If benchmark score doesn't matter to the end user, then why scamsung have to cheat? 

In the end no matter how you mislead the consumer, it is consider wrong.  Who knows what else is scamsung lying to its customer?

Just ask yourself would you buy from a dishonest seller in lowyat forum or any shop out there?  Why not?

It's not just a matter of firing all cores,it's the fact that the CPU only ramps up to that clockspeeed when certain benchmark apps are detected. It doesn't ramp up as high when doing other tasks,that why it's called cheating.

I don waant to correct you but you certaintly don know whats happening here . Better read the article before saying anything and calling ppl bull shit.

obviously he got . But nevermind , its not our responsibility to educate you . ignorance is bliss .

Having code to detect benchmarking software and making cpu to work extra hard to improve the score.  So the phone the actual performance under normal usage is actually lower.  So what does that mean to you?  Trying to mislead the buyer! 

What's next... battery benchmarking, shut down all cores and run for 2 weeks?  And then market to the world that the battery can last 2 weeks?

The thermal headroom is raised by pushing all cores to run at 2.3GHz as long as the benchmark app is detected but it's not raised when you use other apps so that it doesn't kill battery life or doesn't overheat making it not a representative of real world performance. Geekbench is supposed to simulate real world app performance [it gets it's scores by running apps] . This makes one expect the device to perform on a level matching it's benchmark scores when it reality it won't perform that way due to thermals,battery etc etc.

An analogy is like when you buy a pc graphics card expecting a certain level of performance after seeing benchmarks done only to find out later that the card only performs that good in benchmarks,not in games. If that's not cheating then I don't know what it is.

It reminds me of the whole Nvidia 3dmark "optimization" from years back.

The detail "report"

http://arstechnica.com/gadgets/2013/10/gal...es-by-up-to-20/

i have not finish the whole thing yet cos it not really bothers me. but what make them suspicious is as follow

it's about how the test are carried out i would say

when running the test, of course each of the phone to be tested shall be running on the same condition, then only your comparison is fair and square.

i'm not sure about the geekbench. but from my understanding, it's like it test each of the phone not only in peak performance, but also test the phone during idle time, when at this stage the CPU is suppose to shut off 3 of the 4 core. but they claim samsung alter this idle time during these benchmark test and note3 will not stay idle even if it should and thus firing all 4 cores... increasing the overall benchmark score than it should.

and they claim a phone should not behave this way as in normal day to day usage, there will be idle time when 3 of the 4 cores are shut down...

all other phones are tested this way and it is unfair when samsung tweaked note3 to behave otherwise solely for benchmark testing purpose.

this is my understanding la

haha... i dunno...

like i said, it's about how the phone are tested... they have LG G2, tested in "normal" way whereby in some apps the phone shut down 3 of 4 cores. but they claim samsung never shut down its core... so, it is not a fair comparison already and said samsung cheated cos they tweaked the note3 to behave differently specifically when running benchmark app

So u are one of them who kena con and happily say I love to be conned..😄

When the s800 2.3ghz processor is manipulated means it is running faster than 2.3ghz. The problem is consumer will think this phone is faster than others when it may be not. Essentially you are not comparing the same processor with others anymore.

Read the article..

It not only increase idle speed but also core speed during benchmark. Meaning it no longer runs at 2.3 but higher.

That's why there is a 20% increase compared to similar s800 2.3.