Jump to content

Question

Posted

Samsung caught optimizing its phones to score higher on benchmarks

 

It can be tricky figuring out which mobile device to buy. Some get longer battery life than the competition, while others offer more storage, memory, or special features like hands-free operation.

But some folks like to get the model with the fastest processor

Share this post


Link to post
Share on other sites

82 answers to this question

  • 0

Posted

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...


Maybe it is just me, but I find that people who have to toss out insults to prove their point not worth listening to.

People need to relax. This is a discussion...we are all adults, at least I hope so. I wouldnt be surprised if more phone companies do things like this. Samsung is just a hot target since they are the largest and probably looked at the closest.

Share this post


Link to post
Share on other sites
  • 0

Posted

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!

 

That's not the problem here at all. People are glancing over the article and thinking this is just the equivalent of what Intel does with their TurboBoost. (to be fair, the OP didn't explain this well either)

 

The problem is that... say you had a processor that ran at 2.4GHz, with TurboBoost that could spot you up to 3.2GHz when you needed it. Fantastic, right? Now imagine if it ONLY jumped up to 3.2GHz when you were running benchmark software; hitting only up to 2.8GHz with any other demanding program.

 

That's the problem here, at least as best as I've understood it out to be. Why would they bother advertising beyond what is actually tangible? (rhetorical question, we all know the answer)

Share this post


Link to post
Share on other sites
  • 0

Posted

That's not the problem here at all. People are glancing over the article and thinking this is just the equivalent of what Intel does with their TurboBoost. (to be fair, the OP didn't explain this well either)

 

The problem is that... say you had a processor that ran at 2.4GHz, with TurboBoost that could spot you up to 3.2GHz when you needed it. Fantastic, right? Now imagine if it ONLY jumped up to 3.2GHz when you were running benchmark software; hitting only up to 2.8GHz with any other demanding program.

 

That's the problem here, at least as best as I've understood it out to be. Why would they bother advertising beyond what is actually tangible? (rhetorical question, we all know the answer)

 

Except its already been shown that real applications DO in fact hit the 533 MHz mark.

Share this post


Link to post
Share on other sites
  • 0

Posted

Sounds very similar to what AMD and Nvidia do when their drivers detect benchmarks being run. Nothing new here really, it's hardly even a story
1 person likes this

Share this post


Link to post
Share on other sites
  • 0

Posted

Except its already been shown that real applications DO in fact hit the 533 MHz mark.

 

I'm not arguing either side, I'm merely trying to explain the situation. :ermm:

Share this post


Link to post
Share on other sites
  • 0

Posted

That's not the problem here at all. People are glancing over the article and thinking this is just the equivalent of what Intel does with their TurboBoost. (to be fair, the OP didn't explain this well either)

 

The problem is that... say you had a processor that ran at 2.4GHz, with TurboBoost that could spot you up to 3.2GHz when you needed it. Fantastic, right? Now imagine if it ONLY jumped up to 3.2GHz when you were running benchmark software; hitting only up to 2.8GHz with any other demanding program.

 

That's the problem here, at least as best as I've understood it out to be. Why would they bother advertising beyond what is actually tangible? (rhetorical question, we all know the answer)

It's not entirely the same. The limiting factor with smartphones is battery life. If that wasn't a limiting factor, then Samsung would have no issues with running the CPU or GPU at full speed. It's possible but it isn't feasible.

Share this post


Link to post
Share on other sites
  • 0

Posted

Sounds very similar to what AMD and Nvidia do when their drivers detect benchmarks being run. Nothing new here really, it's hardly even a story

 

right, everybody cheats. nvidia,apple,intel,amd,sony,samsung, etc... you know for example when nvidia is pimping their latest mobile gpu,how they say its 2x the performance? it may be true that you can get 2x the performance, but what they dont tell you is,at what power envelope? in the real world, the gpu is not running full on blast,because that would just destroy the battery,therefore real world performance is not 2x,not even close.

Share this post


Link to post
Share on other sites
  • 0

Posted

It's not entirely the same. The limiting factor with smartphones is battery life. If that wasn't a limiting factor, then Samsung would have no issues with running the CPU or GPU at full speed. It's possible but it isn't feasible.

 

Except that's not the argument. Again, (and please correct me if I'm wrong) it's that the full speed is not tangible unless the program being used is benchmarking software...

Share this post


Link to post
Share on other sites
  • 0

Posted

Except that's not the argument. Again, (and please correct me if I'm wrong) it's that the full speed is not tangible unless the program being used is benchmarking software...


I believe it has been shown other apps can utilise the full speed. Making this a completely pointless, non-issue

Share this post


Link to post
Share on other sites
  • 0

Posted

Except its already been shown that real applications DO in fact hit the 533 MHz mark.

 

 

Sounds very similar to what AMD and Nvidia do when their drivers detect benchmarks being run. Nothing new here really, it's hardly even a story

Finally someone with some sense!

Share this post


Link to post
Share on other sites
  • 0

Posted

I believe it has been shown other apps can utilise the full speed. Making this a completely pointless, non-issue

 

That's what Lord Method Man said, and as I replied to him, I'm not arguing against that lol.

 

I'll just see myself out though before I'm quoted into a conversation loop again. :pinch:

Share this post


Link to post
Share on other sites
  • 0

Posted

who on earth ever benches below full performance? whats the point? all this is rubbish

Share this post


Link to post
Share on other sites
  • 0

Posted

Except that's not the argument. Again, (and please correct me if I'm wrong) it's that the full speed is not tangible unless the program being used is benchmarking software...

I wasn't arguing with you. You gave an example and I pointed out why it wasn't a fair comparison. A 2.8-3.2 GHz desktop CPU isn't limited by battery life so there's no reason for it to not run at full speed when a demanding application is running. Some smartphones use processors that are down-clocked to save battery life.

 

As you can see below, there are two brief spikes to 532 MHz with the camera app but aside from that, only certain benchmarking apps can run at the higher speeds for long periods of time. It's not a huge jump from 480 MHz (less than 10%); however, the power requirements are higher. Also, it's mentioned in the article that "power scales quadratically with voltage" so even a 9.77% increase like that wouldn't be feasible. It isn't necessarily cheating because the speeds are definitely attainable. If anything, it's just slightly misleading but it's not something the average consumer would complain about. And it certainly isn't something a smartphone reviewer should be overly concerned about. After all, it makes no sense to benchmark a CPU or GPU below its maximum performance.

 

 

GPUclk_575px.png

Share this post


Link to post
Share on other sites
  • 0

Posted

I wasn't arguing with you. You gave an example and I pointed out why it wasn't a fair comparison. A 2.8-3.2 GHz desktop CPU isn't limited by battery life so there's no reason for it to not run at full speed when a demanding application is running. Some smartphones use processors that are down-clocked to save battery life.

 

As you can see below, there are two brief spikes to 532 MHz with the camera app but aside from that, only certain benchmarking apps can run at the higher speeds for long periods of time. It's not a huge jump from 480 MHz (less than 10%); however, the power requirements are higher. Also, it's mentioned in the article that "power scales quadratically with voltage" so even a 9.77% increase like that wouldn't be feasible. It isn't necessarily cheating because the speeds are definitely attainable. If anything, it's just slightly misleading but it's not something the average consumer would complain about. And it certainly isn't something a smartphone reviewer should be overly concerned about. After all, it makes no sense to benchmark a CPU or GPU below its maximum performance.

 

 

GPUclk_575px.png

 

Precisely! They never advertise "Mhz" in their ads!

Share this post


Link to post
Share on other sites
  • 0

Posted

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

Oh god, I remember that. The team filming a woman cycling, from a van.

Share this post


Link to post
Share on other sites
  • 0

Posted

I wasn't arguing with you. You gave an example and I pointed out why it wasn't a fair comparison. A 2.8-3.2 GHz desktop CPU isn't limited by battery life so there's no reason for it to not run at full speed when a demanding application is running. Some smartphones use processors that are down-clocked to save battery life.

 

As you can see below, there are two brief spikes to 532 MHz with the camera app but aside from that, only certain benchmarking apps can run at the higher speeds for long periods of time. It's not a huge jump from 480 MHz (less than 10%); however, the power requirements are higher. Also, it's mentioned in the article that "power scales quadratically with voltage" so even a 9.77% increase like that wouldn't be feasible. It isn't necessarily cheating because the speeds are definitely attainable. If anything, it's just slightly misleading but it's not something the average consumer would complain about. And it certainly isn't something a smartphone reviewer should be overly concerned about. After all, it makes no sense to benchmark a CPU or GPU below its maximum performance.

 

Honestly, I wasn't thinking with a desktop in mind, but rather with laptops, particularly the Macbook Pro:

 

13" Base Model:

  • 2.5GHz dual-core Intel Core i5
  • Turbo Boost up to 3.1GHz

My point was that if you could only hit that 3.1GHz on benchmarks, otherwise only hitting 2.8GHz or some equivalent of that, would it be fair to say Apple's advertisements aren't exactly right? :huh:

 

Obviously, you don't want to be pulling that much power all the time. That wasn't my point at all. I'm simply trying to establish the points being driven from the articles, that's all.

Share this post


Link to post
Share on other sites
  • 0

Posted

http://www.anandtech.com/show/7187/looking-at-cpugpu-benchmark-optimizations-galaxy-s-4

 

not saying this is accurate or not, but if its as described, it is kinda cheap.

Sorry, I thought you said you were done here after the personal attack of calling me a "dumbass" after I pointed out that you do have a devotion and a trend towards pro-apple, anti-android posts.

Share this post


Link to post
Share on other sites
  • 0

Posted

Sorry, I thought you said you were done here after the personal attack of calling me a "dumbass" after I pointed out that you do have a devotion and a trend towards pro-apple, anti-android posts.

on dozens of occasions you have said the same thing over and over and over.... and i have stated the same thing over and over and over and you just don't seem to get it. Your opinion of my favoritism towards IOS is indeed flawed and maybe now you can see that it CLEARLY is not true. It was wrong of me to say dumbass, I AM SORRY. Honestly, I am. Somewhat gets annoying (not from you) but from people that just assume since you don't worship the SAMSUNG god that you MUST be an APPLE zeolot which is immature and annoying. once again, please accept my apologies, i was wrong.

Share this post


Link to post
Share on other sites
  • 0

Posted

Fair comments, apology accepted.  However, the threads I see you post in, tend towards a trend of anti-android / pro-apple and quite often with nonsense.  This thread is pure nonsense that YOU are misunderstanding - not others.  Apps DO run at full speed, as demonstrated - thus the phone is only reacting to the demands placed upon it by battery conservation.

 

If batteries were not an issue, the phone would run at 100% all the time.  But batteries are an issue, so they scale it down.  If something wants access to the full power of the processor, it is granted.  The benchmarking app is one such demand.  Basically, it would be false benchmarking to judge it at anything under 100%.

1 person likes this

Share this post


Link to post
Share on other sites
  • 0

Posted

Fair comments, apology accepted.  However, the threads I see you post in, tend towards a trend of anti-android / pro-apple and quite often with nonsense.  This thread is pure nonsense that YOU are misunderstanding - not others.  Apps DO run at full speed, as demonstrated - thus the phone is only reacting to the demands placed upon it by battery conservation.

 

If batteries were not an issue, the phone would run at 100% all the time.  But batteries are an issue, so they scale it down.  If something wants access to the full power of the processor, it is granted.  The benchmarking app is one such demand.  Basically, it would be false benchmarking to judge it at anything under 100%.

Couldn't have said it better.

Share this post


Link to post
Share on other sites
  • 0

Posted

If batteries were not an issue, the phone would run at 100% all the time.  But batteries are an issue, so they scale it down.  If something wants access to the full power of the processor, it is granted.  The benchmarking app is one such demand.  Basically, it would be false benchmarking to judge it at anything under 100%.

according to math, you are wrong, all apps are given only a maximum of 90.2 % with the exception of only benchmarks apps.

Share this post


Link to post
Share on other sites
  • 0

Posted

according to math, you are wrong, all apps are given only a maximum of 90.2 % with the exception of only benchmarks apps.

Evidence posted above says otherwise. Oh and welcome :)

Share this post


Link to post
Share on other sites
  • 0

Posted

its what i belive, sorry if it hurts your feelings. I don't know if i am inturpretting it incorrectly or if others are. they way i am reading it is fabricated results. I am sure you also believe what you believe too.

it's not a matter of what I "believe", its just a simple matter of it being an industry standard of phones down-clocking to save on battery. There is no need to be rude. :/  Just saying. I think you may have fallen into the trap of this article's author trying to pull a fast one on the S4's reputation.... I can see the temptation if someone dislikes Samsung. But this is a generally known feature of smart phone hardware 

1 person likes this

Share this post


Link to post
Share on other sites
  • 0

Posted

it's not a matter of what I "believe", its just a simple matter of it being an industry standard of phones down-clocking to save on battery. There is no need to be rude. :/  Just saying. I think you may have fallen into the trap of this article's author trying to pull a fast one on the S4's reputation.... I can see the temptation if someone dislikes Samsung. But this is a generally known feature of smart phone hardware 

like I said before, he's just arguing for the sake of arguing. smurfs are purple you know...

1 person likes this

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!


Register a new account

Sign in

Already have an account? Sign in here.


Sign In Now
Sign in to follow this  
Followers 0

  • Recently Browsing   0 members

    No registered users viewing this page.