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

hajime

macrumors 604
Original poster
Jul 23, 2007
7,728
1,215
Hello, I went to a phone store that had both iPhone 4S and the Galaxy Note on display side by side. After playing with the Galaxy Note for a few minutes and then moved to the iPhone 4S, opening Pages, Keynote and Numbers on the iPhone seemed to take a long time. Of course, I can't compare as the Galaxy Note does not has these applications. However, I also found that I could not zoom in or out while taking video using the iPhone 4S.

Then, I went to an Apple Store. The staff told me that it is not possible to zoom in or out while taking the video using the iPhone 4S. Why? This seems to be kind of a basic function to me. How come there is no App to do that?

I also tried Pages, Keynote and Numbers on the iPhone 4S in the Apple Store. Even the Apple staff closed other applications, it took quite awhile to load the applications and the demo templates. The staff said that waiting for 6-8 seconds is normal. Is that so? The occasional lag on Android seems to be about 3-4 seconds.
 

DodgeV83

macrumors 6502a
Feb 8, 2012
879
6
Hello, I went to a phone store that had both iPhone 4S and the Galaxy Note on display side by side. After playing with the Galaxy Note for a few minutes and then moved to the iPhone 4S, opening Pages, Keynote and Numbers on the iPhone seemed to take a long time. Of course, I can't compare as the Galaxy Note does not has these applications. However, I also found that I could not zoom in or out while taking video using the iPhone 4S.

Then, I went to an Apple Store. The staff told me that it is not possible to zoom in or out while taking the video using the iPhone 4S. Why? This seems to be kind of a basic function to me. How come there is no App to do that?

I also tried Pages, Keynote and Numbers on the iPhone 4S in the Apple Store. Even the Apple staff closed other applications, it took quite awhile to load the applications and the demo templates. The staff said that waiting for 6-8 seconds is normal. Is that so? The occasional lag on Android seems to be about 3-4 seconds.

Those are very complex apps, similar to opening up a big game. If those apps were on Android it technically would take longer, since the iPhone 4S has a much faster CPU/GPU and a much faster disk read speed.

If you wants real comparison, try opening a graphically intensive game like Real Racing 2 on the Note. It will take a few second as well. I tested this with the HTC Rezound and the iPhone 4S opened it faster.
 

lilo777

macrumors 603
Nov 25, 2009
5,144
0
Those are very complex apps, similar to opening up a big game. If those apps were on Android it technically would take longer, since the iPhone 4S has a much faster CPU/GPU and a much faster disk read speed.

If you wants real comparison, try opening a graphically intensive game like Real Racing 2 on the Note. It will take a few second as well. I tested this with the HTC Rezound and the iPhone 4S opened it faster.

That's a false information. Many Android phones (Galaxy Note included) have faster CPUs than iPhone 4S. iPhone 4S has ARM A9-dual core CPU clocked at 800MHz. Galaxy Note has dual-core 1.4 GHz ARM Cortex-A9 CPU, so it is about 75% faster. GPU does nothing for loading apps. That's why Samsung phones starting from Galaxy SII and on (their better models) all load apps and browse web faster than iPhone 4S.
 

cynics

macrumors G4
Jan 8, 2012
11,959
2,154
That's a false information. Many Android phones (Galaxy Note included) have faster CPUs than iPhone 4S. iPhone 4S has ARM A9-dual core CPU clocked at 800MHz. Galaxy Note has dual-core 1.4 GHz ARM Cortex-A9 CPU, so it is about 75% faster. GPU does nothing for loading apps. That's why Samsung phones starting from Galaxy SII and on (their better models) all load apps and browse web faster than iPhone 4S.

Do they though? All we have are benchmarks to compare and the 4S almost always dominates them.

None of us have the means for more accurate testing.

Things I run on my android devices "feel" they run the same but that's by no means accurate. Even if the results are skewed by the software then logic would dictate the end experience would be better.

Comparing benchmarks across platforms is kind of silly regardless. App devs have to use different programming which will mess up any results for comparison.

Real world I found the devices very similar. A game like shadowgun runs the same between my android tablet and my 4s. However when I over clock the tablet from 1 ghz to 1.7 it loads the game significantly faster then my 4s. But even over clocked and loading things faster then my 4s the 4s still has a better linpack score. The tablet will triple to quadruple all benchmarks when over clocked but still shadows the 4s. So how useful are benchmarks if they do not reflect real world performance?

I usually try to avoid this discussion because people get so set on comparing useless numbers it's not worth the argument.
 

lilo777

macrumors 603
Nov 25, 2009
5,144
0
Do they though? All we have are benchmarks to compare and the 4S almost always dominates them.

None of us have the means for more accurate testing.

Things I run on my android devices "feel" they run the same but that's by no means accurate. Even if the results are skewed by the software then logic would dictate the end experience would be better.

Comparing benchmarks across platforms is kind of silly regardless. App devs have to use different programming which will mess up any results for comparison.

Real world I found the devices very similar. A game like shadowgun runs the same between my android tablet and my 4s. However when I over clock the tablet from 1 ghz to 1.7 it loads the game significantly faster then my 4s. But even over clocked and loading things faster then my 4s the 4s still has a better linpack score. The tablet will triple to quadruple all benchmarks when over clocked but still shadows the 4s. So how useful are benchmarks if they do not reflect real world performance?

I usually try to avoid this discussion because people get so set on comparing useless numbers it's not worth the argument.

The benchmarks clearly indicate how slow iPhone 4S is. Here is a chart from anandtech.com for a CPU intensive SunSpider benchmark:

46105.png
 

lilo777

macrumors 603
Nov 25, 2009
5,144
0
Once again another useless benchmark. Sunspider for android is java based, not so with ios. So what is actually being compared?

SunSpider is Java based for all platforms. What's wrong with that? You can't test hardware without software and Java is as good a software as any for that purpose. Are you saying that anandtech is stupid and you are the smart one?
 

Wicked1

macrumors 68040
Apr 13, 2009
3,283
14
New Jersey
Hello, I went to a phone store that had both iPhone 4S and the Galaxy Note on display side by side. After playing with the Galaxy Note for a few minutes and then moved to the iPhone 4S, opening Pages, Keynote and Numbers on the iPhone seemed to take a long time. Of course, I can't compare as the Galaxy Note does not has these applications. However, I also found that I could not zoom in or out while taking video using the iPhone 4S.

Then, I went to an Apple Store. The staff told me that it is not possible to zoom in or out while taking the video using the iPhone 4S. Why? This seems to be kind of a basic function to me. How come there is no App to do that?

I also tried Pages, Keynote and Numbers on the iPhone 4S in the Apple Store. Even the Apple staff closed other applications, it took quite awhile to load the applications and the demo templates. The staff said that waiting for 6-8 seconds is normal. Is that so? The occasional lag on Android seems to be about 3-4 seconds.


I believe it is because the iPhone has a fixed camera and therefore only a Digital Zoom, I can not answer about the Note but I am thinking it has the same thing, I have not seen any newer phone with an Optical Zoom.
 

hot spare

macrumors 6502
Aug 22, 2011
340
66
SunSpider is Java based for all platforms. What's wrong with that? You can't test hardware without software and Java is as good a software as any for that purpose. Are you saying that anandtech is stupid and you are the smart one?

You should read more, and not just a few random graphs. You do realize that it depends on the Javascript and the underlying OS. It's not a perfect test for native raw power. Have you ever realized that by just updating to ICS, there is a huge change in SunSpider scores:

gsmarena_004.gif


Same happens with BrowserMark:

gsmarena_005.gif


This is for SGS2 - both Note and SGS2 use same SoC with different frequencies.

The same happened with iOS4 was upgraded to iOS5.

screen-shot-2011-10-12-at-6-18-56-pm.png


Do you think the SoC in iPhone4 just got better with upgrade iOS 5? So, what explains such substantial increase in scores? Just better Javascript performance due to optimization. SunSpider/BrowserMark do NOT perform native test on the SoC.

You will be better served by using GeekBench. It does not depend on underlying OS. Also, it's available across both platforms. Maybe they also have this for WP7, not sure though.
 

mrsir2009

macrumors 604
Sep 17, 2009
7,505
156
Melbourne, Australia
It's so funny to watch people get on their high horse and preach about stuff.

Driving and talking on the phone is only as a 2nd offense. You have to be breaking the law already to get cited.

Speeding isn't. Why don't we bitch at the op because he probably speeds too!!

I've sped, littered, made a illegal lane change, drove with a tail light out. Which phone do you guys think is right for me?!?

Littering isn't cool. You're making the country you live in look like **** hole.
 

lilo777

macrumors 603
Nov 25, 2009
5,144
0
You should read more, and not just a few random graphs. You do realize that it depends on the Javascript and the underlying OS. It's not a perfect test for native raw power. Have you ever realized that by just updating to ICS, there is a huge change in SunSpider scores:

Image

Same happens with BrowserMark:

Image

This is for SGS2 - both Note and SGS2 use same SoC with different frequencies.

The same happened with iOS4 was upgraded to iOS5.

Image

Do you think the SoC in iPhone4 just got better with upgrade iOS 5? So, what explains such substantial increase in scores? Just better Javascript performance due to optimization. SunSpider/BrowserMark do NOT perform native test on the SoC.

You will be better served by using GeekBench. It does not depend on underlying OS. Also, it's available across both platforms. Maybe they also have this for WP7, not sure though.

Nobody uses GeekBench for anything but Apple products. And yes, I sure knew about SubSpider score dependency on JavaScript engine implementation. The, of course, Apple's implementation is regarded as one of the best. That's why iPhone 4S had higher SunSpider scores than Samsung Galaxy SII before it got ICS. But that was true only for stock web browser. SGSII beat iPhone 4S handily when using Mozilla.

So, there is a dependency but the consensus is that current versions of both JavaScript engines are roughly on par.
 

cynics

macrumors G4
Jan 8, 2012
11,959
2,154
You should read more, and not just a few random graphs. You do realize that it depends on the Javascript and the underlying OS. It's not a perfect test for native raw power. Have you ever realized that by just updating to ICS, there is a huge change in SunSpider scores:

Image

Same happens with BrowserMark:

Image

This is for SGS2 - both Note and SGS2 use same SoC with different frequencies.

The same happened with iOS4 was upgraded to iOS5.

Image

Do you think the SoC in iPhone4 just got better with upgrade iOS 5? So, what explains such substantial increase in scores? Just better Javascript performance due to optimization. SunSpider/BrowserMark do NOT perform native test on the SoC.

You will be better served by using GeekBench. It does not depend on underlying OS. Also, it's available across both platforms. Maybe they also have this for WP7, not sure though.

Here is my real world testament to your post.

Linpack test on my 4S

6c0f4760-92f5-1cf5.jpg


Linpack test on my 4S AFTER a random app update

6c0f4760-9320-526e.jpg


NOTHING changed on the phone just the app did.

Aka benchmarking is only useful when comparing the same OS at the same TIME.

Granted I'm sure a lot of benchmarks have some credibility however on NEW devices is anyone actually saying "man I wish my x phone was as fast as my y phone"?

And if you are you also have to ask yourself what's going on with the phone. iOS safari is smoother to scroll the android browser most of the time but in the same respect safari FREEZES the downloading and rendering of a web page to prioritize UI, android does not. There was a time when I used benchmarks as buying tool but no longer.
 

DodgeV83

macrumors 6502a
Feb 8, 2012
879
6
The benchmarks clearly indicate how slow iPhone 4S is. Here is a chart from anandtech.com for a CPU intensive SunSpider benchmark:

Image

SunSpider benchmarks are not CPU benchmarks, they are Browser benchmarks.

The SunSpider website states:

It is designed to compare different versions of the same browser, and different browsers to each other.

Other sources state that SunSpider was designed:

to measure the performance of the JavaScript engine of a web browser.

When Apple upgraded Safari on the iPhone 4, the SunSpider scores were nearly 3x faster, scoring much better than most of the newer dual-core Android devices at the time.

If you're basing your assertions on Browser Javascript benchmarks, you aren't going to have accurate conclusions.

----------

Here is my real world testament to your post.

Linpack test on my 4S

Image

Linpack test on my 4S AFTER a random app update

Image

NOTHING changed on the phone just the app did.

Aka benchmarking is only useful when comparing the same OS at the same TIME.

Granted I'm sure a lot of benchmarks have some credibility however on NEW devices is anyone actually saying "man I wish my x phone was as fast as my y phone"?

And if you are you also have to ask yourself what's going on with the phone. iOS safari is smoother to scroll the android browser most of the time but in the same respect safari FREEZES the downloading and rendering of a web page to prioritize UI, android does not. There was a time when I used benchmarks as buying tool but no longer.

Hey Cynics, I addressed your post in the other thread, but assumed you missed it when you didn't respond.

After reading a little more about this, it seems the numbers are inflated if the problem size is too low, something about the processors of today being too fast to accurately judge with such a small problem set. Your screenshot of the app pre-update is using a much lower problem set. I was able to reproduce what you're seeing, even after the app update, by lowering the problem set to 200. Once the problem set is above a certain level, things are equalled out.

The Android Linpack app runs at a 550 Problem Size. My iPhone 4S gets 80.86 Max Mflop/s without multithreading, and it stays consistent even when I choose a 2000 Problem Size.

image4.png


As I've been saying, the iPhone 4S CPU is way ahead of the competition:

htc-one-x-linpack.png


This video shows the HTC One X Multi-threaded Linpack benchmark. The highest he got was 117, lowest 106.

http://www.youtube.com/watch?feature=player_detailpage&v=Jiw2kWSU9s4#t=236s

I ran it three times in Multi-threaded mode, for comparison's sake:

image5.png


Again the iPhone 4S comes out on top.
 

DodgeV83

macrumors 6502a
Feb 8, 2012
879
6
That's a false information. Many Android phones (Galaxy Note included) have faster CPUs than iPhone 4S. iPhone 4S has ARM A9-dual core CPU clocked at 800MHz. Galaxy Note has dual-core 1.4 GHz ARM Cortex-A9 CPU, so it is about 75% faster. GPU does nothing for loading apps. That's why Samsung phones starting from Galaxy SII and on (their better models) all load apps and browse web faster than iPhone 4S.

If you are familiar with the Intel vs AMD wars in the PC industry, you'd know that the CPU clock speed is often very deceiving. If you'd like to read this from a more official source, as I'm just a random guy on the internet, check out the link below:

http://arstechnica.com/ask-ars/2011...ip-between-cpu-clockspeed-and-performance.ars

Question: Intel's Sandy Bridge launch just brought its desktop CPU line up to 3.8GHz, but I remember that the Pentium 4 got up to 3.8GHz before being cancelled. So why is it that Sandy Bridge is just now getting to the clock speed levels that the Pentium 4 was at years ago? And how is it that Sandy Bridge still manages to outperform the older Pentium 4, even though it has a lower clock speed?

In a nutshell, the Pentium 4 took many more clock cycles to do the same amount of work as the original Pentium, so its clockspeed was much higher for the equivalent amount of work. This is one core reason why there's little point in comparing clockspeeds across different processor architectures and families—the amount of work done per clock cycle is different for each architecture, so the relationship between clockspeed and performance (measured in instructions per second) is different.

Unless you have a source stating that the Apple designed A5 architecture and family is a 100% exact replica as the ones you're comparing here made by Samsung, you cannot state that clockspeed is a relevant indicator. When you consider the raw CPU benchmarks I posted above, it don't seem likely.

If you're basing your assertions on the rated clockspeed of the processer, and Browser Javascript benchmarks, you aren't going to have accurate conclusions
 

hajime

macrumors 604
Original poster
Jul 23, 2007
7,728
1,215
I believe it is because the iPhone has a fixed camera and therefore only a Digital Zoom, I can not answer about the Note but I am thinking it has the same thing, I have not seen any newer phone with an Optical Zoom.

You can zoom in or out while taking video using the Note. That is a very useful feature.
 

lilo777

macrumors 603
Nov 25, 2009
5,144
0
If you are familiar with the Intel vs AMD wars in the PC industry, you'd know that the CPU clock speed is often very deceiving. If you'd like to read this from a more official source, as I'm just a random guy on the internet, check out the link below:

http://arstechnica.com/ask-ars/2011...ip-between-cpu-clockspeed-and-performance.ars





Unless you have a source stating that the Apple designed A5 architecture and family is a 100% exact replica as the ones you're comparing here made by Samsung, you cannot state that clockspeed is a relevant indicator. When you consider the raw CPU benchmarks I posted above, it don't seem likely.

If you're basing your assertions on the rated clockspeed of the processer, and Browser Javascript benchmarks, you aren't going to have accurate conclusions

As I already mentioned AMD/Intel comparisons are useless in this context. These two companies use absolutely different chip architectures and the architecture determines how much calculations CPU does per clock cycle. The architecture of Apple, NVIDIA, Broadcom, TI, Samsung etc. CPUs is identical and it is developed by ARM. Apple does not develop the architecture. With that in mind, clock speed gives us very good idea about the relative performance of these CPUs.
 

DodgeV83

macrumors 6502a
Feb 8, 2012
879
6
Comparing benchmarks across platforms is kind of silly regardless. App devs have to use different programming which will mess up any results for comparison.

Real world I found the devices very similar. A game like shadowgun runs the same between my android tablet and my 4s. However when I over clock the tablet from 1 ghz to 1.7 it loads the game significantly faster then my 4s. But even over clocked and loading things faster then my 4s the 4s still has a better linpack score. The tablet will triple to quadruple all benchmarks when over clocked but still shadows the 4s. So how useful are benchmarks if they do not reflect real world performance?

I usually try to avoid this discussion because people get so set on comparing useless numbers it's not worth the argument.

But if all apps on those platforms need to use that programming (Java for Android or Objective-C on its for example), then it is relevant to know what theoretical peak horsepower is available to the devs on each platform.

Regarding over clocking, mobile apps generally aren't designed to scale with more power (I don't think anyway) so overclocking would only affect loading times and possibly frame rate.

It depends on how you define real-world performance. I have a machine with a super fast loading time (CPU), but once the game starts its unplayable (GPU). Remember, there are a lot of things that affect performance, including Memory system design, which Apple chips are typically good at. Also, ShadowRun was optimized more for Android due to the developer's agreements (including additional funding they received) with NVidia, so that's not the best example.

I agree that benchmarks don't always translate to real world performance on different platforms. For example, the Xbox 360 benchmarks slower than the PS3, but due to the difficulties developing for the Cell processor, games that were on both platforms typically looked and performed better on the Xbox. However, that's not the case here...

To look at real-world performance, I prefer to look at the best each platform has to offer. I don't think it's a coincidence the platform that has the best looking games and best performing OS also happens to have the best benchmarks :)
 

DodgeV83

macrumors 6502a
Feb 8, 2012
879
6
As I already mentioned AMD/Intel comparisons are useless in this context. These two companies use absolutely different chip architectures and the architecture determines how much calculations CPU does per clock cycle. The architecture of Apple, NVIDIA, Broadcom, TI, Samsung etc. CPUs is identical and it is developed by ARM. Apple does not develop the architecture. With that in mind, clock speed gives us very good idea about the relative performance of these CPUs.

Unless you have a source stating that the Apple-designed A5 architecture and family is a 100% exact replica as the ones you're comparing here made by Samsung, you cannot state that clockspeed is a relevant indicator. When you consider the raw CPU benchmarks I posted above, it don't seem likely.

----------

You can zoom in or out while taking video using the Note. That is a very useful feature.

That's digital zoom, not really useful as it significantly degrades quality. If you want it anyway, there are some iPhone apps that allow you to zoom while doing some magic to try and reduce the degregation, but I've never been impressed with digital zoom on any device.
 

lilo777

macrumors 603
Nov 25, 2009
5,144
0
Unless you have a source stating that the Apple-designed A5 architecture and family is a 100% exact replica as the ones you're comparing here made by Samsung, you cannot state that clockspeed is a relevant indicator. When you consider the raw CPU benchmarks I posted above, it don't seem likely.

----------



That's digital zoom, not really useful as it significantly degrades quality. If you want it anyway, there are some iPhone apps that allow you to zoom while doing some magic to try and reduce the degregation, but I've never been impressed with digital zoom on any device.


In general I rely on anandtech.com for this type of information. You can read their take on ARM architectures here. Comparison with different architectures is here. On the same page youw ill find Linpack benchmark charts for Android phones. I noticed that anandtech never uses linpack benchmarks for iphone. I do not know why but there must be a reason. As I understand, Linpack benchmark scores are also very dependent on software (in case of Android - on Dalvik VM).

anandtech have special page where they show various benchmark scores for different phones. You can use this page to compare SunSpider benchmark scores for iPhone and Android phones but Linpack scores are provided only for Android.
 

hajime

macrumors 604
Original poster
Jul 23, 2007
7,728
1,215
After reading the specifications of the Galaxy S III, I still prefer the Galaxy Note. Anybody knows why the front-facing camera for the iPhone 4S, Galaxy Note and Galaxy S III are still below 2M? I think there are phones that use 12M rear-facing camera. Why these three phones still use a 8M camera?

As for the memory expansion, the Galaxy SIII can be expanded to 64GB while the Galaxy Note can be expanded to 32GB. The former uses microSD while the latter uses microSDHC. Why the phones cannot get more memory? Is it due to the limitation of the phone's hardware, the OS or the availability of higher capacity memory cards? Once the higher capacity memory cards become available, can I increase the memory of these two phones further?
 

hajime

macrumors 604
Original poster
Jul 23, 2007
7,728
1,215
Today I came across the Lumina 610. After trying it for about half an hour, I found it very easy to use and the user interface is nice. Even it is an inexpensive phone, I did not experience the lag I had with the expensive Galaxy Note and the iPhone 4S. Now I can understand why Steve Wozniak praised it. I don't need to wait for the iPhone 5. I probably choose between the Lumina 900, 610 or the Galaxy Note. Not sure if Windows Phone 7.5 has better security than iOS and Android. Any idea?
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.