bow to the new king!!! benchmarking inside note 3 takes all octo core can use all 8

[ReallyBigFeet]

I'm doubtful of this as the GP edition of the GS4 and touchwiz version have pretty much the same benchmarks and perform almost identical.

Disassembly of the Samsung firmware actually shows the coding tags Samsung put in place to detect benchmarking apps and crank up the CPU accordingly just for those. Not sure if this is at all impacted by Touchwiz....if I were coding it, I'd put it at the firmware level.

 

[Peterg2]

Samsung stoops to new lows with the benchmark cheating.

It is safe to say all results from the Note 3 can be disregarded.

arstechnica.com/gadgets/2013/10/galaxy-note-3s-benchmarking-adjustments-inflate-scores-by-up-to-20/

It is not just a Samsung issue. Others, including HTC, have been "gaming" the benchmarks.

I quote:

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

 

[charlesdayton]

https://www.macrumors.com/2013/10/0...rking-scores-phil-schiller-calls-shenanigans/

 

[user-name-here]

But the question remains: why would Samsung deliberately take the time to spoof their code specifically to manipulate benchmarking scores? Is it possible that they realize how many consumers are gullible enough to believe that the "specs matter the most" message that they intentionally use this instead of bettering their products?

Why would you invest the time to do this unless you were being specifically manipulative in some way?

Because spec geeks are potential customers and, by nature of them looking at numbers instead of experience, can be quite gullible.

 

[F123D]

True. Like how some members on this forum were boasting that the new 5s was 40x faster than the previous iPhone.

 

[adder7712]

Disassembly of the Samsung firmware actually shows the coding tags Samsung put in place to detect benchmarking apps and crank up the CPU accordingly just for those. Not sure if this is at all impacted by Touchwiz....if I were coding it, I'd put it at the firmware level.

Games could benefit a lot from the boosted CPU mode.

 

[ReallyBigFeet]

Games could benefit a lot from the boosted CPU mode.

If the device makers would give us a "turbo" switch to turn on/off per app, I'd agree. But rather than focus this into a useful feature, they only chose to focus on benchmark spoofing. That's the question that still remains...why bother?

Anandtech has a good article on this front page today. Virtually all the major players EXCEPT Motorola and Apple appear to be doing this.

http://www.anandtech.com/show/7384/state-of-cheating-in-android-benchmarks

I guess we can pretty much throw out the "better specs" argument entirely at this point as almost all the ones are involved in spoofing the tests.

 

[Assault]

If the device makers would give us a "turbo" switch to turn on/off per app, I'd agree. But rather than focus this into a useful feature, they only chose to focus on benchmark spoofing. That's the question that still remains...why bother?

Anandtech has a good article on this front page today. Virtually all the major players EXCEPT Motorola and Apple appear to be doing this.

http://www.anandtech.com/show/7384/state-of-cheating-in-android-benchmarks

I guess we can pretty much throw out the "better specs" argument entirely at this point as almost all the ones are involved in spoofing the tests.

Does this really matter? I have yet to see any data that shows any phone "overclocking" to generate inaccurate data, but rather ramping up to max cpu freqs during these whitelisted benchmarks. Ironically, even Anandtech notes that other benchmark apps that are not explicitly mentioned in the code were behaving the exact same way as whitelisted apps.

So, is this called spoofing numbers or just giving max performance numbers? Because when playing a game like Dead Trigger, wouldn't you want to know what max numbers are for CPU,GPU and FPS, to let the user know the game will run smoothly? Who gives a crap about normal everyday benchmark numbers when you are using powerpoint, web surfing or reading an ebook?

Or does this all boil down to "my phone is better than yours, and if it isn't it is because you cheated on benchmark numbers"? This is just plain silly.

P.S. I'm not directing this at you 'reallybigfeet'. It's more of a rhetorical post to comment on this discussion as a whole. Your quote just happened to fit best with my comments.

 

[cynics]

I don't know if I'd call that cheating or not. Is there a way or has someone figured out a way to disable its ability to prime itself then run the same benchmark?

 

[Peterg2]

I don't know if I'd call that cheating or not. Is there a way or has someone figured out a way to disable its ability to prime itself then run the same benchmark?

One benchmark has been revised to prevent this:

https://play.google.com/store/apps/details?id=com.pctvtv.android.ttsx