Galaxy-Nexus VS iPhone 4S ?

[cynics]

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. 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.

Image

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

Image

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:

Image

Again the iPhone 4S comes out on top.

Maybe I'm missing something. But I could consistently get the same results with the old app. After update and plugging the old problem size into it I get consistent results, however they are WAAAY different like you showed.

My argument wasn't what's faster it was the inconsistency in the benchmark results. Doing the same "problem" should have the same solution.

Unless I'm missing something. I'm at work so I'm not 100% focused on this lol.

 

[Bugeyeblue]

I'd go with the Galaxy Nexus. I could fill a book with all the reasons why I'd make the decision, but here are some of my big reasons:

1. 4G LTE.
   Which you have to turn off if you want the battery to last for more the 6 hours.
2. Free wireless tethering.
   Same on iPhone. You have to root the nexus to get tethering for free, you have to jailbreak the iPhone to get tethering for free.
3. Widgets and custom Notification Bar status indicators.
   No argument from me here.
4. Google Navigation.
   Again, no argument. A great android feature.
5. Ability to change default programs/choose from multiple programs to perform a task.
   This is true. Part of open source that is nice.
6. Custom fields in contacts sync properly on Android when using Exchange (i.e. Gmail). Custom fields do not sync on the iPhone.
   I don't think I've had that problem.
7. LED notification light. Have a missed call/voicemail/email? Green light. Whatsapp message? White light. Facebook message/text? Blue light. Low battery? Red light. I love being able to just look at my device without having to turn on the screen and know if anything is waiting for me.
   As far as I saw on the gnex, you only get white. Is there any way to change the light color?
8. Google sync is compatible with anything and everything, vs iCloud which is limited to Apple devices.
   True, but google syncs fine with iPhone.
9. Better integration via the sharing API.
10. Better integration of apps...i.e. Google Voice integrating into the native dialer and auto-detecting international numbers to intercept those calls and dial them via Google Voice rather than your carrier, or apps like Fox to Phone or Chrome to Phone.
11. Open source platform.
    Always a benefit here.

Replies in the quote... You have some good arguments and points, but I just can't get over the insane battery life issues and having to turn all auto synching/push notifications/updates to be able to use the phone for a whole day.

 

[DodgeV83]

Maybe I'm missing something. But I could consistently get the same results with the old app. After update and plugging the old problem size into it I get consistent results, however they are WAAAY different like you showed.

My argument wasn't what's faster it was the inconsistency in the benchmark results. Doing the same "problem" should have the same solution.

Unless I'm missing something. I'm at work so I'm not 100% focused on this lol.

The old version didn't differentiate between multi-threaded and single-threaded, that (or something else) must've contributed to the inconsistencies that occur when using a small problem set. The real test would be to increase your problem set on the old app to 550, or even 1000, and see if it's consistent with the results of the new app.