lapped up by Neowin's editors as a Samsung failing as well.

Actually, it was here on NeoWin as well as many other places, and was debunked that Microsoft had hacked their JavaScript engine to favor benchmarks since it was performing those same optimizations for all code. Just because Microsoft is faster in some areas than their rivals and you do not like that does not mean they are cheating.

If you read my earlier post here, then why is it pathetic?

Maybe you should stop being pathetic instead and find out what benchmarks tools are supposed to do ?

Exynos said,
If you read my earlier post here, then why is it pathetic?

Maybe you should stop being pathetic instead and find out what benchmarks tools are supposed to do ?

Was that really necessary? Resulting to personal attacks? That is what I'd call pathetic.

Exynos said,
If you read my earlier post here, then why is it pathetic?

Maybe you should stop being pathetic instead and find out what benchmarks tools are supposed to do ?

No they haven't.

So by Ars Technica's definition, all car manufacturers are cheating when they list their 0-100 km/h acceleration time on their cars when they have to give the cars full power to be able to do that?

In their eyes they are cheating because they aren't doing a 0-100 km/h acceleration time in a neighborhood where lots of peoples lives / where normal conditions are.

When you are testing your GPU under 3DMark, CPU under IntelBurnTest, RAM under AIDA and HDD under CrystalDiskMark and so on, they don't give you the results under normal usage where you opens Firefox, Winamp, Dropbox, DAEMON Tools and use Windows Explorer. They will give you the results where the hardware have been pushed to it's limits.

And everybody knows that those programs as i meantioned over + many many other apps doesn't even use 15% of the full potential of the hardware in our computers / smartphones today.

Everyone who does benchmarks and have done it all the years bechmarks tools have been out makes sure to utilize the hardware on a computer 100%.

Ars Technica's article is totally useless and more of a butthurt response to Samsung because they makes sure to let the benchmark tools to use 100% of the power the phone has to deliver while the other phone manufacturers doesn't do it.

It's not Samsung's fault that manyof the other phone manufacturers doesn't do like everyone else does.

EDIT: http://www.slashgear.com/galax...igans-accusations-01299884/

Exynos said,
So by Ars Technica's definition, all car manufacturers are cheating when they list their 0-100 km/h acceleration time on their cars when they have to give the cars full power to be able to do that?

By Ars Technica's definition, a Car manufacturer would be cheating their acceleration/fuel mileage numbers by removing everything but the driver's seat in the car to make it lighter and thus appear to perform better, however when consumers buy the cars they'd find out that *oh surprise* they don't get the same numbers, because, honestly, who drives a car only with the driver's seat inside? (who uses a Smartphone with all its cores always turned on at the maximum frequency all the time? No one of course, you'd deplete your battery in a couple of hours)

A misrepresentation of performance IS an intent to deceit, at least under the Laws of my country.

If Samsung had a disclaimer saying "we disabled the part of the chip to slow down for power optimization when running the benchmarks" I don't think that anyone would complain. But to embed in the firmware/OS a check on the file name or other unique ID to alter how a benchmark program is run and only for those benchmarks programs and not inform the user, that is the problem.

And car manufacturers do list multiple power outputs in their stats sheets. They will often list the time to 0-100km/h and 0-60mph. They will list the amount of horsepower at the wheel as well as at the axle (in imperial as well as metric) because they are very different.

Those stats for the GPU are usually on desktop computers where power is not an issue and it makes no difference anyway because GPUs do not have speedstep like technology they run at full all the time.

Why would Ars be butthurt? If another phone maker were complaining then I could see it but Ars does not make phones, they are a tech site.

This is why the car companies are having their MPG sticker ratings re-evaluated, because the rules of the game do not match customer expectations, and the bottom line here is that when you see performance measurements, the CUSTOMER expects to see those results.

If a company wants to go "balls out" in their metrics, then they need to clearly explain to consumers that "you will never see this level of performance in your apps, this is just the best that this processor can do". But they don't. So they're wrong.

It is in the same way that MPG is being changed to reflect real road usage, not "best case"... people are upset in general that when a car says it does X/highway and Y/street, they expect to see those numbers. You can't sell a best case scenario... you sell reality.

Wrong, this is along the lines of GTR.

You have a car that is super fast and gets great lap times, but is limited when not at a race track.

So you buy the car thinking its faster than it is on the street (normal apps), and it only unlocks the speed on track (benchmark apps).

There is a legal rule of evidence when comparing the performance of GTR vs normal Cars.
People suing a manufacturer because their car isn't as fast as the GTR version would be thrown out of court, because the usage intent and therefore performance of each car is evident.

Hiding a setting which makes a device go full on during benchmarks while representing the results as being its normal performance is a misrepresentation and therefore is an intent of deceit.

WhatTheSchmidt said,
If Samsung had a disclaimer saying "we disabled the part of the chip to slow down for power optimization when running the benchmarks" I don't think that anyone would complain. But to embed in the firmware/OS a check on the file name or other unique ID to alter how a benchmark program is run and only for those benchmarks programs and not inform the user, that is the problem.

No, it's not the problem, because any benchmark apps are supposed to run at full speed / power independent on what method Samsung is using to let some benchmark apps run.

That's the point. The whole point of a benchmark application is to test the full potential of the hardwares.

Exynos said,

No, it's not the problem, because any benchmark apps are supposed to run at full speed / power independent on what method Samsung is using to let some benchmark apps run.

That's the point. The whole point of a benchmark application is to test the full potential of the hardwares.

Wrong, it is to test the default setup under which all apps must follow, if any special tweak is made, then it is not testing the default device, but a modified one. Special tweaks are a cheat, no matter how you put it.

It's the same as running games with McAfee running, only to turn it off for a benchmark. Sure your score will be higher, but if you leave McAfee on when you game, that score doesn't represent actual performance.

Exynos said,

No, it's not the problem, because any benchmark apps are supposed to run at full speed / power independent on what method Samsung is using to let some benchmark apps run.

That's the point. The whole point of a benchmark application is to test the full potential of the hardwares.

No, benchmarks are supposed to let the user what they can experience under normal circumstances. If it were to test the full potential of the hardware then we would have people throwing liquid nitrogen coolers on their chips, overclocking their CPU, memory, and GPU, and then we would be wondering why my latest and greatest Haswell chip is running much slower thanthe identical Haswell of some other user. There is nothing wrong with benchmarking in the non-user scenarios but they need to inform the user that is happening.

SierraSonic said,
Wrong, it is to test the default setup under which all apps must follow, if any special tweak is made, then it is not testing the default device, but a modified one. Special tweaks are a cheat, no matter how you put it.

It's the same as running games with McAfee running, only to turn it off for a benchmark. Sure your score will be higher, but if you leave McAfee on when you game, that score doesn't represent actual performance.

WhatTheSchmidt said,

No, benchmarks are supposed to let the user what they can experience under normal circumstances. If it were to test the full potential of the hardware then we would have people throwing liquid nitrogen coolers on their chips, overclocking their CPU, memory, and GPU, and then we would be wondering why my latest and greatest Haswell chip is running much slower thanthe identical Haswell of some other user. There is nothing wrong with benchmarking in the non-user scenarios but they need to inform the user that is happening.


So 3DMark / PCMark is not to there to test the full potential of the GPU / computer and just to let you know how good the GPU / hardware are under descent tasks?

Simple, you don't know what a benchmarking tool is. PERIOD. A benchmark tool is to determine how powerfull the hardwares is at it's full power output, not to see how they perform under normal usage. Everyone knows this.

The only ones that doesn't want to admit this is the Samsung haters that knows Samsung basicly whoops the other phone manufacturers ass in performance.

If you want to determine how the phones perform in normal usage, then you can watch YouTube movies like Samsung Galaxy S4 / Note 3 vs LG G2 usage tests and others like that.

Edited by Exynos, Oct 2 2013, 12:27am :

When I am using a device, computer, car, etc. I don't care about the theoretical performance values I want to know how it will perform when I use it. If the device never runs in full speed mode then what good is that extra perf to me?

And I am not a Samsung hater. I have an ATIV.

WhatTheSchmidt said,
When I am using a device, computer, car, etc. I don't care about the theoretical performance values I want to know how it will perform when I use it. If the device never runs in full speed mode then what good is that extra perf to me?

And I am not a Samsung hater. I have an ATIV.


If you want to know how it's in normal usage, then you simply don't use benchmark tools.

Do you know the difference from normal usage and doing benchmarking?

Exynos said,

If you want to know how it's in normal usage, then you simply don't use benchmark tools.

Do you know the difference from normal usage and doing benchmarking?

The size of the users ePeen? I care about what the car does in real world usage, you know what I hate, buying a car that gets 50mpg but only really gets 34mpg.

What do you care about more, the fact that your car can possibly maybe get 50mpg if every perfect variable was matched, or if you realistically get the 37 on average.

Do you care about unrealistic max performance that is never really attainable or do you care about how the device actually runs?

SierraSonic said,
The size of the users ePeen? I care about what the car does in real world usage, you know what I hate, buying a car that gets 50mpg but only really gets 34mpg.

What do you care about more, the fact that your car can possibly maybe get 50mpg if every perfect variable was matched, or if you realistically get the 37 on average.

Do you care about unrealistic max performance that is never really attainable or do you care about how the device actually runs?


Just because you don't see the differences from normal usage to benchmarking, it doesn't mean everyone else does it.

I care about what my computer / smartphone can deliver in performace at 100% power output.

Exynos said,

Just because you don't see the differences from normal usage to benchmarking, it doesn't mean everyone else does it.

I care about what my computer / smartphone can deliver in performace at 100% power output.

So you care about what it cannot do?

Again you care about the 500hp the company states the engine has from the factory, not the 380hp the car actually makes at the wheels.

SierraSonic said,
So you care about what it cannot do?

Again you care about the 500hp the company states the engine has from the factory, not the 380hp the car actually makes at the wheels.


I care what my computer is capable of doing like the benchmarks apps let me do.

Why shouldn't i care about how powerfull my computer really are?

If Intel say that i can get 3.3 GHz on my CPU and i only get 3 GHz, then i will request a new CPU on my warranty because that CPU is broken then.

If Corsair say that i can get my CPU temprature on my CPU down to 50 degrees celcius with the Corsair H60 like i have now and i only get it down to 56 degrees celcius, i wouldn't care as their numbers would just be an indication of an optimal computer system. Because there is alot of other things in our computer that can change the temprature on different things.

However, if someone gets 10000 3DMark score with almost the indentical computer as i have, also that he are using the same CPU, GPU, Motherboard, PSU, RAM and CPU cooler as me and then i only gets 7000 points in 3DMark, then i know there are something wrong with my computer somewhere.

Edited by Exynos, Oct 2 2013, 2:05pm :

Exynos said,

I care what my computer is capable of doing like the benchmarks apps let me do.

Why shouldn't i care about how powerfull my computer really are?

You dont care about how fast your computer really is, you only care about the theoretical speed... I can guarantee that you never run your setup with no apps open all services shut off gui disabled, etc all the time.
What you care about is the max specs that are never used, not the average specs which are what you use. Again, the 500hp car is actually the same 380RHP car, it's just that the 380 is the more accurate number reflecting real world performance instead of theoretical.

Also, if your benchmark score is different, it is most likely because they are using tweaks to boost their score. You would be suprised what difference simple graphic driver settings make, such as changing high quality options to high performance options, which of course is great for benchmarks, but usually somehow magically changes back to high quality if you game.

Edited by SierraSonic, Oct 2 2013, 2:33pm :

SierraSonic said,
You dont care about how fast your computer really is, you only care about the theoretical speed... I can guarantee that you never run your setup with no apps open all services shut off gui disabled, etc all the time.
What you care about is the max specs that are never used, not the average specs which are what you use. Again, the 500hp car is actually the same 380RHP car, it's just that the 380 is the more accurate number reflecting real world performance instead of theoretical.

No, i care about what my computer are good for at the full potential. Just because you only cares about your normal usage performance, it doesn't mean everyone else does the same.

Also, i have been overclocking my Intel Core i5-2500k CPU to 4.8 GHz stable with my Corsair H60 Liquid Cooler and taken my GPU speed from 900 MHz core speed and 4200 MHz memory speed to 980 MHz core speed and 4600 MHz memory speed. And ofc the DDR3 RAM modules from 1866 MHz to 2100 MHz.

So yes, i do care about how my computer performs at the full potential, even in an overclocked state.

Exynos said,

No, i care about what my computer are good for at the full potential. Just because you only cares about your normal usage performance, it doesn't mean everyone else does the same.

Also, i have been overclocking my Intel Core i5-2500k CPU to 4.8 GHz stable with my Corsair H60 Liquid Cooler and taken my GPU speed from 900 MHz core speed and 4200 MHz memory speed to 980 MHz core speed and 4600 MHz memory speed. And ofc the DDR3 RAM modules from 1866 MHz to 2100 MHz.

So yes, i do care about how my computer performs at the full potential, even in an overclocked state.

Overclocks are different than this though, this is literally disabling things normally running, forcing a full speed state that no other apps have access to and tweaks that change how it runs during the benchmarks vs how it actually runs... do you not get that this is the difference between how you run your computer with its overclocks, and then additional tweaks that even your setup doesn't have to make it seem EVEN faster?

SierraSonic said,
Overclocks are different than this though, this is literally disabling things normally running, forcing a full speed state that no other apps have access to and tweaks that change how it runs during the benchmarks vs how it actually runs... do you not get that this is the difference between how you run your computer with its overclocks, and then additional tweaks that even your setup doesn't have to make it seem EVEN faster?

Still doesn't matter, because 2.3 GHz as the Galaxy Note 3 runs at is still 2.3 GHz. And running a benchmark app will still be within that speed.

What's the problem?

Arkos Reed said,

By Ars Technica's definition, a Car manufacturer would be cheating their acceleration/fuel mileage numbers by removing everything but the driver's seat in the car to make it lighter and thus appear to perform better, however when consumers buy the cars they'd find out that *oh surprise* they don't get the same numbers, because, honestly, who drives a car only with the driver's seat inside? (who uses a Smartphone with all its cores always turned on at the maximum frequency all the time? No one of course, you'd deplete your battery in a couple of hours)

A misrepresentation of performance IS an intent to deceit, at least under the Laws of my country.

In this case it still doesn't matter. The tools are designed to push the CPU/GPU to 100% of its power capability. The user isnt going to push it that far in normal usage. Even though the CPU is at 2.3Ghz, the average app will never run it at that speed anyways. The point is, benchmarks are false in that they tell you what your device if capanble of even if you never use it which most won't.

Samsung allowing for the apps to push the hardware even above 100% shows the hardware is even better than it is allowed to be normally.

Which means once XDA cracks the bootloader and I can install a ROM which will allow me to OC the chipset, that means I can run my phone 20% faster than normal without hurting anything.

Who cars about the battery. You can charge the device back up...right?

Exynos said,

Still doesn't matter, because 2.3 GHz as the Galaxy Note 3 runs at is still 2.3 GHz. And running a benchmark app will still be within that speed.

What's the problem?

You are correct, HARDWARE WISE, software wise you are wrong. Tell me, do you think it is okay to create a special testing field with special requirements AND THEN state that those special results are what happens in the regular environment? No, because those result will never happen without those special requirements.

What Samsung, and others, have been found doing is creating a LIE situation, that creates a higher score using special situation that is impossible to achive in the default firmware, no matter what tweaks you make. Now if you flash your device to ALWAYS use that state, using a custom rom, then it is okay, as long as your benchmark reflects "S4 with OCmod Rom", that way when a person using the default S4 rom won't be like "OMG his S4 is faster than mine, even though they are the same!!!!.!>.!!!!"

HipHopSinceFriday said,

In this case it still doesn't matter. The tools are designed to push the CPU/GPU to 100% of its power capability. The user isnt going to push it that far in normal usage. Even though the CPU is at 2.3Ghz, the average app will never run it at that speed anyways. The point is, benchmarks are false in that they tell you what your device if capanble of even if you never use it which most won't.

Samsung allowing for the apps to push the hardware even above 100% shows the hardware is even better than it is allowed to be normally.

Which means once XDA cracks the bootloader and I can install a ROM which will allow me to OC the chipset, that means I can run my phone 20% faster than normal without hurting anything.

Who cars about the battery. You can charge the device back up...right?

Exactly, except you won't say your S4 is 20% faster, you will say your modded S4 is 20% faster, big difference.

Arkos Reed said,

By Ars Technica's definition, a Car manufacturer would be cheating their acceleration/fuel mileage numbers by removing everything but the driver's seat in the car to make it lighter and thus appear to perform better, however when consumers buy the cars they'd find out that *oh surprise* they don't get the same numbers, because, honestly, who drives a car only with the driver's seat inside? (who uses a Smartphone with all its cores always turned on at the maximum frequency all the time? No one of course, you'd deplete your battery in a couple of hours)

A misrepresentation of performance IS an intent to deceit, at least under the Laws of my country.

There is no misrepresentation. Benchmarks are a misrepresentation because the number are based on hardware capability, not how the device is actually going to be used.

You obviously don't understand what benchmarking tools do.
Misrepresentation is illegal in pretty much every country. That is why this isn;t such.

HipHopSinceFriday said,
There is no misrepresentation. Benchmarks are a misrepresentation because the number are based on hardware capability, not how the device is actually going to be used.

You obviously don't understand what benchmarking tools do.
Misrepresentation is illegal in pretty much every country. That is why this isn;t such.

I'm just going to state that real world results are not misrepresentation while theoretical 'max' results are if they have to stoop to hidden settings and silent activation of cores are.

There is a reason why the EPA keeps changing the way MPG is calculated, to more represent real world mileage, not the lies that companies got by specifically designing their cars for benchmarks. Just like there's a reason "Stealth Benchmarks" exist. To get the REAL WORLD numbers not some fantasy numbers.

Edited by SierraSonic, Oct 2 2013, 4:13pm :

SierraSonic said,
I'm just going to state that real world results are not misrepresentation while theoretical 'max' results are if they have to stoop to hidden settings and silent activation of cores.

There is a reason why the EPA keeps changing the way MPG is calculated, to more represent real world mileage, not the lies that companies got by specifically designing their cars for benchmarks.


Ofc real world usage is not misrepresentation, but that's not the point. The point is that every benchmark apps that are out today are testing the full potential and not the real world usage of every hardwares in any computers.

Why else do you think Futuremark keeps developing the 3DMark benchmark apps?

It's to set new boundaries to squeze everything out of you computer everytime there is coming new GPU's and CPU's.

If i just would like to see how my computer perform under real world usage, i wouldn't install the newest 3DMark then, i would run 3DMark 2006 then. But since the newest 3DMark benchmark app is there to squeze out everything of you computers today, then it's there to do that, not to give you the results on real usage.

If i want to see how fast my CPU can render an MKV movie, then i will see it under full load and not under real usage load lol.

This whole article is WAY WAY out of control. First we have tons of crybabies who doesn't like that Samsung optimizes it's ROM's better than other brands and can get better benchmark scores out of it. Secondly, we have a bunch of noobs here who doesn't even know what the main purpose of a benchmark app is.

Talk about maximum failure.

Exynos said,

Ofc real world usage is not misrepresentation, but that's not the point. The point is that every benchmark apps that are out today are testing the full potential and not the real world usage of every hardwares in any computers.

Why else do you think Futuremark keeps developing the 3DMark benchmark apps?

It's to set new boundaries to squeze everything out of you computer everytime there is coming new GPU's and CPU's.

If i just would like to see how my computer perform under real world usage, i wouldn't install the newest 3DMark then, i would run 3DMark 2006 then. But since the newest 3DMark benchmark app is there to squeze out everything of you computers today, then it's there to do that, not to give you the results on real usage.

If i want to see how fast my CPU can render an MKV movie, then i will see it under full load and not under real usage load lol.

This whole article is WAY WAY out of control. First we have tons of crybabies who doesn't like that Samsung optimizes it's ROM's better than other brands and can get better benchmark scores out of it. Secondly, we have a bunch of noobs here who doesn't even know what the main purpose of a benchmark app is.

Talk about maximum failure.

You do realize that all those bench-marking software test the setup of the computer as it is and do not make tweaks to your power settings or video card settings or anything in anyway... right?

What Samsung is doing isn't optimizing their rom to be the fastest it could it, they are literally detecting if bench-marking apps are open and then they change the way the rom operates JUST FOR THE BENCHMARK.

When ATi and Nvidia detected bench-marking applications and changed setting they were caught and changed their ways. If it wasn't ok back then, why would it be now?

Also, grow up, and quit calling people names (noobs, crybabies).

Edited by SierraSonic, Oct 2 2013, 4:54pm :

SierraSonic said,
I'm just going to state that real world results are not misrepresentation while theoretical 'max' results are if they have to stoop to hidden settings and silent activation of cores.

There is a reason why the EPA keeps changing the way MPG is calculated, to more represent real world mileage, not the lies that companies got by specifically designing their cars for benchmarks. Just like there's a reason "Stealth Benchmarks" exist. To get the REAL WORLD numbers not some fantasy numbers.

Huh? Dude, if you install and run the proper application, those so-called silent cores are going to run.

Let me give you an example. Lets look at Infinity Blade III for iOS. The iPhone only has 2 cores. So in reality, if the dev ported that app over to Android without making any significant changes, it would still only use 2 cores. Hwoever, since Android phones do have 4 cores, then the dev could simply add a java script that tells the app that if the game is loaded in a phone with 4 cores, to use all 4 cores where needed and if the phone only has to cores just use the preset 2 cores.

Its that simple. Samsung made a java routine that simply made sure in the benchmark all 4 cores were used. Under a typical benchmark, maybe the the routine wouldnt have required all 4 cores. So what.

Have you even driven new cars? There are cars out there that have 6 cylinders. When the car is doing normal driving, only 4 cylinders are being used. When the car needs more power like when you press on the accelerator harder to make a pass, the other 2 cylinders are engaged to gave more power. Once you don't need it, the 2 turn back off. Is that cheating?

mram said,
This is why the car companies are having their MPG sticker ratings re-evaluated, because the rules of the game do not match customer expectations, and the bottom line here is that when you see performance measurements, the CUSTOMER expects to see those results.

If a company wants to go "balls out" in their metrics, then they need to clearly explain to consumers that "you will never see this level of performance in your apps, this is just the best that this processor can do". But they don't. So they're wrong.

It is in the same way that MPG is being changed to reflect real road usage, not "best case"... people are upset in general that when a car says it does X/highway and Y/street, they expect to see those numbers. You can't sell a best case scenario... you sell reality.

False

When car makers do an MPG test, the final score is based on an average score based on doing several driving conditions. If they say in open highway driving at 55mph, that I should expect a gas mileage of 40 MPG, then I should get anywhere from 38 to 42MPG...40 is simply the average...it is not the exact.

Again this is based on how they tested, not how you drive. As a fact the only way to get such mileage is to use the cruise control because your car will mostly stay at 55 unless there are lots of hills. When you drive with your foot, it gets tired and you fluctuate the speed more often up an down. Thus your mileage may vary.

This is nowhere close to the benchmarking tests. It is not the same. In this phone the CPU speed is 2.3Ghz per core PERIOD. In the tests, the Samsung device called up a java file that simply ran the cores at 2.3Ghz for all the tests. If you look at the tests, even though all 4 cores were at 2.3Ghz, they were not at a full 100%. Which means the phone for example in a video test could play the video at 30fps with the cores running at 2.3Ghz but they were at only 3% usage. Which means in real world usage the phone would play the same video with the CPU at a much lesser clock and still perform as expected. PERIOD. They didn't cheat, they simply past the test with flying colors instead of just simply passing the test.

SierraSonic said,
You do realize that all those bench-marking software test the setup of the computer as it is and do not make tweaks to your power settings or video card settings or anything in anyway... right?

What Samsung is doing isn't optimizing their rom to be the fastest it could it, they are literally detecting if bench-marking apps are open and then they change the way the rom operates JUST FOR THE BENCHMARK.

When ATi and Nvidia detected bench-marking applications and changed setting they were caught and changed their ways. If it wasn't ok back then, why would it be now?

Also, grow up, and quit calling people names (noobs, crybabies).


You do realize that apps like Winamp 5, Mozilla Firefox, Google Chrome, Google Drive, Dropbox and so on will under no situations use more than 30-40% max out of the full potential of the hardware in any modern computers today?

Oh what happens when i run any benchmark apps on my computer, yes the load on everything shoots to 100%. Because that's the main purpose of the benchmark apps.

OMGWTF, those benchmark apps use H4XES AND CHEATS OMGOMG, because the benchmark apps should not run at 100% because some other benchmark apps doesn't. WTF.................

Yeah i hope you get my point here. As it is now, you just gets more and more ridiculous and funny for each time you reply here.

And no, i wont stop using the word crybabies and noobs here when you and some others here still continues to go in a massive circlejerking about something you clearly have no clues about. So maybe you should grow up and get teached about what benchmarking is before you speak nonsense?

STOP IT.

Edited by Exynos, Oct 2 2013, 4:17pm :

HipHopSinceFriday said,
Huh? Dude, if you install and run the proper application, those so-called silent cores are going to run.

Let me give you an example. Lets look at Infinity Blade III for iOS. The iPhone only has 2 cores. So in reality, if the dev ported that app over to Android without making any significant changes, it would still only use 2 cores. Hwoever, since Android phones do have 4 cores, then the dev could simply add a java script that tells the app that if the game is loaded in a phone with 4 cores, to use all 4 cores where needed and if the phone only has to cores just use the preset 2 cores.

Its that simple. Samsung made a java routine that simply made sure in the benchmark all 4 cores were used. Under a typical benchmark, maybe the the routine wouldnt have required all 4 cores. So what.

Have you even driven new cars? There are cars out there that have 6 cylinders. When the car is doing normal driving, only 4 cylinders are being used. When the car needs more power like when you press on the accelerator harder to make a pass, the other 2 cylinders are engaged to gave more power. Once you don't need it, the 2 turn back off. Is that cheating?

First off, porting doesn't work like that, they always need to make optimizations to ports to work better.

Second, The benchmark apps do take the hardware to the max, in their current state. The thing that Samsung is doing is changing the way the device runs once it detects these apps.

Third, the cars will always run with the cylinder deactivation activated, during testing and actual driving, it's not like these companies give special instructions to the car to run better with the cylinders only disabled while they are being tested. That is the difference. The car will run the same during these tests and on the road. So no, it's not cheating.

What Samsung is doing is detecting if someone is doing MPG tests, and completely disables the cylinders, changes the tune to use less fuel, and have less emissions than it really would, which is cheating.

Exynos said,

You do realize that apps like Winamp 5, Mozilla Firefox, Google Chrome, Google Drive, Dropbox and so on will under no situations use more than 30-40% max out of the full potential of the hardware in any modern computers today?

Oh what happens when i run any benchmark apps on my computer, yes the load on everything shoots to 100%. Because that's the main purpose of the benchmark apps.

OMGWTF, those benchmark apps use H4XES AND CHEATS OMGOMG, because the benchmark apps should not run at 100% because some other benchmark apps doesn't. WTF.................

Yeah i hope you get my point here. As it is now, you just gets more and more ridiculous and funny for each time you reply here.

And no, i wont stop using the word crybabies and noobs here when you and some others here still continues to go in a massive circlejerking about something you clearly have no clues about. So maybe you should grow up and get teached about what benchmarking is before you speak nonsense?

STOP IT.

I never said they shouldn't run under 100%, I said they shouldn't change system settings to achieve more than the 100% that is available to apps by default.

SierraSonic said,
The thing that Samsung is doing is changing the way the device runs once it detects these apps.

They have to, if not the throttling system will throttle the speed down on the benchmarking apps witch makes the benchmark scores a waste. Or the whole point in running a benchmark then will be wasted.

It's just a way to go around the speed throttling that is mainly made to save battery. But hey, running benchmark apps is not there to save your battery.

Get a clue, KTNXBAI.

HipHopSinceFriday said,
False

When car makers do an MPG test, the final score is based on an average score based on doing several driving conditions. If they say in open highway driving at 55mph, that I should expect a gas mileage of 40 MPG, then I should get anywhere from 38 to 42MPG...40 is simply the average...it is not the exact.

Again this is based on how they tested, not how you drive. As a fact the only way to get such mileage is to use the cruise control because your car will mostly stay at 55 unless there are lots of hills. When you drive with your foot, it gets tired and you fluctuate the speed more often up an down. Thus your mileage may vary.

This is nowhere close to the benchmarking tests. It is not the same. In this phone the CPU speed is 2.3Ghz per core PERIOD. In the tests, the Samsung device called up a java file that simply ran the cores at 2.3Ghz for all the tests. If you look at the tests, even though all 4 cores were at 2.3Ghz, they were not at a full 100%. Which means the phone for example in a video test could play the video at 30fps with the cores running at 2.3Ghz but they were at only 3% usage. Which means in real world usage the phone would play the same video with the CPU at a much lesser clock and still perform as expected. PERIOD. They didn't cheat, they simply past the test with flying colors instead of just simply passing the test.

No, they cheated, they programmed their cars to specifically look better on the tests, that is why the tests keep changing to more reflex real world performance, and what people expect.

The problem with your logic is that you believe the 100% Samsung forces these apps to run at is the 100% they would normally run at, which isn't true as these apps are forced to run out of system specs.

Exynos said,

They have to, if not the throttling system will throttle the speed down on the benchmarking apps witch makes the benchmark scores a waste. Or the whole point in running a benchmark then will be wasted.

It's just a way to go around the speed throttling that is mainly made to save battery. But hey, running benchmark apps is not there to save your battery.

Get a clue, KTNXBAI.

No, they don't because people do not care about how their apps could run, they care about how they will run at the default 100%, not the unrealistic 100%.

What is so hard to get about that these apps are being given boosts to show better scores than they should? If only benchmarking apps are given the boost to stay at 100% and games aren't, the games won't run at the speed the benchmarks falsely say they should be running at.

SierraSonic said,
No, they don't because people do not care about how their apps could run, they care about how they will run at the default 100%, not the unrealistic 100%.

What is so hard to get about that these apps are being given boosts to show better scores than they should? If only benchmarking apps are given the boost to stay at 100% and games aren't, the games won't run at the speed the benchmarks falsely say they should be running at.


They are given the boost because the benchmark apps are supposed to run at full speed, not half the speed or 75% of the maximum speed. But as the throttling system will throttle the speed down rather than keeping it at max all the time, then Samsung, HTC and LG and many others have made sure to let the benchmarking apps and other apps that needs the full speed available be able to go around the throttling system.

Every benchmarking apps on a computer does the same, so why is it wrong the way Samsung does it then?

SierraSonic said,
I never said they shouldn't run under 100%, I said they shouldn't change system settings to achieve more than the 100% that is available to apps by default.

Again, learn what speed throttling is and come back to me and say why some apps are allowed to go around the speed throttling system.

It's a logic behind it all.

Exynos said,

They are given the boost because the benchmark apps are supposed to run at full speed, not half the speed or 75% of the maximum speed. But as the throttling system will throttle the speed down rather than keeping it at max all the time, then Samsung, HTC and LG and many others have made sure to let the benchmarking apps and other apps that needs the full speed available be able to go around the throttling system.

Every benchmarking apps on a computer does the same, so why is it wrong the way Samsung does it then?
Again, learn what speed throttling is and come back to me and say why some apps are allowed to go around the speed throttling system.
It's a logic behind it all.

First off, do not say all apps are like this, it's not the apps that are doing this, it's the cheat that Samsung, and others, are using to detect benchmark apps. Benchmarks on the computer do NOT change the power profile or disable downclocking, it is up to the user to change those settings.

I'm just saying benchmarks should represent the way the system runs, not using cheat tactics to give it better numbers, than the system performs at.

Listen, I am not disagreeing that benchmarks are there to test the systems performance at 100%, I am disagreeing with the fact that certain companies are changing the way the system performs JUST for benchmarks, which causes these benchmarks to be offset from where they are supposed to be.

Read this, it should better explain how this is wrong.
http://www.anandtech.com/show/...ating-in-android-benchmarks

SierraSonic said,
First off, do not say all apps are like this, it's not the apps that are doing this, it's the cheat that Samsung, and others, are using to detect benchmark apps. Benchmarks on the computer do NOT change the power profile or disable downclocking, it is up to the user to change those settings.

I'm just saying benchmarks should represent the way the system runs, not using cheat tactics to give it better numbers, than the system performs at.

Listen, I am not disagreeing that benchmarks are there to test the systems performance at 100%, I am disagreeing with the fact that certain companies are changing the way the system performs JUST for benchmarks, which causes these benchmarks to be offset from where they are supposed to be.

Read this, it should better explain how this is wrong.
http://www.anandtech.com/show/...ating-in-android-benchmarks


First off, stop being arrogant and read what i write. I'm getting more and more tempted to just report you because you are not here to discuss, but to troll and make stupid arguments.

OMG, are you for real?, first you say that you agree that benchmarks are there to test the systems performance at 100%, and then you say it's a cheat that Samsung, HTC and LG are making sure that the benchmarking apps actually use 100% of the hardwares potential because the throttling system are made so the benchmark apps can't run at full speed.

Make up your mind?

Exynos said,

First off, stop being arrogant and read what i write. I'm getting more and more tempted to just report you because you are not here to discuss, but to troll and make stupid arguments.

OMG, are you for real?, first you say that you agree that benchmarks are there to test the systems performance at 100%, and then you say it's a cheat that Samsung, HTC and LG are making sure that the benchmarking apps actually use 100% of the hardwares potential because the throttling system are made so the benchmark apps can't run at full speed.

Make up your mind?

Report me? HAHA Good laugh.

What are you not getting, the 100% the benchmark apps is NOT the same 100% apps get to use, how am I trolling or not listening to you when I am trying to explain to you that the 100% that samsung is forcing the benchmark apps to use may use all the hardware, but its UNFAIR because NO APPS OTHER THAN THE BENCHMARK APPS WILL HAVE ACCESS TO THAT MUCH OF THE HARDWARE, AND SINCE THEY DO BENCHMARK WITH BETTER SPECS THAN ALL THE OTHER APPS HAVE ACCESS TO THE BENCHMARK IS SHOWING A FALSE SCORE MADE TO INFLATE THE SPEED OF WHICH ALL APPS WILL RUN AT.

The 100% the benchmark use is faster than any other app will get to use because of the default throttling being disable JUST FOR benchmarking. This makes the benchmarks run faster than any other app, thus it's not showing the true potential of the hardware THAT ALL APPS CAN USE. That is why it is a lie, cheating, etc!

The article shows that on average these benchmarks inflate the score by 4.4%, so all other apps will run 4.4% slower than the benchmark app says because they won't use that much hardware.

Go ahead and try to report me for explaining this to you, I am sure the moderator will have a good laugh. They too will most likely refer you to the link and show you that same link that shows these devices inflate their score just to look better, but not run better for EVERY APP.

For a benchmark to have any meaning, it must show the TRUE potential of the device, not an inflated one that works better ONLY BECAUSE OF SPECIAL SETTINGS THAT APPLY TO BENCHMARK APPS ONLY.

Edited by SierraSonic, Oct 2 2013, 9:08pm :

SierraSonic said,
Report me? HAHA Good laugh.

What are you not getting, the 100% the benchmark apps is NOT the same 100% apps get to use, how am I trolling or not listening to you when I am trying to explain to you that the 100% that samsung is forcing the benchmark apps to use may use all the hardware, but its UNFAIR because NO APPS OTHER THAN THE BENCHMARK APPS WILL HAVE ACCESS TO THAT MUCH OF THE HARDWARE, AND SINCE THEY DO BENCHMARK WITH BETTER SPECS THAN ALL THE OTHER APPS HAVE ACCESS TO THE BENCHMARK IS SHOWING A FALSE SCORE MADE TO INFLATE THE SPEED OF WHICH ALL APPS WILL RUN AT.

The 100% the benchmark use is faster than any other app will get to use because of the default throttling being disable JUST FOR benchmarking. This makes the benchmarks run faster than any other app, thus it's not showing the true potential of the hardware THAT ALL APPS CAN USE. That is why it is a lie, cheating, etc!

The article shows that on average these benchmarks inflate the score by 4.4%, so all other apps will run 4.4% slower than the benchmark app says because they won't use that much hardware.

Go ahead and try to report me for explaining this to you, I am sure the moderator will have a good laugh. They too will most likely refer you to the link and show you that same link that shows these devices inflate their score just to look better, but not run better for EVERY APP.


I have already reported another one here for trolling now, so don't be so sure that i wont do the same on you.

Again, explain to me why you agree that benchmarking apps are supposed to run at full speed (100%) and then at the same time says it's a cheat that Samsung and the others are making sure to let the named benchmarking apps in their script to actually run at 100% on their phones so the benchmarking apps doesn't get speed throttled, witch again gives false benchmarking results?

Are you saying benchmarking app are allowed to run at full speed or not?

If yes, then what's wrong with the script Samsung have made to let the benchmarking apps to run at full speed?

Make up your mind.

Exynos said,

I have already reported another one here for trolling now, so don't be so sure that i wont do the same on you.

Again, explain to me why you agree that benchmarking apps are supposed to run at full speed (100%) and then at the same time says it's a cheat that Samsung and the others are making sure to let the named benchmarking apps in their script to actually run at 100% on their phones so the benchmarking apps doesn't get speed throttled, witch again gives false benchmarking results?

Are you saying benchmarking app are allowed to run at full speed or not?

If yes, then what's wrong with the script Samsung have made to let the benchmarking apps to run at full speed?

Make up your mind.

They are allow to run full speed, THAT ANY OTHER APP CAN ACHIEVE, if they are a special class of software that performs better ONLY because there is a 'special script" to boost their performance, that is what I call a cheat.

If every other app is throttled, then the throttled speed is the max speed of the device. If Samsung wanted it to run faster, then all apps should run faster at the cost of battery life, but if they are using a profile which saves battery life in favor performance and all apps are required to follow it, then so should the benchmarking apps.

That is why I keep saying that the benchmarks 100% is different than the other apps 100%, because it is truly showing a false speed. Can you explain to me why you think that one app that shows the speed of the device is allowed to run faster than any other app can still be used as a reference of the speed of the device?

so basically, when it's running certain programs, it speeds up the CPU? I thought computers were doing this for a long time?

i mean, if it can speed it up when the benchmark is running, what's stopping it from speeding up when a game or anything else is running?

Buttus said,
so basically, when it's running certain programs, it speeds up the CPU? I thought computers were doing this for a long time?

i mean, if it can speed it up when the benchmark is running, what's stopping it from speeding up when a game or anything else is running?

They have a hardcoded list of benchmark apps in the firmware. It doesn't do it for normal apps. It's purely to make it look faster in reviews than it actually will perform normally.

What keeps them from doing that is killing the battery. If the Note 3 always ran in this "boost mode" (anyone know if this is just overclocking?) then the battery would surely die much quicker. Doing it this way makes it seem as if the Note 3 is faster as well as getting great battery life, when in reality it's cheating.

Computers that aren't running benchmarks don't do this for other apps either unless it is taxing the CPU.

So, I am not really seeing the issue here. Now if they were overclocking it during benches, I would have more reason to hate on them.

chAos972 said,

They have a hardcoded list of benchmark apps in the firmware. It doesn't do it for normal apps. It's purely to make it look faster in reviews than it actually will perform normally.


They don't do it on many of the normal apps because it's not needed. They wont run any better as they do normally over how they would run if the speed on the CPU / GPU was boosted to the maximum. The only thing it would do is to drain the battery faster without reasons.

chAos972 said,

They have a hardcoded list of benchmark apps in the firmware. It doesn't do it for normal apps. It's purely to make it look faster in reviews than it actually will perform normally.

Which is exactly what benchmarking software does anyways. It pushes hardware well an above what a typical user will push hardware anyways.

chAos972 said,

They have a hardcoded list of benchmark apps in the firmware. It doesn't do it for normal apps. It's purely to make it look faster in reviews than it actually will perform normally.

So what? Benchmarking tools aren't normal apps. What is your point? How many consumers use benchmarking apps as an everyday application? That's right, none.

Benchmarking tools are just that. It is a tool designed to push hardware above and beyond what a typical user would do which would be about 99% of Samsung's user-base. Fact is, the people who buy these product, have never seen or care about benchmarks.

I'm glad the media is coming down on them harder this time. They should've taken the hint last time that it's not acceptable.

Yep they did the same with the Galaxy S4. It needs to stop. It's like the earlier days of Nvidia and ATI with benchmark driver optimisations.

You can read my earlier post here (the long post) about this and explain to me why this is not acceptable when Samsung does this like everyone else who does benchmarks seriously?

Exynos said,
You can read my earlier post here (the long post) about this and explain to me why this is not acceptable when Samsung does this like everyone else who does benchmarks seriously?

I read it, though respectfully disagree. I don't think it's acceptable just because that's what another industry does. Consumers want accurate benchmarks and Samsung is perverting them.

It's either everyone does it, or no one does it. Right now, all but one is the latter.

Except that these aren't driver optimizations. These are just putting the CPU on performance mode for the benchmark.

Which is how benchmarks should be done anyway.

Other OEMs aren't smart enough to do that apparently. Samsung understands that benchmarks sell things, regardless of how stupid it is.

chAos972 said,

I read it, though respectfully disagree. I don't think it's acceptable just because that's what another industry does. Consumers want accurate benchmarks and Samsung is perverting them.

It's either everyone does it, or no one does it. Right now, all but one is the latter.


So running benchmarks at full power is cheating and not right?

Isn't the idea of benchmarks to measure what the hardware is capable of?

Do you know what the main purpose of a benchmark tool is?

Seems to me that you aren't living on the same planet as us others are doing.

Exynos said,

So running benchmarks at full power is cheating and not right?

Isn't the idea of benchmarks to measure what the hardware is capable of?

Do you know what the main purpose of a benchmark tool is?

Seems to me that you aren't living on the same planet as us others are doing.

Yes, it provides a value/values that can be used to compare things. These numbers aren't accurate to what the consumer will get in everyday use (unless all they want the device for is to run benchmarks all day), so the numbers are valueless, the benchmark is pointless, and Samsung (along with any company that fudges these numbers) is full of s**t.

Exynos said,
You can read my earlier post here (the long post) about this and explain to me why this is not acceptable when Samsung does this like everyone else who does benchmarks seriously?

Here comes our resident astroturfer again

Boosting CPU/GPU performance to alleviate performance bottlnecks in specific applications (with an "eight core" processor *cough*) is par for the course. Boosting frequencies to have better scores in a hardcoded lists of benchmarks has always been frowned upon : for example the nVidia/ATI texture filtering "optimizations" back in the day (remember the Quack experiment?)

A benchmark is meaningful only if everyone plays on the same level. Shadily boost your results and you defeat the benchmark's purpose.
Oh, and legally, this kind of "shenanigans" are called intent to deceive.

Arkos Reed said,

Here comes our resident astroturfer again

Boosting CPU/GPU performance to alleviate performance bottlnecks in specific applications (with an "eight core" processor *cough*) is par for the course. Boosting frequencies to have better scores in a hardcoded lists of benchmarks has always been frowned upon : for example the nVidia/ATI texture filtering "optimizations" back in the day (remember the Quack experiment?)

A benchmark is meaningful only if everyone plays on the same level. Shadily boost your results and you defeat the benchmark's purpose.
Oh, and legally, this kind of "shenanigans" are called intent to deceive.


Again, you talk like you have absolutely no experience what so ever about this thing.

Samsung isn't boosting anything with it's CPU or GPU. They just make sure to let them run on full power on what the CPU and GPU is able to deliver on default speeds like everyone else does with benchmarks on computers.

Explain why this isn't the way benchmarks should be run?

virtorio said,
Yes, it provides a value/values that can be used to compare things. These numbers aren't accurate to what the consumer will get in everyday use (unless all they want the device for is to run benchmarks all day), so the numbers are valueless, the benchmark is pointless, and Samsung (along with any company that fudges these numbers) is full of s**t.

If you have an intensive app, it is supposed to run faster, which is what a benchmark is. No one runs their phone at 100% all day long, so no, you won't get that in the real world on a regular basis, but it does show what it can peak at. You guys don't really know what you're even complaining about.

Who cares about the points, you'll still hit your max frequencies, WHEN it's needed.

Exynos said,

Again, you talk like you have absolutely no experience what so ever about this thing.

Samsung isn't boosting anything with it's CPU or GPU. They just make sure to let them run on full power on what the CPU and GPU is able to deliver like everyone else does with benchmarks on computers.

Explain why this isn't the way benchmarks should be run?

You see, it happens I do have some experience on the matter, unlike you, I do benchmarks on a regular basis to buy hardware for a local government in Europe.

Let's take the most recognized and respected benchmark in the computing world, spec.org's SPEC CPU2006.
They have two sets of results in their benchmarking suite for their Integer and FPU tests :
One of the result sets presents scores based on applications compiled with parameters like anyone would find in a standard configuration, which is the most common case.
The other set of scores presents performance on the same applications compiled with parameters specifically optimized for the CPU/Server architecture, these scores are higher, the parameters tightly controlled by the vendor and they are PUBLIC, they don't hide anything

Two sets, standard and high perf, both using PUBLISHED parameters, there's no cheating, no hiding bull**** like you are implicitly condoning to cover for your employer (because honestly, not even a fanboy would be so rabid to defend a brand)

Arkos Reed said,

You see, it happens I do have some experience on the matter, unlike you, I do benchmarks on a regular basis to buy hardware for a local government in Europe.

Let's take the most recognized and respected benchmark in the computing world, spec.org's SPEC CPU2006.
They have two sets of results in their benchmarking suite for their Integer and FPU tests :
One of the result sets presents scores based on applications compiled with parameters like anyone would find in a standard configuration, which is the most common case.
The other set of scores presents performance on the same applications compiled with parameters specifically optimized for the CPU/Server architecture, these scores are higher, the parameters tightly controlled by the vendor and they are PUBLIC, they don't hide anything

Two sets, standard and high perf, both using PUBLISHED parameters, there's no cheating, no hiding bull**** like you are implicitly condoning to cover for your employer (because honestly, not even a fanboy would be so rabid to defend a brand)


Again, you still don't get the point. The point is that anyone who make games who have a Benchmark tool for it's games will run it's benchmarks to test the full potential of the CPU / GPU, not to run it when there is under average with stress on the CPU / GPU.

Do you think PCMark runs your CPU, GPU, RAM and so on on half speed under the benchmark?

Do you think HDD benchmarks runs their benchmarks on a big file to see what your HDD or SSD delivers under 70% read & write speed instead of 100%?

Do you think 3DMark runs the GPU at an average / normal speed when the point of the 3DMark is to run the GPU at it's full potential?

Again, why are the way Samsung gets the benchmark tools to behave on their Galaxy phones any different than what 3DMark, PCMark, CrystalDiskMark, IntelBurnTest, Prime95, Super PI and so on does on our computers?

This is nothing new and it's used by everyone, even HTC and LG to as explained here: http://www.slashgear.com/galax...igans-accusations-01299884/

Exynos said,

Again, you still don't get the point. The point is that anyone who make games who have a Benchmark tool for it's games will run it's benchmarks to test the full potential of the CPU / GPU, not to run it when there is under average with stress on the CPU / GPU.

Do you think PCMark runs your CPU, GPU, RAM and so on on half speed under the benchmark?

Do you think HDD benchmarks runs their benchmarks on a big file to see what it delivers under 70% load instead of 100%?

Do you think 3DMark runs the GPU at an average / normal speed when the point of the 3DMark is to run the GPU at it's full potential?

Again, why are the way Samsung gets the benchmark tools to behave on their Galaxy phones any different than what 3DMark, PCMark, CrystalDiskMark, IntelBurnTest, Prime95, Super PI and so on does?

This is nothing new and it's used by everyone, even HTC and LG to as explained here: http://www.slashgear.com/galax...igans-accusations-01299884/

You are comparing apples to oranges.

You are comparing smartphones, constrained by a set of limits, such as battery life and internal temperature, to PCs, which run on DC, have active cooling and are generally built to perform as fast as is possible.

By the way, Sequential speed based HDD benchmarks perform better on big files, proof among others that you are talking uninformed pseudo PR nonsense

Oh and because others do it makes it right? You're the butthurt one here little boy.

Do everyone a favor and stop digging, the walls may crumble down on you.

Who cares how fast something can go if you won't have access to that power outside of the benchmarking tool? It's bad enough that people use benchmarking tests to determine which phone to buy when they won't even be using that much power realistically, but then to have a test where the hardware exceeds what it's expected limitations are for real use, is even more misleading.

Why show me things that I won't be able to do? That's just nonsense advertising.

Also, two wrongs don't make a right. Samsung is still awful for doing this.

Oh you know, the force is strong with this troll / Samsung astroturfer.

I bet he's still trying to reply, grasping at straws to concoct one more inept silly rebuttal. However I'm glad I read this thread and spent some time dismantling his nonsense, I hope I have cleared the minds of a few readers along the way.

Exynos said,

Again, you still don't get the point. The point is that anyone who make games who have a Benchmark tool for it's games will run it's benchmarks to test the full potential of the CPU / GPU, not to run it when there is under average with stress on the CPU / GPU.

The problem is that (according to the article) you won't ever be getting the performance shown in the benchmark when using the device.

If the device was actually peaking at the same point when using any other resource intensive app (eg. games) it would be perfectly OK, but if I understand correctly it won't. You only get that performance when the device detects a specific benchmarking app running.

If Samsung wanted to show the peak performance of their device (even though you wouldn't be able to get that when using the device yourself) they could have disclosed how to benchmark it under both scenarios: maximum performance and actual performance for apps you'll be running.

Arkos Reed said,

You are comparing apples to oranges.

You are comparing smartphones, constrained by a set of limits, such as battery life and internal temperature, to PCs, which run on DC, have active cooling and are generally built to perform as fast as is possible.

By the way, Sequential speed based HDD benchmarks perform better on big files, proof among others that you are talking uninformed pseudo PR nonsense

Oh and because others do it makes it right? You're the butthurt one here little boy.

Do everyone a favor and stop digging, the walls may crumble down on you.


You still haven't explained why benchmark apps shouldn't run on full power?

And you still haven't explained why everyone does this including HTC and LG to while they aren't supposed to do it according to you?

When you can't explain this, you can't be taken seriously at all.

ichi said,

The problem is that (according to the article) you won't ever be getting the performance shown in the benchmark when using the device.

If the device was actually peaking at the same point when using any other resource intensive app (eg. games) it would be perfectly OK, but if I understand correctly it won't. You only get that performance when the device detects a specific benchmarking app running.

If Samsung wanted to show the peak performance of their device (even though you wouldn't be able to get that when using the device yourself) they could have disclosed how to benchmark it under both scenarios: maximum performance and actual performance for apps you'll be running.


Just because most of the apps isn't runned at full power because it's not needed as they already run as good as it's possible to run them at 1/3 of the power of the CPU and GPU have, it doesn't mean all benchmark apps have to do the same.

Does Firefox, Winamp, Dropbox, Google Drive and so on on a descent computer today use all of the power on the CPU and GPU to run?

No they don't, so why should they run on full power just to waste alot of watt or battery on a laptop every hours just because benchmark apps are designed to use every part of a computer hardwares to the limits, aka 100%?

Edited by Exynos, Oct 1 2013, 11:00pm :

Arkos Reed said,
Oh you know, the force is strong with this troll / Samsung astroturfer.

I bet he's still trying to reply, grasping at straws to concoct one more inept silly rebuttal. However I'm glad I read this thread and spent some time dismantling his nonsense, I hope I have cleared the minds of a few readers along the way.


Yeah, you think you have explained why you are right. But i'm happily gonna tell you that you are both wrong and you don't know how benchmarks apps are supposed to work. It's that simple.

I have been running benchmarks apps on all of my computers for all of the 10-12 years i have had a computer, so i'm pretty sure on how benchmark apps is supposed to work.

Exynos said,

Just because most of the apps isn't runned at full power because it's not needed as they already run as good as it's possible to run them at 1/3 of the power of the CPU and GPU have, it doesn't mean all benchmark apps have to do the same.

Does Firefox, Winamp, Dropbox, Google Drive and so on on a descent computer today use all of the power on the CPU and GPU to run?

No they don't, so why should they run on full power just to waste alot of watt or battery on a laptop every hours just because benchmark apps are designed to use every part of a computer hardwares to the limits, aka 100%?

Most apps do indeed not need the phone to run at 100%, and it's completely fair and logic that the phone doesn't run at 100% as you'll still get great performance with better battery life.

Still, the benchmark's score with the Note3 is achieved by running the phone at 100%, which is OK if you want to show the performance of the phone's components but is not representative of the actual performance of the phone when used by customers.

The reason Samsung is being called on this is because there's an expectation for phone benchmarks to be representative of the performance you'll be getting out of the device, which is not the case here.

ichi said,

Most apps do indeed not need the phone to run at 100%, and it's completely fair and logic that the phone doesn't run at 100% as you'll still get great performance with better battery life.

Still, the benchmark's score with the Note3 is achieved by running the phone at 100%, which is OK if you want to show the performance of the phone's components but is not representative of the actual performance of the phone when used by customers.

The reason Samsung is being called on this is because there's an expectation for phone benchmarks to be representative of the performance you'll be getting out of the device, which is not the case here.


I don't know, but a benchmark tool is not here to show you the representative of the actual performance of the phone when used by customers. It's there to show you the full potential of the hardware at 100% power.

Do you know the difference from a benchmarking tool to a normal usage situation?

adrynalyne said,
Benchmarks by reviewers of PCs aren't accurate to what a customer will get everyday either.

Exactly. This is something the Samsung haters doesn't want to see or admit.

Exynos said,

So running benchmarks at full power is cheating and not right?

Isn't the idea of benchmarks to measure what the hardware is capable of?

Do you know what the main purpose of a benchmark tool is?

Seems to me that you aren't living on the same planet as us others are doing.

When it's a mobile device and it isn't feasible to run at full power 100% of the time, it is absolutely cheating. Let's face it, these devices weren't designed that way and provided you can carry a car battery with you to supply power through the day, you're likely to run into extreme thermal throttling to further rain on your parade. You, your name and behavior have earned you the title of Asshat. Congratulations.

Furthermore, something this article doesn't state is that with previous Samsung devices, they boosted the clock speeds to unobtainable levels. Yes, you heard that right, unless you were running a benchmark app you could never obtain these clocks short of rooting and OCing your device; not for games, or any other demanding apps. Still think it isn't cheating?

Edited by mantequillas, Oct 2 2013, 12:00am :

Exynos said,

Exactly. This is something the Samsung haters doesn't want to see or admit.

Nonsense. If everyone did the tests on phones in the same form and fashion, I wouldn't care. Apples and oranges though.

And boo-friggin-hoo at Samsung haters. It wasn't very long ago everyone was supporting them (see Apple vs Samsung). If there's any heat they're getting today, it's due to their own actions.

mantequillas said,

When it's a mobile device and it isn't feasible to run at full power 100% of the time, it is absolutely cheating. Let's face it, these devices weren't designed that way and provided you can carry a car battery with you to supply power through the day, you're likely to run into extreme thermal throttling to further rain on your parade. You, your name and behavior have earned you the title of Asshat. Congratulations.

Furthermore, something this article doesn't state is that with previous Samsung devices, they boosted the clock speeds to unobtainable levels. Yes, you heard that right, unless you were running a benchmark app you could never obtain these clocks short of rooting and OCing your device; not for games, or any other demanding apps. Still think it isn't cheating?


Doesn't matter what it is. I simply don't care if it's a calculator or an e-reader or a smartphone. If i want to see what those devices can deliver in performance, i will run some benchmark apps to find that out, because that's the only way to see what those devices are able to give us in performance at full power. Yes, mark my words 'FULL POWER'. I hope that word in caps lock tells you the whole thing.

dead.cell said,

Nonsense. If everyone did the tests on phones in the same form and fashion, I wouldn't care. Apples and oranges though.

And boo-friggin-hoo at Samsung haters. It wasn't very long ago everyone was supporting them (see Apple vs Samsung). If there's any heat they're getting today, it's due to their own actions.


Doing benchmark on a computer is the same as doing it on a smartphone. Both things use an OS, GPU, CPU & RAM.

Exynos said,

Doing benchmark on a computer is the same as doing it on a smartphone. Both things use an OS, GPU, CPU & RAM.

Duh.

We know how benchmarks work. What you don't seem to get is why it matters that hardware for one phone should be allowed to drive faster than what it would normally operate outside of a benchmark, when others are not doing the same.

If all phones were measured in this manner, then that's totally cool beans. No one would care! However, how would you explain Samsungs measurements to be anywhat at all comparable to others, if others aren't doing the same? They don't! Thus, is makes it misleading, plain and simple.

I'm not going to argue this further. If you disagree, fine. But Samsung is still awful for doing this, plain and simple. (need proof? see as evidenced by... well, numerous reputable websites)

dead.cell said,

Duh.

We know how benchmarks work. What you don't seem to get is why it matters that hardware for one phone should be allowed to drive faster than what it would normally operate outside of a benchmark, when others are not doing the same.

If all phones were measured in this manner, then that's totally cool beans. No one would care! However, how would you explain Samsungs measurements to be anywhat at all comparable to others, if others aren't doing the same? They don't! Thus, is makes it misleading, plain and simple.

I'm not going to argue this further. If you disagree, fine. But Samsung is still awful for doing this, plain and simple. (need proof? see as evidenced by... well, numerous reputable websites)


As it have been pointed out several times already, others like HTC and LG are doing the same: http://www.slashgear.com/galax...igans-accusations-01299884/

And quess why?

Yes, because it's normal under benchmarking.

dead.cell said,
If it's normal, then why did Samsung deny doing this?

Samsung denied that they was boosting their CPU and GPU out of what the default speed / power was, not because they was letting benchmark apps to use the full potential of the hardware. Because doing that is just normal as you can see when HTC and LG are doing the same to like everyone does today with benchmarking apps.

dead.cell said,
Except if it was normal, we wouldn't be trying to figure out who's doing this, and who's not. Still misleading.

If this is misleading, then are the rest of the benchmark tools out there are misleading to. Right?

Do you actually know the main purpose of a benchmarking tool?

Exynos said,

I don't know, but a benchmark tool is not here to show you the representative of the actual performance of the phone when used by customers. It's there to show you the full potential of the hardware at 100% power.

Do you know the difference from a benchmarking tool to a normal usage situation?

The point of benchmarks is measuring performance, not specifically at 100% but under any condition where the measurement might be useful.

Eg. if I wanted to know where resource intensive Android games would be running better I would be benchmarking phones under the exact conditions those games would be running. Boosting the hardware to it's full capacity would give scores that have little to nothing to do with the actual performance I'd be getting when playing games.

This would be different if they had some sort of API call where you could set the device in high performance mode for any app that requests it, but as it is they are relying on a whitelist of benchmarking apps. What you'll get is inconsistent benchmark results if you happen to run a benchmark that's not whitelisted (as shown when they ran a renamed Geekbench).

Edited by ichi, Oct 2 2013, 8:05am :

chAos972 said,

I read it, though respectfully disagree. I don't think it's acceptable just because that's what another industry does. Consumers want accurate benchmarks and Samsung is perverting them.

It's either everyone does it, or no one does it. Right now, all but one is the latter.

Actually consumers could give to shakes about benchmarks. They mean nothing.

Samsung made an app so that when benchmarks are done, it allows the hardware to run 20% faster.

Question, isnt the point of benchmarks is to push the hardware to it highest possible point to see what the hardware can do? YES!!!!

So...who cares if in normal usage you will never use that extra 20%. We all know no one will. The point of benchmarks is to see exactly wjhat the hardware is capable of, now what we will actually do in real life.

The reality of fact is, benchmrks do exactly what I said. They show what the hardware can do, its not saying you will use it that way, but if you want too like if you install a custom ROM< you do get an idea how hardware you can push the hardware.

Fact is, everyone is hating on Samsung and trying ot make them look bad because they are kicking EVERYONE butt in the market and bad press is suppose to hurt them. Just a like a great politician who is accused of screwing around on his partner so he can get bad press and fall from grace.

I dont give a crap what ARCS says, I am still buying the device because I have a Note II. Even tho I am a tech nut, I already that ALL benchmarks are nothing but BS. Anyone who even takes any benchmark score seriously should have their heads examine.

All that matter is, setting a phone net to another and doing simple tasks and see which one does them better. That is what a true benchmark is...what the eyes can see and what the hands can feel. Not some stupid software doing crap you can't see or feel.

Its a person who gets accused of a crime and then plays crazy. Metal diseases are BS. To easy to fake when you need to. But Cancer? Can be seen with your eyes and felt with your hands in many case. Something you cant fake
Benchmarking is a bunch of artificial garbage.

adrynalyne said,
Except that these aren't driver optimizations. These are just putting the CPU on performance mode for the benchmark.

Which is how benchmarks should be done anyway.

Other OEMs aren't smart enough to do that apparently. Samsung understands that benchmarks sell things, regardless of how stupid it is.

Someone who gets it. +100 for you sir.

HipHopSinceFriday said,
Someone who gets it. +100 for you sir.

Had to give adrynalyne a +1 to my self :-D

And yes, it takes like 3-4 clicks to set my Galaxy S4 in 'Performance' mode, or to the Performance governor witch lets me run the phone at 1.9 GHz constantly.

Arkos Reed said,

Here comes our resident astroturfer again

Boosting CPU/GPU performance to alleviate performance bottlnecks in specific applications (with an "eight core" processor *cough*) is par for the course. Boosting frequencies to have better scores in a hardcoded lists of benchmarks has always been frowned upon : for example the nVidia/ATI texture filtering "optimizations" back in the day (remember the Quack experiment?)

A benchmark is meaningful only if everyone plays on the same level. Shadily boost your results and you defeat the benchmark's purpose.
Oh, and legally, this kind of "shenanigans" are called intent to deceive.

The CPU in the phone is clocked at 2.3Ghz. In all the tests the CPU simply ran at 2.3Ghz...how is that cheating?

Who knows. Apparently methods of benchmarks a phone are considered cheating if you put the CPU to full speed.

I could understand if they overclocked it, or loaded the benchmark into a ram disk (I did that for a rom as a joke and it took off like wildfire in other roms), but this? This is silly to get upset about. Other OEMs should have thought of it themselves instead of relying on the default cpu governor. All they had to do was switch to performance governor during the benchmark and then back when done.

Which is probably how Samsung did it. Detect an app by name, switch to performance governor, and then go. No different than how those who root their phones and benchmark. Except they usually overclock too.

adrynalyne said,
Who knows. Apparently methods of benchmarks a phone are considered cheating if you put the CPU to full speed.

I could understand if they overclocked it, or loaded the benchmark into a ram disk (I did that for a rom as a joke and it took off like wildfire in other roms), but this? This is silly to get upset about. Other OEMs should have thought of it themselves instead of relying on the default cpu governor. All they had to do was switch to performance governor during the benchmark and then back when done.

Which is probably how Samsung did it. Detect an app by name, switch to performance governor, and then go. No different than how those who root their phones and benchmark. Except they usually overclock too.

I disagree. There is a reason why its cheating and a reason why its not. Benchmarks are suppose to push the device to its limits, so that you know what the hardware is capable of.

The past test was to show a score based on actual usage. What is the point of that? Fact - Side by side the Galaxy Note 3 mops the floor with the iPhone 5S. The SunSpider and Antutu scores mean nothing. All it means is, Apple has something in the browser coding that resolves rendering pages made with Java. The fact is, even if Samsung's browser has older java code for dealing with it, the fact that the CPU clocked up helps it push right through it. Look an the GN3 and the 5S side by side in a page made with Java. It still loads faster on the GN3...that is fact.

The point of a benchmark is, to be able to brag that my phone is better than yours. Even without the benchmarks which I never use anyways, I know the Note 3 is better than a 5S. 4 cores is better, 3GB of RAM is better, the multi-tasking is better, the resolution is better, the screen size is better, the fact it has a digitized stylus makes It better, the gimmicky features in TouchWiz are better, the fact that Android 4.3 doesn't slow down an older phone like the GS3 and GN2 shows that Android is better vs iOS which has made my iPhone 4 slower than a jar of molasses.