Samsung Again Caught Inflating Benchmarking Scores, Phil Schiller Calls 'Shenanigans'

[840quadra]

Have you read those articles? GPU is overclocked NOT ONLY for benchmarks but for some selected apps too. For example, camera app is executed in the same mode. This looks like a primitive implementation of turbo mode where overclocking is based on app type rather than on thermal conditions. Obviously Phil is too dumb to understand this.

Well lets not jump to conclusions on what he means / thinks based on a 1-word tweet. .

That said, I agree, the phone appears to be adapting in a way similar to Intel (and other companies) go into Turbo Boost under special conditions. It is quite possible that the thermal condition trigger is patented, or otherwise protected, meaning Samsung had to go an other route for their processors.

I also agree with many previous posts, the fact that the phone was able to reach those speeds, means, it is a valid test, regardless of the trigger for the speed bump.

As far as the iPhone is concerned, I am sure it is only a matter of time before those devices have more adaptive processor speeds, or cores than they currently do. Likely utilizing technology of Samsung, or other chip manufacturers.

 

[TTile]

But doesn't that mean the device IS in fact capable of going that fast, if coders write their apps to take advantage of it?

No, there are very specific reasons for not overclocking processors.

 

[everything-i]

Why not add code to games that does what what the code in Android did? Its an open standard, the designers can probably do that if they can find the right calls.

Because your battery would last half the time if they did. When the processor is over clocked like this it uses far more power.

 

[samcraig]

While it's not right to "forge" scores - ultimately all that matters is how the phone performs for the individual using it. Personally I don't look at specs or benchmarks - I use the phone. The way it performs from using apps to taking pictures to surfing the net - all important to me. Some # or technical jargon might win pissing contests - but at the end of the day are meaningless to the end user.

 

[Rafterman]

Why would you want a company which is deceitful and lies to get off scott-free?

Like when Apple lied about the iPhone 4's signal strength problems being just faulty signal bar code?

 

[iAlphard]

The only mistake Samsung made, they should put benchmark boost mode in the settings. If the hardware capable of doing that and no rendering or nor any test was bypassed, it is actually not cheating. It is being smart.

 

[Impuls3101]

CPU performance is honestly excellent. The Galaxy Note 3 is more or less the fastest Android smartphone we've tested up to this point. In the situations where we can do cross platform (OS/browser) comparisons, it isn't quite as fast as the iPhone 5s but in some cases it comes close. I should mention that the Note 3 (like many other Android devices - SGS4, HTC One) detects certain benchmarks and ensures CPU frequencies are running at max while running them, rather than relying on the benchmark workload to organically drive DVFS to those frequencies. Max supported CPU frequency is never exceeded in this process, the platform simply primes itself for running those tests as soon as they're detected. The impact is likely small since most of these tests should drive CPU frequencies to their max state regardless (at least on the CPU side), but I'm going to make it a point to call out this behavior whenever I see it from now on.

http://www.anandtech.com/show/7376/samsung-galaxy-note-3-review/4

 

[nzalog]

I can't believe a company that cheated in the past and steals other people's work is cheating...
/s

I lost no respect for Samesung, because I haven't had any in a lonng time now.

 

[Tones2]

This thread cracks me up. I think I just reached the point of oversatuation with all this pissy-fighting. We are now to the point that we have TWO great phone companies making GREAT phones and it doesn't really matter much anymore whose is slightly faster at what time - they are both FAST enough. It's more about screen size and ecosystem preference - there is no "better" but just preference. Can't we are just stop all of this non-sense. Jeez....

As for the cheating and questionable business practice thing, neither party (like any corporation) is innocent, and I could care less. Just give me good phones and devices and keep the competition level high to continue to push the bounderies.

 

[I am Andy]

No Samsung

Before and after pics or it didnt happen.

I've already done that.
The only leftover is a TV but this will not last that long.
When buying new things I'm ignoring Shamesung.
Btw. have you read all those Samsung trolls
It's funny to see their reaction on any pro-Apple or anti-Samsung news.
Do they really think that we don't see that???

 

[ActionableMango]

Are there people that choose a phone based on CPU benchmarks?

 

[Jman13]

Anyone else notice that the Note 3's 'normal' mode has a higher overall score than the iPhone 5S? Per the results on MacWorld.com. I didn't buy Geekbench app and test myself. Only like 2 points though.

It's essentially the same, but with a slight edge to the 5S.

Here's the Note 3 as tested without the benchmark overclocking, vs. my iPhone 5S (just ran Geekbench 3 on my phone):

Overall: Note 3: 2,487 iPhone 5S: 2,568 Slight advantage to my 5S. (3%)
Integer: Note 3: 2,800 iPhone 5S: 2,874 Slight advantage to my 5S. (2.6%)
Floating Point: Note 3: 2,677 iPhone 5S: 2,666 Very slight advantage to Note 3 (0.4%)
Memory: Note 3: 1,481 iPhone 5S: 1,761 Big advantage to the 5S. (18.9%)

Overall, the 5S tests slightly faster...at least my 5S tests slightly faster. Which is incredible given it's a dual core chip vs. quad-core, plus much smaller battery requirements with similar battery life.

 

[JAT]

No, it's more nefarious than that. It's like increasing the boost on a turbocharged car specifically for cars that go to car mags to be tested for 0-60 times and 1/4 mile and then selling cars to the public that are much slower as the higher boost levels aren't available to regular consumers as in long-term boosting causes overheating problems.

I'd say it's more like claiming your car can do 60000 mph...cause the Earth is moving that fast.

 

[mbh]

But the cores did produce those results, but they just aren't in line with the unwritten rule of what a benchmark should be testing for? Got it.

Kind of like how the unwritten rule about measuring a piece of lumber is to find out how long it is. You could use the length of the tree it was cut from, but that's not really the point, is it?

 

[TTile]

Are there people that choose a phone based on CPU benchmarks?

Talk to any S4 owner...

 

[slapppy]

Stop jerking around. My comment was aiming toward samsung PRODUCTS, not COMPONENTS.

It's typical of Apple haters and Samsung Apologist. Even if the last component is a .02 cents piece of plastic strip, the same argument posted will be used. LOL

 

[SoldOnApple]

I'm personally banning ALL samsung device in my house. Washer machine, ssd, screen, tv and ....

And processors?

 

[GQB]

This is pretty hilarious actually. Why not just always run the CPU this way

Sure, if you want (at least) 20% reduced battery life... probably more.

 

[JAT]

Are there people that choose a phone based on CPU benchmarks?

There are people that claim to.

 

[samcraig]

This thread cracks me up. I think I just reached the point of oversatuation with all this pissy-fighting. We are now to the point that we have TWO great phone companies making GREAT phones and it doesn't really matter much anymore whose is slightly faster at what time - they are both FAST enough. It's more about screen size and ecosystem preference - there is no "better" but just preference. Can't we are just stop all of this non-sense. Jeez....

As for the cheating and questionable business practice thing, neither party (like any corporation) is innocent, and I could care less. Just give me good phones and devices and keep the competition level high to continue to push the bounderies.

 

[I am Andy]

This thread cracks me up. I think I just reached the point of oversatuation with all this pissy-fighting. We are now to the point that we have TWO great phone companies making GREAT phones and it doesn't really matter much anymore whose is slightly faster at what time - they are both FAST enough. It's more about screen size and ecosystem preference - there is no "better" but just preference. Can't we are just stop all of this non-sense. Jeez....

Of course there's a 'better'.
And most people know. Don't mix up Samsungs fantasy numbers and reality.

 

[iSee]

Does this seem more of a feature only to me?
When a particularly demanding task, like a benchmark, is executed, the CPU releases more power. I call this "efficiency".
And these numbers came out from the CPU's calculating power, not from nowhere.. how is this fake?

The problem is that's not what Samsung did. Samsung boosts the CPU based NOT on the load but on the package name of the app running.

Ars simply renamed Geekbench 3 to Stealthbench and benchmark scores went down 20%.

Your Samsung device will perform better when running benchmarks but nothing else. It's pretty clear this is a sneaky way for Samsung to claim great performance and great battery life and keep the size (and possibly price) of the phone down. It's pathetic, of course. And ultimately hurts their brand. The next Samsung device that comes out and performs well, the geekerati will largely assume Samsung is pulling another fast one.

 

[Tones2]

Overall, the 5S tests slightly faster...at least my 5S tests slightly faster. Which is incredible given it's a dual core chip vs. quad-core, plus much smaller battery requirements with similar battery life.

Remembering the biggest battery drain is NOT the CPU but rather the display - which is dependent on screen size. So the Note 3 larger battery is offset by the much larger screen size, so the simtlar battery lives make sense.

----------

Of course there's a 'better'.
And most people know. Don't mix up Samsungs fantasy numbers and reality.

Yep. This makes a lot of sense.

 

[iAlphard]

The ignorance level in this thread is too damn high and why the hell i even went here. Oh God why....

 

[samcraig]

Of course there's a 'better'.
And most people know. Don't mix up Samsungs fantasy numbers and reality.

It's not about numbers. It's about personal preference.

Both Samsung and Apple make outstanding phones. Both Android and iOS are excellent OSes. There are pluses and minuses to both hardware and software platforms.

Don't mix up your fantasy and reality. See what I did there?