Alleged iPhone X benchmarks pop up, blows away the Samsung Galaxy S8 in every regard

AppleInsider

With a single-core score more than double that of the Galaxy S8, the "iPhone 10,5" thought to be the iPhone X, leaves the Android device in the dust -- and is very close to the 13-inch entry-level i5 MacBook Pro.

The alleged benchmarks for the iPhone X popped up on Tuesday, and was first spotted by Twitter user. The details of the device as far as what we know about core count is accurate.

#iPhoneX spotted on GFXBench.https://t.co/lSPKKfAX5i#AppleEvent pic.twitter.com/CHTy9TLf87

-- Kid Pool (@KidPoolDead)

September 12, 2017

Other details about the processor include 32KB of L1 instruction cache, 32KB of L1 data cache, and 8MB of L2 cache. The motherboard is identified as a D211AP.

However, the "24 MHz" speed listed for the processor is clearly off -- and is either an indication of a fake, or a processor not properly identified by the benchmarking tool. Similar results with misidentified speeds have popped up before on the Geekbench Browser, most notably with new Intel processors.

There is only the lone benchmark for the device at present. AppleInsider has reached out to Geekbench for details on the provenance of the measurement.

For comparison, the A10 Fusion iPhone 7 Plus generates a score of 3514 in single-core mode, and 5970 in multi-core calculations. The new 12.9-inch iPad with A10X Fusion processor scores 3924 in single-core, and 9371 in multi-core.

The Samsung Galaxy S8 scores 2024 in single-core, and 6279 in multi-core, with eight cores.

Benchmarking an ARM processor in a phone and comparing the numbers to an x86 processor, like in the MacBook Pro, is perilous because of architectural differences and the optimizations between processor families. The more linear comparison is mobile to mobile -- and ideally on the same operating system.

Join AppleInsider as we cover Apple's special media event live from Apple Park today at 10 a.m. Pacific. Make sure to follow us on Facebook, Twitter, YouTube, and Instagram for all the latest news, analysis, photos, video and more.

engdev

Repost (posted about this 20 minutes ago):

Looks good, although, only a ~15% increase in single core performance is a bit disappointing.

From my understanding Geekbench isn't the best to compare x86 and ARM, so don't go throwing out your Macbooks just yet. There's a reason websites like Anandtech don't use Geekbench.

ravnorodom

Drooling.... want one.

engdev

AppleInsider said:

Benchmarking an ARM processor in a phone and comparing the numbers to an x86 processor, like in the MacBook Pro, is perilous because of architectural differences and the optimizations between processor families. The more linear comparison is mobile to mobile -- and ideally on the same operating system.

Kudos for highlighting that, all to often people assume their phone/tablet is more powerful than their laptop, when there are many other factors that would weigh in.

It should also be noted that Geekbench will actually pause the benchmark to stop thermal throttling, so everything seen is mostly peak performance and may not be sustained in long term.

Geekbench inserts a pause (or gap) between each workload to minimize the effect thermal issues have on workload performance. Without this gap, workload that appear later in the benchmark would have lower scores than workload that appear earlier in the benchmark.

Source: Geekbench

maestro64

The issue the android guys have is their hardware maybe fast, however, once you put Android OS on the hardware it cripple it, as time goes on, it slows down. Google has a mess of an operating system, which is not optimized for the hardware it is running on, lack good memory management and the list goes on. Android is doing exactly what Wintel did, Intel had to keep making faster and faster hardware just so the OS did not make it run slower than the pervious generation products. Windows was bloatware and Android is the modern version of bloatware for mobile phones.

edited September 2017

melgross

There has been an earlier benchmark circulating. So this is the second.

but the numbers seem strange. While the single core rating seems about right, though a bit less than I expected, the multi core number is very odd. The reason why the A10x has such good multiprocessing numbers is because it has three cores. And the multi core efficiency from these three cores is very good indeed. But here, we have an impossible situation unless what we’ve been reading about the A11 is wrong, which is possible. 

According to the leak, and what’s being reported by several people, is that Apple is using an odd 2 high performance core and 4 high efficiency core design. It’s assumed that it works about the same as the A10, with the high and performance cores not working at the same time.

if that’s true, there’s is simply no way that the multi core number can be equal to double the single core score. But here, it’s way greater than the single core score. So something is wrong. Either people are wrong, and Apple is using a 6 core 3+3 arrangement - same as the A10X, or, they’ve decided to copy the ARM Big/Little arrangement where all cores work when performance is needed. The latter is considered to be a problem, one that Apple avoided in their design.

it will be interesting to see what’s true in a bit over an hour from now.

edited September 2017

r1senfa17h

EngDev said:

Repost (posted about this 20 minutes ago):

Looks good, although, only a ~15% increase in single core performance is a bit disappointing.

From my understanding Geekbench isn't the best to compare x86 and ARM, so don't go throwing out your Macbooks just yet. There's a reason websites like Anandtech don't use Geekbench.

The iPhone 7 Plus has a single-core Geekbench 4 score of 3332. Looks more like a ~22% increase in single-core performance.

edited September 2017

engdev

melgross said:

There has been an earlier benchmark circulating. So this is the second.

but the numbers seem strange. While the single core rating seems about right, though a bit less than I expected, the multi core number is very odd. The reason why the A10x has such good multiprocessing numbers is because it has three cores. And the efficiency from this three cores is very good indeed. But here, we have an impossible situation unless what we’ve been reading about the A11 is wrong, which is possible. 

According to the leak, and what’s being reported by several people, is that Apple is using an odd 2 high performance core and 4cligh efficiency core design. It’s assumed that it works about the same was as the A10, with the high and performance cores not working at the same time.

if that’s true, there’s is simply no way that the multi core number can be equal to double the single core score. But here, it’s way greater than the single core score. So something is wrong. Either people are wrong, and Apple is using a 6 core 3+3 arrangement - same as the A10X, or, they’ve decided to copy the ARM Big/Little arrangement where all cores work when performance is needed. The latter is considered to be a problem, one that Apple avoided in their design.

it will be interesting to see what’s true in a bit over an hour from now.

Geekbench is reading it as a 6 core processor, so all 6 cores are running at once.

For the A10 Fusion and A10X Fusion, it would be 2 big or 2 small and 3 big or 3 small respectively. Hence Geekbench would show the A10 Fusion as 2 cores and the A10X Fusion as 3 cores.

Source (A10 Fusion): Geekbench
Source (A10X Fusion): Geekbench

engdev

r1senfa17h said:

EngDev said:

Repost (posted about this 20 minutes ago):

Looks good, although, only a ~15% increase in single core performance is a bit disappointing.

From my understanding Geekbench isn't the best to compare x86 and ARM, so don't go throwing out your Macbooks just yet. There's a reason websites like Anandtech don't use Geekbench.

The iPhone 7 Plus has a single-core Geekbench 4 score of 3332. Looks more like a ~22% increase in single-core performance.

You're right, although I recalled seeing numerous scores around 3500. Perhaps people just wanted to post the best scores they could find.

Edit: 

Here's the A10 Fusion at ~3500: Geekbench

edited September 2017

lkrupp

EngDev said:

Repost (posted about this 20 minutes ago):

Looks good, although, only a ~15% increase in single core performance is a bit disappointing.

From my understanding Geekbench isn't the best to compare x86 and ARM, so don't go throwing out your Macbooks just yet. There's a reason websites like Anandtech don't use Geekbench.

Well, for a certain groupthink, anything Apple does is disappointing. Thanks for the negative point of view.

engdev

lkrupp said:

EngDev said:

Repost (posted about this 20 minutes ago):

Looks good, although, only a ~15% increase in single core performance is a bit disappointing.

From my understanding Geekbench isn't the best to compare x86 and ARM, so don't go throwing out your Macbooks just yet. There's a reason websites like Anandtech don't use Geekbench.

Well, for a certain groupthink, anything Apple does is disappointing. Thanks for the negative point of view.

You're welcome! I'm glad I could be the dark cloud in the sky on your sunny day. /s

tmay

melgross said:

There has been an earlier benchmark circulating. So this is the second.

but the numbers seem strange. While the single core rating seems about right, though a bit less than I expected, the multi core number is very odd. The reason why the A10x has such good multiprocessing numbers is because it has three cores. And the multi core efficiency from these three cores is very good indeed. But here, we have an impossible situation unless what we’ve been reading about the A11 is wrong, which is possible. 

According to the leak, and what’s being reported by several people, is that Apple is using an odd 2 high performance core and 4 high efficiency core design. It’s assumed that it works about the same as the A10, with the high and performance cores not working at the same time.

if that’s true, there’s is simply no way that the multi core number can be equal to double the single core score. But here, it’s way greater than the single core score. So something is wrong. Either people are wrong, and Apple is using a 6 core 3+3 arrangement - same as the A10X, or, they’ve decided to copy the ARM Big/Little arrangement where all cores work when performance is needed. The latter is considered to be a problem, one that Apple avoided in their design.

it will be interesting to see what’s true in a bit over an hour from now.

Someone here suggested based on the leak, that the A11 has a new microarchitecture that allows heterogeneous computing, which would, I think be the answer to your question. Is it the first ARM to do so? I don't know, but the numbers look pretty dang great. That fact that the A11 is exceeding the A10x of the iPad Pro's in a smaller, less thermally capable form factor, the iPhone, and with less battery power available, seems to indicate just that. Huge innovation if so.

melgross

EngDev said:

melgross said:

There has been an earlier benchmark circulating. So this is the second.

but the numbers seem strange. While the single core rating seems about right, though a bit less than I expected, the multi core number is very odd. The reason why the A10x has such good multiprocessing numbers is because it has three cores. And the efficiency from this three cores is very good indeed. But here, we have an impossible situation unless what we’ve been reading about the A11 is wrong, which is possible. 

According to the leak, and what’s being reported by several people, is that Apple is using an odd 2 high performance core and 4cligh efficiency core design. It’s assumed that it works about the same was as the A10, with the high and performance cores not working at the same time.

if that’s true, there’s is simply no way that the multi core number can be equal to double the single core score. But here, it’s way greater than the single core score. So something is wrong. Either people are wrong, and Apple is using a 6 core 3+3 arrangement - same as the A10X, or, they’ve decided to copy the ARM Big/Little arrangement where all cores work when performance is needed. The latter is considered to be a problem, one that Apple avoided in their design.

it will be interesting to see what’s true in a bit over an hour from now.

Geekbench is reading it as a 6 core processor, so all 6 cores are running at once.

For the A10 Fusion and A10X Fusion, it would be 2 big or 2 small and 3 big or 3 small respectively. Hence Geekbench would show the A10 Fusion as 2 cores and the A10X Fusion as 3 cores.

Source (A10 Fusion): Geekbench
Source (A10X Fusion): Geekbench

That’s not how it read the A10 or A10X, so that’s not a given. If it’s happening, then it needs to be updated for this chip. Just so you know, if the high efficiency cores aren’t activated during performance tests, Geekbench will still know they’re there from reading this info in the chip, but can’t include them in tests.

anton zuykov

EngDev said:

Repost (posted about this 20 minutes ago):

Looks good, although, only a ~15% increase in single core performance is a bit disappointing.

From my understanding Geekbench isn't the best to compare x86 and ARM, so don't go throwing out your Macbooks just yet. There's a reason websites like Anandtech don't use Geekbench.

Anandtevh does not use it, because it is their competitor. It is like to say that BMW does not use Lexus components, therefore Lexus must be shitty on reliability. But the reality is quite the opposite if you look at stats and facts.

edited September 2017

cpsro

If these are real benchmarks, why can't I use the new iPhone for my desktop? It's faster than the old Mac Pro I still use as my main workstation.

anton zuykov

EngDev said:

Repost (posted about this 20 minutes ago):

Looks good, although, only a ~15% increase in single core performance is a bit disappointing.

From my understanding Geekbench isn't the best to compare x86 and ARM, so don't go throwing out your Macbooks just yet. There's a reason websites like Anandtech don't use Geekbench.

Also, when you code for a benchmark, you dont necessarily take into account if it is Intel or Arm or any other type/architecture. You just instead code A TASK that needs to be completed in a shortest amount of time. That is it.  OS and its underlying schedulling algs and other such stuff are not important because they are part of the infrastructure that works with a cpu and memory AS A WHOLE.. Saying that you cant compare certain chips becuase they run different OSs is absolutely stupid, mainly because such approach absolutely misses the point of a comparison. It is like saying that you cant compare two sprinters because they have different height or because they wear clothing of two different manufacturers. 

edited September 2017

spamsandwich

Well, that's unfortunate for Samsung. LOL!

anton zuykov

cpsro said:

If these are real benchmarks, why can't I use the new iPhone for my desktop? It's faster than the old Mac Pro I still use as my main workstation.

What do you actually mean, when you say you wanna use your iPhone as "my desktop"? Aren't there ipads for that that run the same iOS as iPhones do?

herbivore2

EngDev said:

AppleInsider said:

Benchmarking an ARM processor in a phone and comparing the numbers to an x86 processor, like in the MacBook Pro, is perilous because of architectural differences and the optimizations between processor families. The more linear comparison is mobile to mobile -- and ideally on the same operating system.

Kudos for highlighting that, all to often people assume their phone/tablet is more powerful than their laptop, when there are many other factors that would weigh in.

It should also be noted that Geekbench will actually pause the benchmark to stop thermal throttling, so everything seen is mostly peak performance and may not be sustained in long term.

Geekbench inserts a pause (or gap) between each workload to minimize the effect thermal issues have on workload performance. Without this gap, workload that appear later in the benchmark would have lower scores than workload that appear earlier in the benchmark.

Source: Geekbench

It's Intel that would benefit most from measuring peak performance as their CPUs heat up considerably under load and are forced to throttle back. Apple CPUs rarely if ever throttle. 

My iPhone and iPad never heat up unless being charged. Unlike my MacBook which heats up under load all the time. Especially when playing video. 

GPU performance of the iPad Pro trounces all integrated Intel units also. 

My iPad Pro is actually far more useful than any MacBook (pro). Superior battery life, integrated baseband radio, and the real productivity enhancement has been the pencil. 

For server workloads, Intel is quite good although AMD is now building some awfully compelling products. For mobile/portable computing, Apple is far better than Intel. It isn't even close. Core i5 level performance in a handheld device is actually pretty spectacular. Especially when the SOC essentially never throttles and has superior graphics capabilities. 

engdev

anton zuykov said:

EngDev said:

Repost (posted about this 20 minutes ago):

Looks good, although, only a ~15% increase in single core performance is a bit disappointing.

From my understanding Geekbench isn't the best to compare x86 and ARM, so don't go throwing out your Macbooks just yet. There's a reason websites like Anandtech don't use Geekbench.

Anandtevh does not use it, because it is their competitor. It is like to say that BMW does not use Lexus components, therefore Lexus must be shitty on reliability. But the reality is quite the opposite if you look at stats and facts.

I should have been more clear Anandtech doesn't use it for the comparison between x86 and ARM, but it uses Geekbench for certain comparisons such as ARM vs ARM. 

Source: Anandtech

Geekbench isn't consistent for that comparison.

anton zuykov

EngDev said:

anton zuykov said:

EngDev said:

Repost (posted about this 20 minutes ago):

Looks good, although, only a ~15% increase in single core performance is a bit disappointing.

From my understanding Geekbench isn't the best to compare x86 and ARM, so don't go throwing out your Macbooks just yet. There's a reason websites like Anandtech don't use Geekbench.

Anandtevh does not use it, because it is their competitor. It is like to say that BMW does not use Lexus components, therefore Lexus must be shitty on reliability. But the reality is quite the opposite if you look at stats and facts.

I should have been more clear Anandtech doesn't use it for the comparison between x86 and ARM, but it uses Geekbench for certain comparisons such as ARM vs ARM. 

Source: Anandtech

Geekbench isn't consistent for that comparison.

EngDev said:

anton zuykov said:

EngDev said:

Repost (posted about this 20 minutes ago):

Looks good, although, only a ~15% increase in single core performance is a bit disappointing.

From my understanding Geekbench isn't the best to compare x86 and ARM, so don't go throwing out your Macbooks just yet. There's a reason websites like Anandtech don't use Geekbench.

Anandtevh does not use it, because it is their competitor. It is like to say that BMW does not use Lexus components, therefore Lexus must be shitty on reliability. But the reality is quite the opposite if you look at stats and facts.

I should have been more clear Anandtech doesn't use it for the comparison between x86 and ARM, but it uses Geekbench for certain comparisons such as ARM vs ARM. 

Source: Anandtech

Geekbench isn't consistent for that comparison.

What do you mean by saying geekbench is not consistent? Can you elaborate?