iPhone 4S Geekbench Benchmarks ~68% Faster than iPhone 4

[aliasfox]

I seem to recall that my PowerBook G4 12" 867 MHz (which is no longer my primary machine, but I still use regularly) gets mid 40k on Sunspider, which is pretty comparable to what I remember my iPad getting. Surfing appears to be much faster on the iPad though, probably due to not having to parse Flash or swapping data to the HD.

In other words, on the Sunspider benches, the best new phones are coming in at twice as fast as an old G4.

 

[GodWhomIsMike]

I'm excited and nervous. I preordered the 16GB model, and thinking I should have gotten the 32GB model.

 

[Thunderhawks]

I actually agree with the "If you like your phone then great." Everytime I buy a new computer/phone/something like that I'll assess everything in the market and choose the best solution, regardless of brand (Although brand quality can play a factor), that's how I ended up with a MacBook Pro.

About the Droid fans, this is true, to which apple fans have generally responded "Specs don't matter," and this changes once apple takes the lead, as in this case.

I think that everyone should hold their horses on performance until the galaxy HD and Nexus Prime are out. Seems only fair to compare these phones given that they're all released at pretty much the same time. People say the S2 vs iP4 was not balanced due to the age of the iP4. Nexus Prime, android flagship vs iPhone 4S, apple flagship. Seems like a nice comparison.

Only way to be! Check out what you need to do, find the product that best suits your needs and buy whatever you think is best at that moment.
Never look back until your needs change.

Why would anybody waste their time arguing about somebody else's specs, let alone trying to make others feel inferior about their choice based on individual needs?

I can only remember a few times in my life that I bought something and regretted it and that was my fault for not assessing clearly what I needed
and doing research.

Itching to get a new 17" MacBookPro (Or even bigger), but I am convinced the next one is around the corner and will be a MacAirBookPro (LOL what a name) no burner/drive (haven't used it in 2 years) and all SSD etc.etc.

 

[KnightWRX]

Is the SunSpider test really irrelevant?

I understand you when you say it is a test of software but if the goal is to see how it stacks up to the competition it does matter who is fastest even when software is concerned. I know we're talking hardware here but it does matter to me which is faster even if it is just software.

The problem is you have to test with equivalent software if you want a real measure of performance. Android uses the same Webkit component and I believe the same Javascript engine. iOS 5 is latest and greatest, not even released software. Once Android releases the same latest and greatest, then maybe we can get a decent benchmark going. It seems 2.3.7 contains the same version of components and thus the scores are raised accordingly.

As such, the benchmark becomes a Webkit performance benchmark, not a hardware/OS benchmark. The problem is that Javascript is still being heavily optimized and the performance degradation is all in the Javascript stack itself. It's a poor benchmark.

 

[rockrigo]

I figured it would be powerful

only question I have is why was the SGII (and many of the Android phones) not running 2.3.5? It was released last week (via Kies)

 

[NAG]

This pretty much confirms that iPhone 4S only Siri is probably a soft launch and iPad 2 will get it some day (and I wouldn't be surprised if OS X gets Siri eventually...10.8?).

Edit: this assumes the 4S doesn't have some sort of Siri chip that only Siri gets to use.

 

[Oletros]

The problem is you have to test with equivalent software if you want a real measure of performance. Android uses the same Webkit component and I believe the same Javascript engine.

Javascript engine are different on Android (V8) and iOS (Nitro) and Sunspider depends heavily on javascript engine

 

[fenwah]

After reviewing these scores, I have a feeling that working from a 3GS all these years to a 4S will knock my socks off. Please induce coma on me until Friday morning.

 

[NAG]

After reviewing these scores, I have a feeling that working from a 3GS all these years to a 4S will knock my socks off. Please induce coma on me until Friday morning.

The A5 is amazingly fast compared to any other iOS device (and the A4 isn't a slouch).

 

[macridah]

I'm getting iPhone 4S envy. I didn't pre-order but might just line up Friday ...

 

[rockrigo]

just for ****** and giggles, had a friend run SunSpider on their SG I international version with stock 2.3.5 and got 3245 which is quite a difference from what this test shows with 2.3.4 (6134)

 

[kntgsp]

Yep, heavily underclocked APU. Just like we were all saying it was going to have despite all the mouth-breathers claiming otherwise.

The 512 MB RAM isn't too much of a surprise. Isn't Siri supposedly not run locally and has to have a data connection since all the processing is done server-side on Apple's network? Would explain why there's no need to bump the processing power. Doesn't really explain why the iPhone 4 doesn't get Siri though....

Haven't had much time to read over Siri specifics. Gonna have to go to some Googling.

 

[milo]

Don't forget that the iPad has a higher screen resolution and more physical space to devote to battery. It makes perfect sense that it's clocked higher.

I still want to see Siri for iPad 2.

 

[foodog]

More speed is always good, but I still just can't quite believe it took 16 months for this revision or that it required a change in release schedule from the normal July timeframe when the internals are pretty much near identical to the iPad 2 launching back in late March/April, which still ironically beats the newer device in GeekBench scores.

I know iOS 5 was a factor (or more-over Siri) but it didn't delay the iPad 2 launching with basically the same innards.

Tossing up whether to avail of my free carrier upgrade to the 4S when released, thereby extending my contract for another 18 months, or to come off contract and wait this revision out.

I guess I missed the iPad2 launching with iOS5.

 

[gorgeousninja]

Wirelessly posted (Mozilla/5.0 (iPhone; U; CPU iPhone OS 4_3_5 like Mac OS X; en-us) AppleWebKit/533.17.9 (KHTML, like Gecko) Version/5.0.2 Mobile/8L1 Safari/6533.18.5)

So this is officially the fastest phone on the planet?

No I thats the Galaxy SII and soon the Nexus Prime.

Apart from it obviously not being the SII as you can see from the scores. You did bother to read that bit didn't you?
Anyway werent you on here a few weeks ago trying to tell us all how the HTC superincreditastic was going to blow us all away? How did that go? Yeah we thought as much.

 

[tripjammer]

Apple engineers seem to be doing a great job with that ARM processor.

Can't wait to see it in a year .

Just wait till the Apple A6 4 core processor comes out next year in the Ipad 3 next spring...then the Iphone 5 next fall...hopefully we will have 1 Gb of ram by then...

----------

Yep, heavily underclocked APU. Just like we were all saying it was going to have despite all the mouth-breathers claiming otherwise.

The 512 MB RAM isn't too much of a surprise. Isn't Siri supposedly not run locally and has to have a data connection since all the processing is done server-side on Apple's network? Would explain why there's no need to bump the processing power. Doesn't really explain why the iPhone 4 doesn't get Siri though....

Haven't had much time to read over Siri specifics. Gonna have to go to some Googling.

Actually I think Siri is using some Iphone processing...probably one of the cores...and Siri probably will work on the Ipad 2 (update, i read that the Ipad 2 does not use the same type of microphones as the Iphone 4S so nix that device for SIRI, probably Ipad 3 which will also get the Iphone 4S 8MP camera)...Ipad 3 will probably when it gets out of BETA status in the next few months.

 

[dgree03]

Screen Size?

Doesnt screen size have a negative effect on benchmark scores of larger screen phones or tablets? So 3.5 screen vs a 4.3 in screen with the same or similar specs. would the 3.5 always bench higher?

If that is the case then whatever is in the Samsung 8.9 tablet or the iPad 2 is faster than anything on ANY handheld mobile device?

 

[KnightWRX]

Doesnt screen size have a negative effect on benchmark scores of larger screen phones or tablets? So 3.5 screen vs a 4.3 in screen with the same or similar specs. would the 3.5 always bench higher?

If that is the case then whatever is in the Samsung 8.9 tablet or the iPad 2 is faster than anything on ANY handheld mobile device?

Screen size no. Resolution does, but the benchmarks can be run to an off-screen renderer that has a equivalent resolution accross all devices.

 

[TallGuy1970]

The problem is you have to test with equivalent software if you want a real measure of performance. Android uses the same Webkit component and I believe the same Javascript engine. iOS 5 is latest and greatest, not even released software. Once Android releases the same latest and greatest, then maybe we can get a decent benchmark going. It seems 2.3.7 contains the same version of components and thus the scores are raised accordingly.

As such, the benchmark becomes a Webkit performance benchmark, not a hardware/OS benchmark. The problem is that Javascript is still being heavily optimized and the performance degradation is all in the Javascript stack itself. It's a poor benchmark.

While you may be technically correct. The average users don't care about the tech speak you just posted. They see a graph like this, they use a phone that works great for them, and they are happy.

 

[KnightWRX]

While you may be technically correct. The average users doesn't care about the tech speak you just posted. They see a graph like this, they use a phone that works great for them, and they are happy.

Fine, but we're talking tech here. Why do you insist on bringing down the level of conversation ?

We get it : Users don't care about specs. But we do. We like discussing them.

 

[aristotle]

Don't care about battery life and as long as the extra heat doesn't damage the phone it isn't a problem.

My first question was simple.

"Any ideas on how difficult it would be set the clock speed to 1Ghz?"

I wasn't asking "if" or "why" it should be done just if it was possible.

*WOOSH*
Hear that sound? That is the entire concept of the Hz myth flying right over your head. What is important is overall performance, not clock speed. Higher clock speed on those other chips is like an engine that has a lot of RPMs but low acceleration and power.

 

[Rodimus Prime]

Fine, but we're talking tech here. Why do you insist on bringing down the level of conversation ?

We get it : Users don't care about specs. But we do. We like discussing them.

bingo.
I too really want to see the CPU bench marks in floating point. Big reason is the tech geek side of me just likes seeing the numbers.

 

[tripjammer]

This pretty much confirms that iPhone 4S only Siri is probably a soft launch and iPad 2 will get it some day (and I wouldn't be surprised if OS X gets Siri eventually...10.8?).

Edit: this assumes the 4S doesn't have some sort of Siri chip that only Siri gets to use.

The Microphones are subpar in the Ipad 2....the Ipad 3 will get the Siri in the Spring...

 

[TallGuy1970]

Fine, but we're talking tech here. Why do you insist on bringing down the level of conversation ?

We get it : Users don't care about specs. But we do. We like discussing them.

Talk tech all you want. It's sometimes interesting to me, but it would be nice if it stayed a conversation and didn't get into the name calling and insults.

The debates on here about how (insert your favorite phone here) is so much better than (insert your not so favorite phone here) because it runs a 1.2GHz processor vs. a 800MHz processor get really tiring really quick.

 

[KnightWRX]

Talk tech all you want. It's sometimes interesting to me, but it would be nice if it stayed a conversation and didn't get into the name calling and insults.

It was a conversation about tech and there was no insults or name calling (nor has there been).

At this point I'm puzzled by your intervention at all. Did you come in here just to tell us users don't care ?

The debates on here about how (insert your favorite phone here) is so much better than (insert your not so favorite phone here) because it runs a 1.2GHz processor vs. a 800MHz processor get really tiring really quick.

On a clock per clock basis, a 1.2 GHz processor vs a 800 MHz can be better or not, it depends on a lot of factors. However, we're discussing SoCs that all pretty much use the same Cortex A9 design and architecture, so they should be pretty much equivalent as far as instructions per clock go.

I was just pointing out that we lack a benchmark that's accurate and reliable for pure CPU performance, to see if Apple gets better performance out of its implementation vs Qualcomm, nVidia, Samsung or Texas Instruments.

Again, I don't get the point of your intervention now at page 6. This conversation was perfectly smooth and going well. We didn't need you to remind us to stay in line or to remind us how users don't like all this tech talk.