Note 3?s benchmarking ?adjustments? inflate scores by up to 20%

[Ry]

Note 3?s benchmarking ?adjustments? inflate scores by up to 20%

 

[blitz118]

Yawn

 

[apollooff320]

wow...this is on almost every site I visit and the news is just getting here. Anyways Samsung needs to stop with this crap it's very misleading. How about letting all apps and games run at those clocks and not just benchmarks.

 

[thymeless]

That's LAME Samsung, LAME. Why would you act to make owning your product an embarassment?

 

[zach_alt]

This is such a non issue. As the Anadtech review says, most Android phones do this. WHO CARES?!

 

[power5]

Better question is why would any benchmark have cores idling or at 1/8th full speed? So this little code seems to not allow the cores to throttle down once the app is started. I suspect that they just don't need to speed up during a test and never turn down like it would otherwise. Not sure if that is whats giving the 20% boost, but the lack of throttling must help during testing.

 

[Blackintuit]

So when it's in power saving mode it will let 3 of the cores go idle, but when there's an intense workload it will turn on all 4 cores?

Pardon me, but that isn't a bug, it's a feature. You save battery when you don't need it, and when you do need it, all four cores of the CPU will actually be engaged.

Seems pretty normal Linux kernel behavior to me. Which is what Android is built on top of.

If the reviewer's going to complain about anything, they should complain about using synthetic benchmarks in the first place.

 

[mavrrick]

PC Gamers have been doing this a long time. Ask any serious gamer if the leave Intel Speedstep or AMD's Cool n' Quite enabled on there gaming rig.

 

[Tsepz_GP]

AnandTech's Review:
AnandTech | Samsung Galaxy Note 3 Review

"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."

It's also interesting to note that the Galaxy Note 3 appears to outperform all other Snapdragon 800 smartphones we've tested thus far. There's a couple of potential explanations here. First, the Galaxy Note 3 is using newer drivers than any of the other S800 platforms we've tested:

Note 3: 04.03.00.125.077
Padfone: 04.02.02.050.116
G2: 4.02.02.050.141"


Move along folks, nothing to see here.

 

[jpullins]

According to Slashgear HTC and LG have done the same with the G2 and the ONE. So samsung is not the only one they are just getting all the negativity while everyone else get a free pass.

Posted via Android Central App

 

[apollooff320]

So when it's in power saving mode it will let 3 of the cores go idle, but when there's an intense workload it will turn on all 4 cores?

Pardon me, but that isn't a bug, it's a feature. You save battery when you don't need it, and when you do need it, all four cores of the CPU will actually be engaged.

Seems pretty normal Linux kernel behavior to me. Which is what Android is built on top of.

If the reviewer's going to complain about anything, they should complain about using synthetic benchmarks in the first place.

not even the same thing as a PC. A PC doesn't raise frequencies high on just benchmarks. It does it with games also which the Note 3 doesn't. All it does is raises the frequency during benchmarks and nothing more.

 

[Botti]

not even the same thing as a PC. A PC doesn't raise frequencies high on just benchmarks. It does it with games also which the Note 3 doesn't. All it does is raises the frequency during benchmarks and nothing more.

How do you know what it does in the games? Have you tested or do you make such assumption because of some random named non-game program?

Posted via Note3

 

[Cary Quinn]

I know on laptops, you can have power management options to have the device run the cpu in a lower frequency range when on battery than when plugged in.

Android devices have had CPU Tuner as an option since Android 2.1 (on rooted devices). As was pointed out above, the OS has built in power management functions and most, if not all, devices run the CPU in more of a battery saving state by default, but has the option to boost the CPU to full speed if needed. Most of the time it should not be needed, even on games, to get good performance out of the device.

The best takeaway I get from the ARS article (which I am thankful for their testing, even though their conclusions seem biased against Samsung for giving the benchmark tool what it is asking for); is that even with the benchmark being treated as a generic program the Note 3 compares favorably against other devices using the same CPU.

I like the idea of having the most kickass CPU on the market right now in my new phone; but I'd much rather also have the idea that the battery life of that phone is being maintained so I can run it for days at a time without having to plug it in everywhere I go.

 

[Sooks]

I dont understand what people aren't understanding? When I run a benchmark on my PC all cores become unparked the speed hits turbo boost and max out, my graphics card spins up to it highest clock speed. It is doing exactly what it is supposed to do during a benchmark test.

 

[jrham]

I tend to agree with you however I kind of understand what some other are concerned about (the inability to ramp all four cores for other situations). Still though its a benchmark and in the pc world people are using LN2 just to reach insane speeds. What people need to realize is a benchmark mark is stressing the system to see its potential not for its everyday usability.
Sent from my SGH-T889 using AC Forums mobile app

 

[Sooks]

Exactly, and there is no way a 2.3 GHz Quad-Core with 3 GB of RAM wont keep up on day to day task.

 

[JB802]

not even the same thing as a PC. A PC doesn't raise frequencies high on just benchmarks. It does it with games also which the Note 3 doesn't. All it does is raises the frequency during benchmarks and nothing more.

It doesn't raise frequencies at all, it just makes them run at full speed, there is no over clocking, the chips are still running at 2.3 ghz, it just turns off all the power saving features. It doesn't do this with anything else because a) nothing needs the full power of the chip to run properly and b) running like this for any length if time will eat your battery

 

[Kevin OQuinn]

AnandTech's Review:
AnandTech | Samsung Galaxy Note 3 Review

"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."

It's also interesting to note that the Galaxy Note 3 appears to outperform all other Snapdragon 800 smartphones we've tested thus far. There's a couple of potential explanations here. First, the Galaxy Note 3 is using newer drivers than any of the other S800 platforms we've tested:

Note 3: 04.03.00.125.077
Padfone: 04.02.02.050.116
G2: 4.02.02.050.141"


Move along folks, nothing to see here.

There's absolutely plenty to see here. The fact that many manufacturers are doing this doesn't excuse it, either. It wasn't ok with the PC market, and it shouldn't be ok with the mobile market.

 

[Sooks]

There's absolutely plenty to see here. The fact that many manufacturers are doing this doesn't excuse it, either. It wasn't ok with the PC market, and it shouldn't be ok with the mobile market.

Why are we comparing apples to androids ?

The iphone pushes sub 720p .... the note 3 is pushing 2.5 times the pixels... no **** it is going to take a split second faster to load

 

[Kevin OQuinn]

Why are we comparing apples to androids ?

The iphone pushes sub 720p .... the note 3 is pushing 2.5 times the pixels... no **** it is going to take a split second faster to load

I wasn't aware that I compared anything to Apple. I made a statement that companies "cheating" on benchmarks should be unacceptable.

It's a little bit ironic that nVidia isn't cheating on any of them, at least according to Anandtech. Maybe they actually learned something from going through it back in the PC days.