Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.

The benchmarks were essentially done single threaded. But when you run a multi-threaded test where the cores matter and where Android shines, the A7 gets crushed. I don't understand why Anand did not do more multi0-threaded tests. Also some of his tests are questionable and has been called out on it.

58200.png


Reply from Anand about his testing being flawed:
Anand Lal Shimpi - Wednesday, September 18, 2013 - link
Ultimately the benchmarking problem is being fixed, just not on the time scale that we want it to. I figured we'd be better off by now, and in many ways we are (WebXPRT, Browsermark are both steps in the right direction, we have more native tools under Android now) but part of the problem is there was a long period of uncertainty around what OSes would prevail. Now that question is finally being answered and we're seeing some real investment in benchmarks. Trust me, I tried to do a lot behind the scenes over the past 4 years (some of which Brian and I did recently) but this stuff takes time. I remember going through this in the early days of the PC industry too though, I know how it all ends - it'll just take a little time to get there.
 
Last edited:
The benchmarks were essentially done single threaded. But when you run a multi-threaded test where the cores matter and where Android shines, the A7 gets crushed. I don't understand why Anand did not do more multi0-threaded tests. Also some of his tests are questionable and has been called out on it.

Image

Not all of the benchmarks were done single-threaded.
 
The benchmarks were essentially done single threaded. But when you run a multi-threaded test where the cores matter and where Android shines, the A7 gets crushed. I don't understand why Anand did not do more multi0-threaded tests. Also some of his tests are questionable and has been called out on it.

Lets compare it to real actual everyday performance tests from gurus like anandtech.

Funny how these exact same tests were more than fine when the galaxy 4 came out.

58166.png


58181.png


58180.png


58179.png
 
Last edited:
Lets compare it to real actual everyday performance tests from gurus like anandtech.

Funny how these exact same tests when the GS4 came out were more than fine.

Do you understand these where browser benchmarks?? That these were not hardware benchmarks? The one I posted was and the A7 lost. And that's from Anand himself.

That's why Anand got called out. he is using JS browser engine benchmarks and that is an "Apples to Oranges" test.

His quote in the 5S review comments in response to being called out sums it. That he's working on it.

You can't effectively compare different platforms running different JS engines and different browsers and call it a day. That's wrong.
 
Do you understand these where browser benchmarks?? That these were not hardware benchmarks? The one I posted was and the A7 lost.

That's why Anand got called out. he is using JS browser engine benchmarks and that is an "Apples to Oranges" test.

His quote in the 5S comments in response to being called up sums it up. That he's working on.

yeah and if anand didn't take the benchmark then who did? Not like anyone has the 5S yet.....something tells me that benchmark is BS. Multithreaded or not the 5S would beat the 5 and 5c.
 
You can't effectively compare different platforms running different JS engines and different browsers and call it a day. That's wrong.

From the perspective of user experience, of course you can. It doesn't matter one whit what the underlying hardware/software/crap is, if the user experiences better performance on their end, it's a win.
 
Do you understand these where browser benchmarks?? That these were not hardware benchmarks? The one I posted was and the A7 lost. And that's from Anand himself.

That's why Anand got called out. he is using JS browser engine benchmarks and that is an "Apples to Oranges" test.

His quote in the 5S comments in response to being called up sums it up. That he's working on it.

You can't effectively compare different platforms running different JS engines and different browsers and call it a day. That's wrong.

Please. Nobody but fandroids cares about pure hardware specs. It's just laughable when fandroids cannot take being beaten at your own game.

Fandroids need to get over wasting their time posting in every apple related thread about how much they hate apple and how the 500th android handset released this week is better.

I guess you have plenty of time, while you wait for your manufacturer/handset to be finally compatible with x version of android. Not the latest version of course, that will probably come in 6 to 12 months.

I ran a sunspider test on my iphone 5 with ios 7 and this 1 year old phone is faster than the latest (week old) Android with 100 cores.

Your use of a physics test, as if we are going to run autocad or mathlab on our phones, is laughable by any stretch.
 
Last edited:
Do you understand these where browser benchmarks?? That these were not hardware benchmarks? The one I posted was and the A7 lost. And that's from Anand himself.

That's why Anand got called out. he is using JS browser engine benchmarks and that is an "Apples to Oranges" test.

His quote in the 5S review comments in response to being called out sums it. That he's working on it.

You can't effectively compare different platforms running different JS engines and different browsers and call it a day. That's wrong.

Is the Apple the orange?
 
yeah and if anand didn't take the benchmark then who did? Not like anyone has the 5S yet.....something tells me that benchmark is BS. Multithreaded or not the 5S would beat the 5 and 5c.

That benchmark I posted was from Anand. Look at the link.

He himself admits that the test is not perfect and he's working on it. But people need to realize what the test actually measure.

----------

Please. Nobody but fandroids cares about pure hardware specs. It's just laughable when fandroids cannot take being beaten at your own game.

Fandroid? Hardly. I have a MBP 17, MBA, 2 iPads, still have my iPhone 3G, 3Gs, and 4. I work in an all mac shop. So hardly a "Fandroid".

I only responded about the benchmarks and how it is flawed, period.

I could care less if it's the best chip ever made. But you are laughable talking about your sunspider scores when that is exactly what I meant about flawed.
 
That benchmark I posted was from Anand. Look at the link.

He himself admits that the test is not perfect and he's working on it. But people need to realize what the test actually measure.

your right my fault i read the whole review tuesday but didn't remember that one. but you picked one of the only tests that the 5S didn't do well on (this and triangles) and I am betting there is a bug somewhere.
 
Who gives a crap! Just use the phone that you want for your specific needs and not what you think everyone else think is cool because it has certain hardware specs. Seems like we have a bunch of impressionable teenagers who are trying too hard to impress everyone with what they have.:rolleyes:
 
Please. Nobody but fandroids cares about pure hardware specs. It's just laughable when fandroids cannot take being beaten at your own game.

Fandroids need to get over wasting their time posting in every apple related thread about how much they hate apple and how the 500th android handset released this week is better.

I guess you have plenty of time, while you wait for your manufacturer/handset to be finally compatible with x version of android. Not the latest version of course, that will probably come in 6 to 12 months.

I ran a sunspider test on my iphone 5 with ios 7 and this 1 year old phone is faster than the latest (week old) Android with 100 cores.

Your use of a physics test, as if we are going to run autocad or mathlab on our phones, is laughable by any stretch.

Sometimes I'm embarrassed to prefer certain products (Apple ) by those that represent us so loudly....
 
Please. Nobody but fandroids cares about pure hardware specs. It's just laughable when fandroids cannot take being beaten at your own game.

Fandroids need to get over wasting their time posting in every apple related thread about how much they hate apple and how the 500th android handset released this week is better.

I guess you have plenty of time, while you wait for your manufacturer/handset to be finally compatible with x version of android. Not the latest version of course, that will probably come in 6 to 12 months.

I ran a sunspider test on my iphone 5 with ios 7 and this 1 year old phone is faster than the latest (week old) Android with 100 cores.

Your use of a physics test, as if we are going to run autocad or mathlab on our phones, is laughable by any stretch.



calm down.

he is just trying to set the record straight.

you are the one who jumped the gun (and posted misleading benchmarks).


wait until the iphone 5s is out and proper testing can be done.
 
Lets just wait a few more days until proper testing can be done on iPhone 5s and A7 (and some possible bugs fixed if there are any). Then we can talk ;)

Either way all those numbers posted don't really mean a lot. iPhone 5s will beat android competition in all real world tests (browsing, opening apps, games graphics quality and frame rate etc.). iPhone 5 is as fast or faster than Android competition, let alone 5s. :)
 
Tests where Galaxy 4 does well: ABSOLUTELY VALID

Tests where 5S clearly destroys it: MUST BE DESIGNED BY FANBOIIS

Of course!
 
The benchmarks were essentially done single threaded. But when you run a multi-threaded test where the cores matter and where Android shines, the A7 gets crushed. I don't understand why Anand did not do more multi0-threaded tests. Also some of his tests are questionable and has been called out on it.

Image
A
Reply from Anand about his testing being flawed:
This is a 3D/GPU benchmark. It is not a CPU benchmark.

The iPhone 5S is the first phone on the planet with the next-generation Imagination Technologies' Rogue GPU. The drivers for this GPU are still relatively immature/unoptimized (I.e., v1.0). The iPhone 5S' 3D/GPU performance should improve over time as GPU drivers improve and Apple incorporates those improved drivers. Until the drivers are mature/optimized, there will be the occasional 3D/GPU use case where the iPhone 5S underperforms.
 
Last edited:
Fandroid? Hardly. I have a MBP 17, MBA, 2 iPads, still have my iPhone 3G, 3Gs, and 4. I work in an all mac shop. So hardly a "Fandroid".

Based on the similar way most fanatical fandroid types behave and act with these sorts of things, the assumption was valid.
 
I highly doubt the A7 is beating the s800 on the LG g2 and the note 3..... Maybe on apps and browsing where single thread is more important but when it comes to actually using all the cores, The A7 is no where near the ball park of the s800 and the octa. The Lg g2 Put down a 4100 score on geek bench ummm the iPhone 5 had maybe 1400-1700 range the most and barely beat the s3 dual core version. 64 bit doesn't make you jump 2000 points unless you have the ram to back it up too. No I am not a fanboy either I love techonolgy I had all the iPhone's including the 5 and all galaxy's and my current device is the s4 and i even had the Htc one Ill post some pictures of benchmarks from the note 3 and lg g2 n btw fastest phone in the world right now is the Note 3 ;)
 
Another question would be if you use your phone to run artificial benchmarks ...

Benchmark results are always ... benchmarks. They make hardware capabilities comparable because they perform the same task.

Drawing any kind of conclusion from the results of such tasks is nonsense.
 
A lot of apps use 2 cores and it all comes down to real life performance and day to day use... When the s4 was first announced it was the king of benchmarks but in real life use from all the stuff Samsung put on the phone it would stutter till they released a software fix. So you can have a 100 cores but if your is not optimized its still going to lag and run slow. Software and Hardware needs to work together for a smooth and fast experience. It like if you had a 1000hp car you can still lose a race if you dont know how to control it and crash lol
 
Can someone dumb this down for us non techies? I use my phone to (in order of highest occurance rate)
web browser
text
email
phone
So doing the things above, which phone does it the fastest? I and most others dont use to much software that takes alot of power. Please answer in simple terms.
 
Can someone dumb this down for us non techies? I use my phone to (in order of highest occurance rate)
web browser
text
email
phone
So doing the things above, which phone does it the fastest? I and most others dont use to much software that takes alot of power. Please answer in simple terms.


its just irrelevant.

comparing benchmarks of high-end phones is only for geeks who want to feel happy about the choice they made.

it really just translates in differences of nanoseconds between one phone and the other - especially for the tasks you list.
 
And yet can we think of anything that is on Android running quad core that uses so much processing power that the 5s cant do ?

The answer will be no.

Real world test matters not a hardware one.

Samsung throws hardware at stuff but never optimises it with software.
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.