Benchmark tests show new iPad Pro models outperform MacBook Pro in some CPU & GPU tasks

[captain_to_fire]

I'm curious what the results would be if Stardock releases "Ashes of the Singularity" (Ashes of the Benchmark as Linus calls it) for iOS. I hope Linus will review the new iPad Pro but he maybe have to wait for iOS 11 which will bring more functionality like a Finder app, better multitasking, etc.

Sources: 9to5 Mac and Bare Feats

 

Quote

Apple earlier this month refreshed both its iPad Pro and MacBook Pro lineups with faster processors and other performance improvements. The company heavily markets the iPad Pro as a PC replacement, and new benchmarks from blog BareFeats show that the tablet may even outperform the more expensive MacBook Pro in some cases. (emphasis is mine)

 

The blog stacked the 2017 and 2016 MacBook Pro with Touch Bar models up against four different iPad Pro models. Notably, the MacBook Pro models tested were the top-of-the-line machines with Core i7 processors, 16GB of RAM, and Intel graphics. Here’s the specs of all the MacBook Pro and iPad Pro models compared:

• 2017 MacBook Retina 13-inch, 3.5GHz Dual-Core i7 processor, Intel Iris Plus Graphics 650 GPU, 16GB of 2133 MHz LPDDR3 memory, 1TB PCIe based flash storage
• 2016 MacBook Retina 13-inch, 3.1GHz Dual-Core i7 processor, Intel Iris Graphics 550 GPU, 16GB of 2133 MHz LPDDR3 memory, 1TB PCIe based flash storage
• 2017 iPad Pro 12.9 inch (iPad7,2), 2.39GHz A10X processor, 512GB flash storage, 4GB of memory
• 2017 iPad Pro 10.5 inch (iPad7,4), 2.39GHz A10X processor, 512GB flash storage, 4GB of memory
• 2015 iPad Pro 12.9 inch (iPad6,8), 2.26GHz A9X processor, 128GB flash storage, 4GB of memory
• 2016 iPad Pro 9.7 inch (iPad6,4), 2.24GHz A9X, 256GB flash storage, 2GB of memory

When it came to single-core CPU performance, the MacBook Pro reigned supreme, with the 2016 and 2017 models beating out all other iPad Pro models. As for iPad Pro, the 10.5-inch 2017 model came in at the top, followed closely by the 2017 12.9-inch model. For multi-core CPU performance, the 2017 MacBook Pro came in at the top with a score of 10,261. In second place, however, was the 10.5-inch iPad Pro followed by the 2017 12.9-inch model. The 2017 iPad Pros both beat out the 2016 MacBook Pro – a notable feat considering the power and price associated with the MacBook Pro.

 

Then came the Geekbench metal GPU tests. Here, the 10.5-inch iPad Pro beat out all other devices, including the MacBook Pro. The newest 12.9-inch model came in second, followed by the 2016 MacBook Pro. The 2017 MacBook Pro placed fourth, behind its predecessor. Lastly, the devices were run through a pair of GFXBench graphics tests. Yet again, the tests showed the iPad Pro reigning supreme, with the 12.9-inch and 10.5-inch 2017 models edging out both MacBook Pro models.

But one should remember that these are just synthetic benchmarks more often, they're not true representations of real world performance. But then again, this is not the first time a mobile SoC outperformed a desktop processor on a synthetic benchmark. The A9 chip used by the iPhone 6s outperformed an Intel Xeon in Geekbench too. I wouldn't call this fake news but I won't necessarily say that the new iPad Pro is overall faster than the MacBook Pro. It's probably because iOS is still way lighter than macOS. Another reason why the iPad Pro seemed faster is that  it's already rocking their new file system "APFS" while macOS Sierra is still using HFS+ by default and Mac users must wait until macOS High Sierra to use APFS. 

 

But Intel should probably get worried. I'm pretty sure deep down in Apple's R&D labs, Craig Federighi and Jony Ive are probably having prototypes of a MacBook Pro running on an ARM SoC instead of Intel. They once made Mac OS X Tiger and Leopard to run both on Power PC and Intel chips, maybe they're also making a macOS that can run on both Intel and their Ax series SoCs.

 

Hey @LinusTech, if you ever decide to review the iPad Pro, I have a test idea. Have the 10.5 iPad Pro and the MacBook Pro export the same 4K and 1080p/60fps footage via iMovie.

Edited June 16, 2017 by hey_yo_

[ARikozuM]

Dear Apple, 

 

Please release this as your next 6" iPhone. 

 

Unapologetically hoping,

ARM

[captain_to_fire]

1 minute ago, ARikozuM said:

Dear Apple, 

 

Please release this as your next 6" iPhone. 

 

Unapologetically hoping,

ARM

I'm sure they will. Apple's SoC more often outperforms Qualcomm Snapdragon's 8-core in some benchmarks too like single threaded tasks. If the difference between the next iPhone 7s+ and the iPhone 8 is just an AMOLED screen, I'd pick the 7s+.

[The Sloth]

Just now, ARikozuM said:

Please release this as your next 6" iPhone. 

 

if they can shrink those bezels down and get an AMOLED display, i agree. maybe they can fit a headphone jack and NOT make the new iPhone thinner. 

 

i am wondering the validity of geekbench 

[DrMacintosh]

5 minutes ago, nerdslayer1 said:

i am wondering the validity of geekbench 

because Apple can actually design an ARM CPU that kicks the competition in the ass? 

[The Sloth]

Just now, DrMacintosh said:

because Apple can actually design an ARM CPU that kicks the competition in the ass? 

not that, it's more about optimization rather than raw power. 

[DrMacintosh]

Its been well established for a few years now that the iPad is the only great tablet. Apple also has an uncanny ability to design powerful SOCs because you know,

 

dat Apple R&D money  

[DrMacintosh]

Just now, nerdslayer1 said:

it's more about optimization rather than raw power. 

means are irrelevant as long as the ends result in amazing performance. 

 

At least for CPUs. 

[DrMacintosh]

8 minutes ago, nerdslayer1 said:

maybe they can fit a headphone jack

never going to happen. The iPhone has lost the headphone jack forever. 

[captain_to_fire]

Just now, DrMacintosh said:

means are irrelevant as long as the ends result in amazing performance. 

 

At least for CPUs. 

It seems that the Power VR GPU inside the A10X chip is better than Intel Iris Graphics.

[DrMacintosh]

Just now, hey_yo_ said:

It seems that the Power VR GPU inside the A10X chip is better than Intel Iris Graphics.

Is that surprising? The power of Apple money boners is stronger. 

[SCHISCHKA]

but can it run xcode?

this is why it will never replace a laptop

[DrMacintosh]

4 minutes ago, SCHISCHKA said:

this is why it will never replace a laptop

implying that most people want a laptop to code? 

[captain_to_fire]

4 minutes ago, SCHISCHKA said:

but can it run xcode?

this is why it will never replace a laptop

Nope. But it only shows how much RISC microprocessors are slowly catching up with CISC microprocessors.

[LAwLz]

I would just like to remind everyone that Geekbench is a really shitty benchmark.

[Vode]

This is Geekbench. *sigh*

 

Talk about useless article.

[NinerL]

Apple SoC's are second to none 

[Coaxialgamer]

5 hours ago, hey_yo_ said:

Nope. But it only shows how much RISC microprocessors are slowly catching up with CISC microprocessors.

I know this isn't really related to the topic but i had to say something. 

X86 isn't a CISC core. Hasn't been since the pentium pro. It's a RISC core with a CISC wrapper to maintain compatibility with 8086,286,386 and 486. 

Plus, the whole CISC vs RISC argument hasn't been relevant since the 90's. They are design philosophies more than anything, not indicators of performance. 

[VerticalDiscussions]

Hopefully Google figures out the magic too, if they are truly trying to make new custom CPU's for their smartphones.

 

About time we should also see a new Pixel C 2 tablet release this year, perhaps at a smaller pricetag?

 

Google Pixel 2 XL vs iPhone 8 is gonna be ferocious this year for sure.

[Đỗ Đức Huy]

Sees Geekbench

"Fake bench more like it"

Basically it have a very unusual way of weighing score. Too high emphasis on security hash functions which most SOCs have dedicated ICs for, while General CPUs don't.

 It's like taking an Ant miner and showing its bench in Hashing compared to an 6950x and then claiming it to be more powerful than the Extreme edition.

GeekBench should die in a fire

[captain_to_fire]

31 minutes ago, VerticalDiscussions said:

Hopefully Google figures out the magic too, if they are truly trying to make new custom CPU's for their smartphones.

 

About time we should also see a new Pixel C 2 tablet release this year, perhaps at a smaller pricetag?

 

Google Pixel 2 XL vs iPhone 8 is gonna be ferocious this year for sure.

I hope Google comes up with a design that doesn't look like an iPhone.

[VerticalDiscussions]

20 minutes ago, hey_yo_ said:

-snip-

Hopefully the one from this year looks like this, heck it would be the coolest thing ever (small bezels + front facing speakers):

 

[Sauron]

Stop using geekbench for this sort of comparison. It's heavily arm biased and does not reflect real world performance in any significant way.

[Drak3]

9 hours ago, DrMacintosh said:

Its been well established for a few years now that the iPad is the only great tablet.

Surface Pro, Vaio Z Canvas, Galaxy Tab S. Galaxy Tab Pro, Galaxy Tab Pro S, Note 8.0, Note 10.1, Nvidia Shield, Acer Predator 8.

 

9 hours ago, DrMacintosh said:

Apple also has an uncanny ability to design powerful SOCs because you know,

It's not so much that the SoCs are powerful as it is that Google's Android, what the god awful Geekbench is tested with, makes poor use of what's available to it without manufacturers/users heavily tweaking it, or outright replacing it with a custom ROM that is significantly closer to AOSP than Google's butchery job.

9 hours ago, DrMacintosh said:

means are irrelevant as long as the ends result in amazing performance. 

Geekbench isn't indicative of end performance at all.

[Droidbot]

come back when the knockoff Surface Pro can run proper x86 applications

then I'll happily buy one

 

17 minutes ago, Drak3 said:

Surface Pro, Vaio Z Canvas, Galaxy Tab S. Galaxy Tab Pro, Galaxy Tab Pro S, Note 8.0, Note 10.1, Nvidia Shield, Acer Predator 8.

I want a fucking Z Canvas refresh so bad. Just. Add. TB3.