Personal computing discussed

Moderators: David, mac_h8r1, Nelliesboo

 
MethylONE
Gerbil XP
Topic Author
Posts: 398
Joined: Tue Sep 05, 2006 5:30 pm
Location: Orlando

iPhone 5 Benchmarks...

Sun Sep 16, 2012 6:17 pm

Turns out it is a ARMv7 Dual Core, 1ghz CPU.

http://browser.primatelabs.com/geekbench2/1030202
License to kill gophers.
 
JohnC
Gerbil Jedi
Posts: 1890
Joined: Fri Jan 28, 2011 2:08 pm
Location: NY/NJ/FL

Re: iPhone 5 Benchmarks...

Sun Sep 16, 2012 6:23 pm

Geekbench is absolutely useless "benchmark" - its results vary from each run, even on the same exact smartphone model:
http://browser.primatelabs.com/geekbench2/1032717
http://browser.primatelabs.com/geekbench2/1032724

"ARMv7" is most likely a generic identifier (previous A4 and A5 processors in iPhone 4 and iPhone 4S also belong to ARMv7 architecture). Also, the CPU most likely has a variable clock rates, which GeekBench also cannot correctly detect (it can only "approximate" base clock speed and even it may not be the actual one) , so I'd rather wait for Apple (or someone else) to publish more reliable clock speed information, not based on "approximations" of this useless benchmark.
My subscription allows you people to exist on this site and makes me a better human being than you'll ever be
 
MethylONE
Gerbil XP
Topic Author
Posts: 398
Joined: Tue Sep 05, 2006 5:30 pm
Location: Orlando

Re: iPhone 5 Benchmarks...

Sun Sep 16, 2012 8:37 pm

JohnC wrote:
Geekbench is absolutely useless "benchmark" - its results vary from each run, even on the same exact smartphone model:
http://browser.primatelabs.com/geekbench2/1032717
http://browser.primatelabs.com/geekbench2/1032724

"ARMv7" is most likely a generic identifier (previous A4 and A5 processors in iPhone 4 and iPhone 4S also belong to ARMv7 architecture). Also, the CPU most likely has a variable clock rates, which GeekBench also cannot correctly detect (it can only "approximate" base clock speed and even it may not be the actual one) , so I'd rather wait for Apple (or someone else) to publish more reliable clock speed information, not based on "approximations" of this useless benchmark.


Many were speculating that this phone had quad core, it doesn't. Many also said it would be Coretex A15, it isn't. It is a custom spin based on ARMv7. The actual benchmarks I don't care about, just the data related to the hardware.
License to kill gophers.
 
JohnC
Gerbil Jedi
Posts: 1890
Joined: Fri Jan 28, 2011 2:08 pm
Location: NY/NJ/FL

Re: iPhone 5 Benchmarks...

Sun Sep 16, 2012 8:54 pm

Well, you can see that it's based on "ARMv7s" even without the useless "random number"-generated benchmark:
http://www.anandtech.com/show/6292/ipho ... ustom-core
My subscription allows you people to exist on this site and makes me a better human being than you'll ever be
 
MethylONE
Gerbil XP
Topic Author
Posts: 398
Joined: Tue Sep 05, 2006 5:30 pm
Location: Orlando

Re: iPhone 5 Benchmarks...

Mon Sep 17, 2012 10:02 am

JohnC wrote:
Well, you can see that it's based on "ARMv7s" even without the useless "random number"-generated benchmark:
http://www.anandtech.com/show/6292/ipho ... ustom-core


You'll notice in that very article he says he's waiting to see someone test it on Geekbench. Soooo... there ya go.
License to kill gophers.
 
derFunkenstein
Gold subscriber
Gerbil God
Posts: 22655
Joined: Fri Feb 21, 2003 9:13 pm
Location: WHAT?

Re: iPhone 5 Benchmarks...

Mon Sep 17, 2012 10:28 am

That's kinda cool, and in retrospect it makes sense. Surely they didn't just buy PA Semi to make SoCs based on completely off-the-shelf designs.
I do not understand what I do. For what I want to do, I do not do. But what I hate, I do.
 
MethylONE
Gerbil XP
Topic Author
Posts: 398
Joined: Tue Sep 05, 2006 5:30 pm
Location: Orlando

Re: iPhone 5 Benchmarks...

Mon Sep 17, 2012 10:59 am

Anand now has a follow up since the Geekbench results.

http://www.anandtech.com/show/6298/anal ... ch-results
License to kill gophers.
 
JohnC
Gerbil Jedi
Posts: 1890
Joined: Fri Jan 28, 2011 2:08 pm
Location: NY/NJ/FL

Re: iPhone 5 Benchmarks...

Mon Sep 17, 2012 5:03 pm

MethylONE wrote:
JohnC wrote:
Well, you can see that it's based on "ARMv7s" even without the useless "random number"-generated benchmark:
http://www.anandtech.com/show/6292/ipho ... ustom-core


You'll notice in that very article he says he's waiting to see someone test it on Geekbench. Soooo... there ya go.

He likes to blindly trust synthetic benchmarks (even though he himself points out that, for example, he's "seen some odd results" when using GeekBench) - I don't, especially considering the fact that same benchmark program can show about 30% difference in performance on same exact hardware :roll:
My subscription allows you people to exist on this site and makes me a better human being than you'll ever be

Who is online

Users browsing this forum: No registered users and 4 guests