Samsung caught cheating on phone specs


Recommended Posts

from that same article:

 

 

Isn't this normal behavior (for the camera?) eg. throttle up when needed, throttle down when not?

 

Yes, that is normal behavior. What isn't normal is having a list that calls out specific benchmark apps and tweaks CPU/GPU speeds when those apps are launched.

 

The CPU/GPU clocks should vary like your graph depending on what the benchmark app is trying to do (ie waiting at the menu or doing a benchmark), and peak performance given to the benchmark app should be the same as peak performance given to games and other third party apps.

Link to comment
Share on other sites

Benchmarks have never been a great indicator of real-world performance, so who really cares?

 

The same applies to a PC.  You can overclock it to 5GHz, which will give you a longer e-peen benchmark score, but you won't see that extra performance anywhere else.

 

As long as a phone is stable and doesn't lag, then specs and benchmarks mean absolutely nothing.  People should know this by now.

Link to comment
Share on other sites

Minimizing power consumption is a high priority for smartphones. What's the point of running the CPU or GPU at full power when there isn't enough power for it to last more than 3 hours? Samsung isn't cheating here. They're showing the maximum power of the hardware they use. It would be poor design for a messaging app to use the full power of a CPU and GPU. It's simply unnecessary. And aside from games, the only thing that can really push the hardware are benchmarking apps.

  • Like 2
Link to comment
Share on other sites

Minimizing power consumption is a high priority for smartphones. What's the point of running the CPU or GPU at full power when there isn't enough power for it to last more than 3 hours? Samsung isn't cheating here. They're showing the maximum power of the hardware they use. It would be poor design for a messaging app to use the full power of a CPU and GPU. It's simply unnecessary. And aside from games, the only thing that can really push the hardware are benchmarking apps.

Because your e-penis will shrink if your phone doesn't have the maximum specs to roxx0rs joor boxers while texting.

Link to comment
Share on other sites

sound like "An investigation carried out ostensibly to uncover activities but actually used to harass and undermine those with differing views."

Link to comment
Share on other sites

for those that don't grasp the concept of "stretching the numbers", imagine if a car manufacturer had a different set guidelines that it based gas mileage on... (yes they don't get the milage posted as is already) but image if you a manufacturer took about all the interior and spare tire and all the weight it could, say lessened the car 700 lbs... then published those numbers for MPG  as the their benchmark number. Did it actually get the posted "MPG"? yes. Can it be done under any other circumstance? No. 

 

Why I think this is skummy is because it will force all other handset makers to start doing the same thing and then this may lead CPU number game that means almost nothing, misleading and wrong.

Link to comment
Share on other sites

for those that don't grasp the concept of "stretching the numbers", imagine if a car manufacturer had a different set guidelines that it based gas mileage on... (yes they don't get the milage posted as is already) but image if you a manufacturer took about all the interior and spare tire and all the weight it could, say lessened the car 700 lbs... then published those numbers for MPG  as the their benchmark number. Did it actually get the posted "MPG"? yes. Can it be done under any other circumstance? No. 

 

Why I think this is skummy is because it will force all other handset makers to start doing the same thing and then this may lead CPU number game that means almost nothing, misleading and wrong.

Well I guess if everybody does the same thing, then the comparisons will be on an even playing ground.

Link to comment
Share on other sites

samsung already denied and explained this:

 

http://www.techradar.com/news/phone-and-communications/mobile-phones/samsung-denies-rigging-galaxy-s4-benchmarking-results-1170026

 

 

Which is pretty normal behavior in processors if you ask me, downclock when not in use, upclock when juice is required, how is that cheating?

 

because "downclock when not in use" usually means the CPU will ramp up the freq whenever it ... "senses" its being demanded. but instead of that, it increases whenever a few select benchmark apps are ran. in fact, apparently on the exynos octa, all cores stay on at max freq all the time when running the apps, regardless of actual load.

 

stuff like this has probably happened before on GPU benchmarks, I'm assuming.

Link to comment
Share on other sites

for those that don't grasp the concept of "stretching the numbers", imagine if a car manufacturer had a different set guidelines that it based gas mileage on... (yes they don't get the milage posted as is already) but image if you a manufacturer took about all the interior and spare tire and all the weight it could, say lessened the car 700 lbs... then published those numbers for MPG  as the their benchmark number. Did it actually get the posted "MPG"? yes. Can it be done under any other circumstance? No.

 

Oh I grasp the concept, so stop trying to patronise - you fail badly!

 

So, you look at this:  Speaker wattage is ALWAYS shown at peak.  You CANNOT run a speaker at peak for more than a second or so - but it's the accepted norm.  People want to know how far it can be pushed.

 

Also, car manufacturers do test under "ideal" conditions.  Your comparison is also false - it would be akin to unloading all core system services to just focus on the one task - and this is not happening.

 

Keep clutching at straws.  Your iDevice loyalty is so apparent it's just a joke.

  • Like 1
Link to comment
Share on other sites

for those that don't grasp the concept of "stretching the numbers", imagine if a car manufacturer had a different set guidelines that it based gas mileage on... (yes they don't get the milage posted as is already) but image if you a manufacturer took about all the interior and spare tire and all the weight it could, say lessened the car 700 lbs... then published those numbers for MPG  as the their benchmark number. Did it actually get the posted "MPG"? yes. Can it be done under any other circumstance? No. 

 

Why I think this is skummy is because it will force all other handset makers to start doing the same thing and then this may lead CPU number game that means almost nothing, misleading and wrong.

The flaw with that analogy is that there's no such thing as "unnecessary fuel economy". The higher, the better. And the results are reviewed by the government so it's not like manufacturers can fake them and hope that customers get fooled into buying their cars.

 

As for smartphones, it's entirely possible for manufacturers to run their CPUs and GPUs at full power. Of course, that would result in really poor battery life. The fact that Samsung's hardware ran at the speeds mentioned in the article is proof that it's capable of running at that speed. However, Samsung chose to clock things down in the interest of saving battery life. Also, there's a certain range (as well as a maximum speed) that the CPU/GPU operates at. Obviously, most apps don't need such high performance. They run well at lower clocks. It's only when games are introduced that things kick into high gear. Even then, battery life is still a priority. It isn't a priority when the hardware is being benchmarked though.

 

I fail to see how that's "scummy". Would you prefer it if Samsung allowed their CPUs and GPUs to run at full power all the time, regardless of the app? I'm sure a lot of customers would complain about poor battery life.

  • Like 1
Link to comment
Share on other sites

Also, car manufacturers do test under "ideal" conditions.  Your comparison is also false - it would be akin to unloading all core system services to just focus on the one task - and this is not happening.

 

Keep clutching at straws.  Your iDevice loyalty is so apparent it's just a joke.

from what i am reading, more then just the artical posted, this is indeed what they are doing...

and about the idevice loyalty? you sir are one blind dumbass... read a bit more deeply, I won't even give you hint... just try it.... (I own a Samsung Galaxy Note 2, and a Nexus 4 and a Nexus 7, switched my nephew from a iphone to a nexus 4 and switched both my parents from iphones to nexus 4's, AND just last 2 weeks ago, got my daughter a s2x to replace her iphone)

it would seem to me, despite the facts, the OP is just arguing for the sake of doing so :p

some people just can't see things the way they were meant to be seen is all. Its not your fault, just your ability.

Link to comment
Share on other sites

 

some people just can't see things the way they were meant to be seen is all. Its not your fault, just your ability.

Seriously? Almost everyone who has posted a reply has given you facts and examples, yet it is me that lacks the ability?  Ok :D

  • Like 1
Link to comment
Share on other sites

Seriously? Almost everyone who has posted a reply has given you facts and examples, yet it is me that lacks the ability?  Ok :D

the ones that did, missed the concept too.

Link to comment
Share on other sites

for those that don't grasp the concept of "stretching the numbers", imagine if a car manufacturer had a different set guidelines that it based gas mileage on... (yes they don't get the milage posted as is already) but image if you a manufacturer took about all the interior and spare tire and all the weight it could, say lessened the car 700 lbs... then published those numbers for MPG  as the their benchmark number. Did it actually get the posted "MPG"? yes. Can it be done under any other circumstance? No. 

 

Why I think this is skummy is because it will force all other handset makers to start doing the same thing and then this may lead CPU number game that means almost nothing, misleading and wrong.

Car mileage is always rated at certain speeds and highway conditions. Your analogy goes against your own argument.

  • Like 2
Link to comment
Share on other sites

I am shocked how some of you just can not see what the article is saying. I give up. Some just can't see what is being done. Oh well, I'm done.

Link to comment
Share on other sites

I am shocked how some of you just can not see what the article is saying. I give up. Some just can't see what is being done. Oh well, I'm done.

We did read, we are trying to point out to you how that same article is being sensationalist, as the phones operation won't probably be affected in average day to day use.

 

Wait... so my GPU clocks up ( es per the driver) when I play dead space, so that's cheating? Ok.  What anandtech MIGHT have a point is that this should be user toggled, however, how many average joes care about such an issue?

  • Like 1
Link to comment
Share on other sites

I can see the discussion here.  The bigger question is how is the benchmark being applied.

 

The system maxed out can do this.  No application other than the Benchmark can do it.  vs.  The benchmark being used as the Indicator as to what Real World is.

 

Remember it looks like ONLY benchmark Applications can hit this.  Not just full screened apps.

 

I personally would rather have real world benchmarks. :)

Link to comment
Share on other sites

from that same article:

 

GPUclk_575px.png

 

Isn't this normal behavior (for the camera?) eg. throttle up when needed, throttle down when not?

 

 

I can see the discussion here.  The bigger question is how is the benchmark being applied.

 

The system maxed out can do this.  No application other than the Benchmark can do it.  vs.  The benchmark being used as the Indicator as to what Real World is.

 

Remember it looks like ONLY benchmark Applications can hit this.  Not just full screened apps.

 

I personally would rather have real world benchmarks. :)

see my above self-quote, anandtech found that some applications do hit the 533, which IMO is normal! Why would the phone keep @ full load just to waste energy if  it can handle the load without doing so and NOT sacrificing usability ?

  • Like 1
Link to comment
Share on other sites

from that same article:

 

GPUclk_575px.png

 

Isn't this normal behavior (for the camera?) eg. throttle up when needed, throttle down when not?

 

It very much is.  It looks like only the camera app can do that at this point in time.  So we have the Benchmark app and the Camera App(in spurts).  This is neither good nor bad, it simply is.  Now, as long as all of that is known while ranking the "Fastest Cell Phone" and write up and comparison between phones,  I see no real problem with it. 

  • Like 1
Link to comment
Share on other sites

We did read, we are trying to point out to you how that same article is being sensationalist, as the phones operation won't probably be affected in average day to day use.

 

Wait... so my GPU clocks up ( es per the driver) when I play dead space, so that's cheating? Ok.  What anandtech MIGHT have a point is that this should be user toggled, however, how many average joes care about such an issue?

 

The issue here is that the GPU DOESN'T clock up when you play games or do anythingm the max it clocks up to is 480Mhz, the only time it clocks up to 530Mhz, is when it detects a benchmark app. 

 

Basically whenever the phone detects benchmarking it overclocks, because it can't sustain that clock speed during normal operations. and all the reviewers will use the benchmark numbers, that they are not aware are fake to show how the phone is over 10% faster than the competition. 

  • Like 1
Link to comment
Share on other sites

Not shocking that it occurs, however something definitely to keep in mind next time benchmarks are being posted comparing model x to model y....

  • Like 1
Link to comment
Share on other sites

I thought they all did it? Nokia was caught out promoting their stabilizer while filming on the move too :p

Nokis problem was that the 920 was just as good as the dslrs they used .

Link to comment
Share on other sites

LOL at the people who think the CPU and GPU should be going 100% full speed while reading email or something...

 

i want the thing blasting fast when i'm doing something that needs it!

Link to comment
Share on other sites

This topic is now closed to further replies.
  • Recently Browsing   0 members

    • No registered users viewing this page.