Attend XDA's Second Annual Developer Conference, XDA:DevCon 2014!
5,808,261 Members 37,936 Now Online
XDA Developers Android and Mobile Development Forum

[Q] Too low scores from benchmarks

Tip us?
 
Lethargy
Old
#11  
Lethargy's Avatar
Senior Member
Thanks Meter 1,621
Posts: 2,917
Join Date: Jan 2014
Location: Sydney, Australia
Quote:
Originally Posted by msnnsm View Post
I'm pretty sure i'm starting it when it's cold. I didn't touch it for 15 minutes isn't it enough ?
At least use an app to look at CPU temp ._.
Quote:
Originally Posted by kyphur
You ask us to understand your position. Well if you want to benefit from our experience and time then I think it is only fair that you understand our position.
I actively add ignorant people to my ignore list.
 
msnnsm
Old
#12  
msnnsm's Avatar
Senior Member - OP
Thanks Meter 333
Posts: 280
Join Date: Jun 2012
Location: Konya
Quote:
Originally Posted by Lethargy View Post
At least use an app to look at CPU temp ._.
Hmm you're right. Do you know a good one ? I got 28.8k now with elemental x kernel with "Undervolt" feature. To be honest i don't know how much is normal for a nexus 5 device with this kernel.
If you like my work, press thanks
 
eddiehk6
Old
#13  
eddiehk6's Avatar
Senior Member
Thanks Meter 582
Posts: 1,373
Join Date: Dec 2010
Location: London
Quote:
Originally Posted by msnnsm View Post
I don't care about score
Yes do you, that's why you opened the thread..

The post sounds aggressive, but it's really trying to change the way you think about scores in the first place..

Quote:
But others phones get higher and there is something wrong with mine.
No, just because other people get highers 'scores' does not mean there is something wrong with your phone.

Before you started running 'tests', were you experiencing any actual issues?

Quote:
It's getting hot more than others. So that's not like it should be.I hope you understood why i'm trying to fix it and sorry if my english isn't understandable.
If it's getting hot, is this because you're running unnecessary benchmarks...

Or any other activities which are causing it to get hot? This could be an app or hardware issue, both nothing to do with your benchmark. Are you comparing actual temperatures in degrees to identical phones doing exactly the same activities side by side, or does it just feel a bit warmer?

What I'm trying to say is your obsession with the score is causing you unnecessary stress...you are trying to fix something which does not need fixing.

So go back to the original thinking:
- Imagine scores don't exist
- Did the phone actually feel slow in day to day normal usage before you started running benchmarks?

Or:
- Stop running tests, forget about them
- Use the phone normally
- Enjoy using the phone again

Let...it...go...
Device: Nexus 5 | ROM: SlimKat | Recovery: PhilZ Touch
Retired: HTC Desire
_____________________________________________
let's all try to get along ...but remember to search first...
 
msnnsm
Old
#14  
msnnsm's Avatar
Senior Member - OP
Thanks Meter 333
Posts: 280
Join Date: Jun 2012
Location: Konya
Quote:
Originally Posted by eddiehk6 View Post
Yes do you, that's why you opened the thread..

The post sounds aggressive, but it's really trying to change the way you think about scores in the first place..



No, just because other people get highers 'scores' does not mean there is something wrong with your phone.

Before you started running 'tests', were you experiencing any actual issues?



If it's getting hot, is this because you're running unnecessary benchmarks...

Or any other activities which are causing it to get hot? This could be an app or hardware issue, both nothing to do with your benchmark. Are you comparing actual temperatures in degrees to identical phones doing exactly the same activities side by side, or does it just feel a bit warmer?

What I'm trying to say is your obsession with the score is causing you unnecessary stress...you are trying to fix something which does not need fixing.

So go back to the original thinking:
- Imagine scores don't exist
- Did the phone actually feel slow in day to day normal usage before you started running benchmarks?

Or:
- Stop running tests, forget about them
- Use the phone normally
- Enjoy using the phone again

Let...it...go...
The first time run it i just wanted to see how my phone is powerful because i felt a little lag on asphalt 8 yesterday, so yes i wasn't expecting any issues but when i saw i got 17k i think that difference is just huge. I don't care if my device is same at gaming,multitasking with other nexus devices but benchmark makes me think it's not right.

In normal usage; yeah it is fast But I'm not sure my device is how it's supposed to be like this. But if you say it's normal and benchmark's not meaning anything i won't care. But there should be a reason all devices got different scores.
If you like my work, press thanks
 
eddiehk6
Old
#15  
eddiehk6's Avatar
Senior Member
Thanks Meter 582
Posts: 1,373
Join Date: Dec 2010
Location: London
Quote:
Originally Posted by msnnsm View Post
The first time run it i just wanted to see how my phone is powerful because i felt a little lag on asphalt 8 yesterday, so yes i wasn't expecting any issues but when i saw i got 17k i think that difference is just huge. I don't care if my device is same at gaming,multitasking with other nexus devices but benchmark makes me think it's not right.
So it is sort of only the benchmark which is making you think something is not right

You don't need a score to see how powerful your phone is. It will breeze through most tasks just fine. "A little lag" on a graphically intense game is to be expected sometimes.

Gaming -> High CPU load -> heat -> thermal throttle -> bit of lag

Laws of physics

Quote:
In normal usage; yeah it is fast
This is the only real 'benchmark' which is actually important in normal day to day usage.

Quote:
But I'm not sure my device is how it's supposed to be like this. But if you say it's normal and benchmark's not meaning anything i won't care. But there should be a reason all devices got different scores.
Yes there is a reason we all get different scores, the same reason we all get different battery life. There are many variables, not least temperature and thermal throttling.

We have different number of apps installed and running at any one time, things syncing, mods, scripts, ROMs, kernels, all different. No two CPUs and Nexus 5s are exactly identical either. So therefore you get different 'scores'. Even if you had two stock factory reset phones side by side running the same test, they'll get different scores.

And just because your score is lower, it doesnt necessarily mean that something is wrong, or that your phone is always slow. It is just a nominal number from an app at a particular point in time. It still feels fast, and is fast...
Device: Nexus 5 | ROM: SlimKat | Recovery: PhilZ Touch
Retired: HTC Desire
_____________________________________________
let's all try to get along ...but remember to search first...
The Following User Says Thank You to eddiehk6 For This Useful Post: [ Click to Expand ]
 
msnnsm
Old
#16  
msnnsm's Avatar
Senior Member - OP
Thanks Meter 333
Posts: 280
Join Date: Jun 2012
Location: Konya
Quote:
Originally Posted by eddiehk6 View Post
So it is sort of only the benchmark which is making you think something is not right

Even if you had two stock factory reset phones side by side running the same test, they'll get different scores.
It doesn't make any sense
Quote:

And just because your score is lower, it doesnt necessarily mean that something is wrong, or that your phone is always slow. It is just a nominal number from an app at a particular point in time. It still feels fast, and is fast...
Okey, i trust you. Since i just bought phone and didn't test it much really i won't test it or try to improve it for now.
If you like my work, press thanks
 
bblzd
Old
#17  
Senior Member
Thanks Meter 294
Posts: 1,315
Join Date: Nov 2013
Location: Toronto
The way a phone works doesn't lend itself well to benchmarks. That's why manufacturers like Samsung have traditionally "cheated" on benchmarks by ensuring maximum clock speeds are used when they detect a bench marking application.

If you want to maximize your benchmark scores you simply need to disable thermal throttling (or increase its limit) and also use optimized Dalvik copied from another device.
W32
 
TheLastSidekick
Old
#18  
TheLastSidekick's Avatar
Recognized Contributor
Thanks Meter 2,063
Posts: 3,141
Join Date: Aug 2012
Location: Springfield, Massachusetts

 
DONATE TO ME
Quote:
Originally Posted by rootSU View Post
The other thing is to not care about benchmarks...
This. Thermal throttle is a way to keep your phone regulated into normal temperatures specified by the device manufacturer. However, it's generally safe to raise it a bit, mine is set to 85.
Team Inferno (Co-admin)
 
simms22
Old
#19  
simms22's Avatar
Recognized Contributor
Thanks Meter 21,428
Posts: 25,784
Join Date: Jun 2009
Location: BROOKLYN!
aw man, its too bad i wasnt around earlier
Google+ Profile
Twitter Simms22
DerKernel For Trinity kernels and more!

The Following User Says Thank You to simms22 For This Useful Post: [ Click to Expand ]
Thread Tools Search this Thread
Search this Thread:

Advanced Search
Display Modes