[TechTastic] A10X Geekbench 4 Score

[witeken]

I agree that Apple is just catching up to Intel now for ultrabook class performance. However, I consider that absolutely remarkable. If you had told me five years ago (when I bought my iPad 2), that Apple's ARM chips would be competing with Intel's laptop chips soon, I would've thought you were an ID10T.

Apple's chips are competing with Intel's ultrabook chips the same way Intel's ultrabook chips are competing with their desktop chips: about 5x the TDP allows for higher sustainable frequency and core clocks. Nothing remarkable.

The "if I were told 5 years ago" is getting old. And you know what? In fact you *were* told! I remember reading an AnandTech article 3 years ago that compared mobile chips with laptops. Anand has been talking about mobile being the same as pc industry but accelerated for a long time.

So that allowed Intel to release a mobile chip with 2x Apple performance, because the mobile vendors didn't have mature products yet, contrary to Intel. Oops x86 myth.

Edit: Don't forget that tablets can have peak power of something like 10W, so that's only 33% less than 15W, not the -80% delta between desktop chips.

 

[Eug]

Apple's chips are competing with Intel's ultrabook chips the same way Intel's ultrabook chips are competing with their desktop chips: about 5x the TDP allows for higher sustainable frequency and core clocks. Nothing remarkable.

That makes no sense. Are you trying to suggest that Apple's chips have 5X the TDP for similar performance? Cuz it sure sounds that way.

The "if I were told 5 years ago" is getting old. And you know what? In fact you *were* told! I remember reading an AnandTech article 3 years ago that compared mobile chips with laptops. Anand has been talking about mobile being the same as pc industry but accelerated for a long time.

3 years is not 5 years. Five years ago Apple released A5 (iPad 2). Three years ago Apple released A7 (iPad Air).

In those two years, Apple quadrupled the speed of their iPad SoC.

Thus, the perspective had changed dramatically in those 2 short years, because of what Apple was able to accomplish in that short time.

BTW, if that bench is accurate, that also means that CPU speed in Apple tablet SOCs has tripled again in the last three years. Not as fast a ramp up as before but overall after 5 years their tablet SoC is 12X as fast.

So that allowed Intel to release a mobile chip with 2x Apple performance. Oops x86 myth.

Edit: Don't forget that tablets can have peak power of something like 10W, so that's only 33% less than 15W, not the -80% delta between desktop chips.

Or you could say 15 Watts is 50% more power than 10 Watts. That's a pretty big deal IMHO.

 

[witeken]

Apple tablet SOCs has tripled again in the last three years. Not as fast a ramp up as before but overall after 5 years their tablet SoC is 12X as fast.

Or you could say 15 Watts is 50% more power than 10 Watts. That's a pretty big deal IMHO.

Then what do you say of Core m? Like 3x higher single thread than Silvermont in only 1 year .

Everyone on this forum knows power scales quadratically with voltage and linear with frequency for an up-to-cubic trend. And I'm not being scientific or precise, obviously.

 

[Arachnotronic]

Then what do you say of Core m? Like 3x higher single thread than Silvermont in only 1 year .

Everyone on this forum knows power scales quadratically with voltage and linear with frequency for an up-to-cubic trend. And I'm not being scientific or precise, obviously.

Core M isn't even a real SoC.

 

[lolfail9001]

You don't know what you're talking about, simple as that. Geekbench has no dependency on the OS, more on the compiler. But we can't stop you believing they are coming to take you

Considering it uses different compiler for different systems, well. Let's just say that results with Clang will be higher than results with Visual C on same hardware.

 

[Nothingness]

Considering it uses different compiler for different systems, well. Let's just say that results with Clang will be higher than results with Visual C on same hardware.

clang/llvm is used for iOS, MacOS X, Linux and Android, while VS is used for Windows, so it's easy to check.

http://www.geekbench.com/doc/geekbench4-cpu-workloads.pdf

This guy seems to have tested various OS on the same system http://browser.primatelabs.com/user/53371
Mac OS X 64 : 5697
Fedora 24 : 5777
Windows 64 : 5726
Looks pretty similar to me

 

[Arachnotronic]

OK, so this article/source is bogus. They took a performance estimate performed by somebody else and passed off as actual performance results.

Dang.

 

[Nothingness]

OK, so this article/source is bogus. They took a performance estimate performed by somebody else and passed off as actual performance results.

That looked too good to be true, and the scaling was too similar to the one from A9 to A9X. We'll have to be patient.

 

[Arachnotronic]

That looked too good to be true, and the scaling was too similar to the one from A9 to A9X. We'll have to be patient.

The original estimate (which was presented as such -- an estimate) was derived by taking the scaling from A9 to A9X and applying it to A10. That's why it's similar

 

[witeken]

It's actually kind of sad that the AnandTech site's comment system is so much behind the forums. I've just seem their recent article about the semiconductor companies and GloFo.

 

[dark zero]

Lol, no, I would rather look at it a different way: Apple is simply catching up to Intel.

You have a bad memory? When Intel launched Core m, it was twice as fast as Apple A8. It took Apple a whopping two (2!) generations to catch up to Broadwell-y's peak (~2.4GHz, so still a little lower than BDW-Y 2.7) performance... But now Intel is already at 3.6Ghz with Kaby Lake Y.

Apple's not going to get past Kaby Lake Y any time soon, with their short cells, and of course neither will Intel, although they will increase sustained performance obviously.

Now that I'm talking about this 3.6Ghz, didn't we at some point with Broadwell-y think that it would go to 4Ghz, I remember something like that .

Err... That 3.6 GHz is not sustained.... At best will stay at 2.2 GHz ... Apple was always at the same clock rate.

 

[witeken]

Err... That 3.6 GHz is not sustained.... At best will stay at 2.2 GHz ... Apple was always at the same clock rate.

Sustaining clock speeds also depends on thermal management of course.

According to HWiNFO, Skylake-Y can sustain 2Ghz dual core at less than 6W. I don't know what the numbers for Apple are, so I'm just throwing it out.

Only after 4.5 minutes it goes to 4.5W (Xiaomi Air).