Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
First, this is not really six cores, it is two fast cores plus four low power cores. If we look at future Macs: This chip easily beats all dual core Macs (yes, the very cheapest Macs havd. And it is reasonably competitive with quad core Macs, without quite beating them in performance. Where it will beat them is with not very demanding applications; if the four slow cores are powerful enough then they will use a lot less power and improve battery life.

But I’m sure Apple is working on chips with four fast + four slow cores at least. That should easily beat everything quad core, beat six core Macs, and almost draw even with eight cores and I expect that in the first ARM Macs. So: Improved low end Macs with 2+4 cores that you _might_ buy instead of quad core Intel. Improved mid range Macs with 4+4 cores that replace 4 or 6 core Intel Macs and that you _might_ buy instead of an 8 core Intel Mac. And all Intel Macs with fewer than eight cores get dropped.

(Actually, I assume 4+4 cores is close to finished, and they are working on the next size).

It is quite risky comparing a small powered cpu with small set of instructions to a 45~125 watt chip with a lot more instructions (remember atom series with micro ops that is only run at half the performance of their counterpart with CISC at the same clockspeed). I wouldn't be surprised if the next mac with Apple Silicon would run some softwares (including benchmarks) better than the Intel one, while the other run worse.

Regarding the DTK, it is strange that i haven't been able to found benchmark result other than geekbench score. Which is not a good benchmark for a sustained workload, the kind of thing that traditional computer do best compared to phone or tablet. We need a lot more than that to measure real world performance.
 
Cool to see how far technology has come. The Intel X5690 has a single core score of less than half of that at ~630 and multi core around ~3100. Thermal management is another issue entirely, but the fact that mobile devices have more powerful processors than the 5,1 Mac Pro is a pretty amazing.

My Mac Pro 2009 upgraded
Single core 681
Multi core 6974
Cinebench 3407
Metal 51587

We better get that macOS on an iPad then.
 
No, not really. They're about 43% more than your XS Max. Hardly 100% more.


It’s double the raw geek bench numbers. That’s what I was referring to. Apple said themselves 40% faster cpu and 30% graphics. So again, still kinda tempted to buy iPhone 12 based on current rumors, but will most likely wait till iPhone 12S next year.
 
With the Geekbench scores coming out, surely that means the release of the iPad is VERY close? I.e sometime next week possibly?
 
As CPUs get faster, operating systems get slower, so really the purpose of these faster chips is to just keep up
Truer words have never been spoken. Developers are spoiled with ridiculous layers of abstraction that were fantasy just 20 years ago. A lot of them are lazy, too, not coding to the highest possible efficiencies. Why bother tightening up your code when the chips are so fast?

And before anyone starts crying up a hissy fit over my calling developers spoiled and lazy, when was the last time you actually, really, had to truly program anything? Writing scripts and HTML is not coding.
 
Looks plausible.

A11 was a 25.1% jump in single-core performance; A12 20.1%; A13 19.6%. Now, another 19.3%.

That seems great until you include the clock speed:

(SC in GB5)
A10 - 2.34 Ghz - 727
A11 - 2.39 Ghz - 918 (+ 2% in clocks from A10) => 26% improvement
A12 - 2.49 Ghz - 1108 (+ 4% in clocks from A11) => 21%
A13 - 2.66 Ghz - 1327 (+ 7% in clocks from A12) => 20%
A14 - 2.99 Ghz - 1583 (+ 12% in clocks from A13) => 19%

If you subtract the clock increase from every jump in performance, the gains have been gone down with every generation of Ax soc...
A10 --> A11 (24%)
A11 --> A12 (17%)
A12 --> A13 (13%)
A13 --> A14 (7%)

A 7% architectural improvement feels somewhat intel-ish.
 
Impressive results — the new Macs will be beasts.

I fear what Apple will charge for them to make up all those years of R & D.
 
Useless synthetic benchmarks. How about some real workloads like Retroarch, Fortnite, 7zip, etc.?
 
That seems great until you include the clock speed:

(SC in GB5)
A10 - 2.34 Ghz - 727
A11 - 2.39 Ghz - 918 (+ 2% in clocks from A10) => 26% improvement
A12 - 2.49 Ghz - 1108 (+ 4% in clocks from A11) => 21%
A13 - 2.66 Ghz - 1327 (+ 7% in clocks from A12) => 20%
A14 - 2.99 Ghz - 1583 (+ 12% in clocks from A13) => 19%

If you subtract the clock increase from every jump in performance, the gains have been gone down with every generation of Ax soc...
A10 --> A11 (24%)
A11 --> A12 (17%)
A12 --> A13 (13%)
A13 --> A14 (7%)

A 7% architectural improvement feels somewhat intel-ish.
Too bad clock speed and performance don't work like that at all. Increasing clock speed is not lineair with performance.
Also, your chip has to be fast enough to finish it's instructions before the next clock, else a clock increase has literally no point. A clock increase is only useful if you actually increased the performance of you chip.
 
With the Geekbench scores coming out, surely that means the release of the iPad is VERY close? I.e sometime next week possibly?
Yes. Start hoping by this Monday...i guess this coming week, we will have the release of the ipad air 4 and the announcement of the next digital event for 13th October
 
I have the feeling that A14-ish ARM for Macs will be sufficient to be better from the rest, but not more. Apple likes to feed us with small spoons a year, so they can keep us always hungry.. to buy. I'm sure if they want to beat everyone, they have the means and already ARM processor in the working to accomplish that. 2021-2022 will be very interesting to see how Intel and the rest will respond to ARM.. no doubts, ARM is the future for now. :)
 
First, this is not really six cores, it is two fast cores plus four low power cores. If we look at future Macs: This chip easily beats all dual core Macs (yes, the very cheapest Macs havd. And it is reasonably competitive with quad core Macs, without quite beating them in performance. Where it will beat them is with not very demanding applications; if the four slow cores are powerful enough then they will use a lot less power and improve battery life.

But I’m sure Apple is working on chips with four fast + four slow cores at least. That should easily beat everything quad core, beat six core Macs, and almost draw even with eight cores and I expect that in the first ARM Macs. So: Improved low end Macs with 2+4 cores that you _might_ buy instead of quad core Intel. Improved mid range Macs with 4+4 cores that replace 4 or 6 core Intel Macs and that you _might_ buy instead of an 8 core Intel Mac. And all Intel Macs with fewer than eight cores get dropped.

(Actually, I assume 4+4 cores is close to finished, and they are working on the next size).

I'm curious how this compares with Mac Pro 5,1 2010 XEON chips. Has anyone ever tested the A chips against them?
 
If they don't update the iPad Pro in October, I'd be confused as to why anyone would buy it over the Air now. Their iPad lineup is a bit unbalanced at the moment.
They're waiting to transition to mini-LED, which is probably more important than processor upgrade to differentiate between Air and Pro.
 
It is quite risky comparing a small powered cpu with small set of instructions to a 45~125 watt chip with a lot more instructions (remember atom series with micro ops that is only run at half the performance of their counterpart with CISC at the same clockspeed). I wouldn't be surprised if the next mac with Apple Silicon would run some softwares (including benchmarks) better than the Intel one, while the other run worse.
ATOM is CISC and has the same instructions as any other x86-64 chip.

And all x86-64 cpus have micro-ops.

Your post doesn’t make sense.
 
That seems great until you include the clock speed:

(SC in GB5)
A10 - 2.34 Ghz - 727
A11 - 2.39 Ghz - 918 (+ 2% in clocks from A10) => 26% improvement
A12 - 2.49 Ghz - 1108 (+ 4% in clocks from A11) => 21%
A13 - 2.66 Ghz - 1327 (+ 7% in clocks from A12) => 20%
A14 - 2.99 Ghz - 1583 (+ 12% in clocks from A13) => 19%

If you subtract the clock increase from every jump in performance, the gains have been gone down with every generation of Ax soc...
A10 --> A11 (24%)
A11 --> A12 (17%)
A12 --> A13 (13%)
A13 --> A14 (7%)

A 7% architectural improvement feels somewhat intel-ish.
If it worked that way, you still need to compare apples to apples. Why can’t intel increase its clock speed to keep up a 20% per year cadence?

Because it ain’t that easy.
 
If they don't update the iPad Pro in October, I'd be confused as to why anyone would buy it over the Air now. Their iPad lineup is a bit unbalanced at the moment.

If you want to go more than 64gb, the Pro is still the better choice easily. For £40 more (U.K.) you get pretty much better everything - speakers (quad), microphones, cameras, screen (promotion, slightly bigger, slightly smaller bezels, brighter), faceId (preference but I think better for ipad), processor is on par (slower single core, faster multi core), likely better gpu (8 core), 6gb ram.

Other than for the colours, I’d say it’s a no brainier for anyone wanted more than 64gb.
 
  • Like
Reactions: Roode and nordique
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.