• Guest, The rules for the P & N subforum have been updated to prohibit "ad hominem" or personal attacks against other posters. See the full details in the post "Politics and News Rules & Guidelines."

Apple A10 Geekbench 4 Score

Page 3 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

scannall

Golden Member
Jan 1, 2012
1,764
1,202
136
Common sense says that Apple's cores are not as good relative to Intel's cores as this single benchmark says they are. If this were any other company, results of a single benchmark would be met with healthy skepticism.
More likely answer is that even if they had an edge on Skylake, they would stay with x86 for their Mac line. A lot of their customers also need Windows compatibility. Also, to pull off a Rosetta like swap for ARM, it would need a %40ish percent lead to keep people happy enough to run their old software in emulation.
 
Mar 10, 2006
11,719
2,003
126
Common sense says that Apple's cores are not as good relative to Intel's cores as this single benchmark says they are. If this were any other company, results of a single benchmark would be met with healthy scepticism. Because it's Apple, it's interpreted as holy writ.
They really are pretty much Intel class. Seriously, don't try to downplay Apple's achievements here.
 
Mar 10, 2006
11,719
2,003
126
But whatever, it's far too easy for people to rationalise a plethora of weak reasons why Apple wouldn't switch to their supposedly magical, generations ahead of Intel, CPUs in their Mac range. Here's something a lot more difficult: You have a core with as good or better IPC than Intel, with a fraction of the power consumption, albeit at lower clocks. If this scenario is true, then you'd have to be a drooling idiot not to exploit that advantage in the ultra lucrative server market. Guess what?
Why would Apple want to get down and dirty in the server chip market? You need much more than a good CPU core to play in that game.
 

HurleyBird

Platinum Member
Apr 22, 2003
2,278
707
136
More likely answer is that even if they had an edge on Skylake, they would stay with x86 for their Mac line. A lot of their customers also need Windows compatibility. Also, to pull off a Rosetta like swap for ARM, it would need a %40ish percent lead to keep people happy enough to run their old software in emulation.
I don't think Apple gives a much of a hoot about users who put Windows onto their Macs. This is Apple we're talking about.

They wouldn't need parity with Rosetta, just "good enough" performance. The massive improvements in power consumption and/or battery life and/or thinness combined with Apple's marketing might would more than suffice to sell the new Macs. And even if you're right and sales were depressed a bit initially, it wouldn't take too long for the major apps to update to native code. In less than a year, you're enjoying all the benefits of a common platform, superior architecture, and lower hardware costs, with none of the downsides.

They really are pretty much Intel class. Seriously, don't try to downplay Apple's achievements here.
You've downplayed AMD and its Blender result close to the point of absurdity now, but when you replace that with Apple and Geekbench you do the exact opposite. It's doublethink, plain and simple, although I have no doubt you can come up with thin rationalisations to defend your hypocritical viewpoints.

Why would Apple want to get down and dirty in the server chip market? You need much more than a good CPU core to play in that game.
Sure, you need more than just a good CPU. But having a vastly superior architecture to your competitors is about the most important advantage you can have. Meanwhile many relatively small companies have been putting out ARM servers, and the server market is big, big money.
 
Last edited:
Mar 10, 2006
11,719
2,003
126
You've downplayed AMD and its Blender result close to the point of absurdity now, but when you replace that with Apple and Geekbench you do the exact opposite. It's doublethink, plain and simple, although I have no doubt you can come up with thin rationalisations to defend your hypocritical viewpoints.
Could you try to make your point without the personal attacks? Thanks.

Geekbench 4 does show Apple CPUs in the same league as Intel CPUs, but GB4 isn't the only test. Look at the SPEC2k6 results that AT did. The A9X could hang with Core m/Skylake even though Intel CPUs had a pretty big advantage on the compiler side of things.

There's also all of the web based stuff, like Kraken, SunSpider, JetStream, etc. If you run these on a Mac on Safari and on the iPhone on Safari, you can see pretty clearly that Apple is developing competitive CPU cores.

Sure, you need more than just a good CPU. But having a vastly superior architecture to your competitors is about the most important advantage you can have. Meanwhile many relatively small companies have been putting out ARM servers, and the server market is big, big money.
ARM servers, heck the server market, is peanuts compared to iPhone/iPad. Intel, the dominant server chip company, is having a great year if it can grow its $16B server biz by 10%.

iPhone is well north of $100 billion per year in revenue.

Intel is really a small fry compared to Apple, as much of a giant Intel is in its own right.
 
  • Like
Reactions: Nothingness

Eug

Lifer
Mar 11, 2000
23,015
536
126
Apple already tried the server market, and then closed up shop on them many years ago.

It seems enterprise doesn't want to pay the Apple tax for pretty servers and don't want to deal with companies that refuse to provide roadmaps. Surprise, surprise. The server market isn't really a good fit for the currently very secretive Apple who prides itself on aesthetically pleasing designs that they release at big press events. Apple didn't really want to create the massive support matrix either.

To criticize Apple for not pushing its phone chips into servers doesn't make much sense.
 
  • Like
Reactions: Phynaz

stingerman

Member
Feb 8, 2005
100
11
76
They really are pretty much Intel class. Seriously, don't try to downplay Apple's achievements here.
In some ways they are and in essential ways they are higher. Apples developing their processor to accelerate their own needs. For example, the Apple "Brain" that powers their local neural engine depends on hardware logic. They have many other secret regions on their processor for their own needs, not to mention the secure enclave.

And, when it comes to emulators, think about Apple hardwiring a limited "Rosetta" compatibility layer that will buy vendors time to recompile and test their software.. Anyway, having most of the major players on Xcode makes the conversion so much easier. Apple needs to get off Intel and quickly, Intel is putting Apple in a bad position both performance and cost wise.
 

Eug

Lifer
Mar 11, 2000
23,015
536
126
That's just too clunky. Not a great way to sell Macs, esp. in a declining market for computer sales.

The last time around it was a necessity. At this time it isn't.
 

HurleyBird

Platinum Member
Apr 22, 2003
2,278
707
136
Could you try to make your point without the personal attacks? Thanks.
I absolutely did not attack you personally. I attacked your viewpoints. Mainly, that you go above and beyond what could be considered healthy scepticism in one instance, while simultaneously displaying little to no scepticism in a similar situation when there are different players.

Geekbench 4 does show Apple CPUs in the same league as Intel CPUs, but GB4 isn't the only test. Look at the SPEC2k6 results that AT did. The A9X could hang with Core m/Skylake even though Intel CPUs had a pretty big advantage on the compiler side of things.

There's also all of the web based stuff, like Kraken, SunSpider, JetStream, etc. If you run these on a Mac on Safari and on the iPhone on Safari, you can see pretty clearly that Apple is developing competitive CPU cores.
In order, SPEC2K6 is as synthetic as you can get, and although it tended to hang close to Skylake in most subsets in AnandTech's article, there were a few subsets where Skylake was massively faster, and no subsets where Apple had a similar lead.

Next, all that web based stuff isn't isolating the CPU in the slightest. It's testing the entire platform, and is as much a test of the browser and OS as anything else. Unlike most other platforms, Apple is able to optimise IOS web performance very close to the metal. OSX Safari is not IOS Safari. What you're doing is analogous to pointing at DriveClub or The Order 1886 on the PS4 as proof that an HD 7870 is in the same league as, say, the GTX 980.

And lastly, while I don't trust Bapco much more than the makers of GeekBench, there is another cross platform "real world" benchmark that does happen to show the A9X in a very different light.

http://www.pcworld.com/article/3006268/tablets/tested-why-the-ipad-pro-really-isnt-as-fast-a-laptop.html

I take all of these benchmarks with a pinch of salt. You really need a battery of tests to have any kind of confidence, doubly so when you lack a common platform.

ARM servers, heck the server market, is peanuts compared to iPhone/iPad. Intel, the dominant server chip company, is having a great year if it can grow its $16B server biz by 10%.

iPhone is well north of $100 billion per year in revenue.

Intel is really a small fry compared to Apple, as much of a giant Intel is in its own right.
Everything Apple does is peanuts compared to iPhone, and aside from iPhone everything Apple does is peanuts compared to the server market. And of course you list revenue, even though you're fully aware that margins on server processors are ludicrously high.

If Apple's processors are as good you say they are, they would be absolutely disruptive in the server space. Multiple generations ahead of anything Intel or IBM can come up with in terms of perf/watt (and even assuming Apple's clocks stay low, comparable overall performance considering how low clocked those massive core count Xeons are compared to their desktop brethren. In actuality, a 24-core up-clocked A10 would likely eat a 24-core Xeon for lunch). Apple would be able to dominate servers in a way even Intel hasn't been able to, and make money hand over fist (no, not iPhone money, but more than anything else they do). If they could do it, don't you think they would?
 
Last edited:
  • Like
Reactions: RussianSensation

Eug

Lifer
Mar 11, 2000
23,015
536
126
^^^^ In those non-Geekbench tests, iPad Pro does remarkably well IMO. I would not have imagined an Apple ARM chip competing like that against Core even just 2 years ago.

BTW, Geekbench in that article is GB 3, not 4, obviously.

I think the most disappointing chip there is Atom.
 

dawheat

Diamond Member
Sep 14, 2000
3,123
71
91
While interesting, I find it all kind of pointless until some new software takes advantage of all this power in a smartphone. For what 98% of people do (surfing, texting, photos, FB, etc) there will be no discernible difference over the the 6S which was already overpowered. Similar performance with big gains in efficiency would be more useful on a daily basis until something like AR really pushes apps forward.
 

beginner99

Diamond Member
Jun 2, 2009
4,722
1,115
136
I think the most disappointing chip there is Atom.
True and not unexpected. But as the guy says some of the tests are platform tests and hardware with atoms is notorious for slow RAM and even slower storage (especially relevant in the unzip test).
 

IntelUser2000

Elite Member
Oct 14, 2003
7,250
1,839
136
If Apple's processors are as good you say they are, they would be absolutely disruptive in the server space. Multiple generations ahead of anything Intel or IBM can come up with in terms of perf/watt (and even assuming Apple's clocks stay low, comparable overall performance considering how low clocked those massive core count Xeons are compared to their desktop brethren. In actuality, a 24-core up-clocked A10 would likely eat a 24-core Xeon for lunch). Apple would be able to dominate servers in a way even Intel hasn't been able to, and make money hand over fist (no, not iPhone money, but more than anything else they do). If they could do it, don't you think they would?
It wouldn't be that simple. It seems that some people think all you need is one good thing to rule servers while PCs/Laptops/Tablets/Phones are much more complex.

Apple already tried the server market, and then closed up shop on them many years ago.
Power chips are very awesome in servers. However, if you delve down to the consumer side, it would be nothing special. Their awesome SMT4-8 is what makes it awesome, along with surrounding system interfaces and interconnects. Power 8 single thread were on par with Intel chips 3 generations ago(Broadwell generation, equivalent to Ivy Bridge). But that 10-15% difference is what makes us awe and drool on the PC side.

Same with other competitors too. On paper it seems so simple, but they just fall short of the best server chips. Because the vendors do not know everything that makes it an awesome server chip.

In fact, SpecCPU benchmark by Anand is more server/workstation oriented, and Intel chips lead here. You'll see the "IPC gains" of latest Intel chips are better on the server side than on PC. And it lines up pretty well with SpecCPU results too.

In this way, Apple and Intel are polar opposites of each other. The former knows how to captivate consumers, but little of servers. The latter has almost no clue about consumers, but knows quite a bit about servers.

To Arachnotronic: It's bit of unfair to compare in terms of revenue, you see.

Apple sells ~200 million units of iPhone/iPads annually. So in terms of chips its only 200 million of them.

Even in the "repressed" PC market, Intel is selling close to 300 million CPUs annum.

In terms of realistic TAM, Intel is far closer to it than Apple ever was/is. But because the ASP of "premium" CPUs are only in the range of low-cost phones, in terms of absolute revenue Apple is the bigger winner by far.

It may seem impressive for manufacturers like Apple to sell $600 phones constantly, but very impressive too that Intel can sell just a piece of silicon for $300 to quite a few as well. Apple may just be bringing Intel down to a more realistic level since they were too dominant in that space.
A cheap phone is in the range of "premium" CPU in terms of costs.
 
Last edited:

antihelten

Golden Member
Feb 2, 2012
1,764
274
126
And lastly, while I don't trust Bapco much more than the makers of GeekBench, there is another cross platform "real world" benchmark that does happen to show the A9X in a very different light.

http://www.pcworld.com/article/3006268/tablets/tested-why-the-ipad-pro-really-isnt-as-fast-a-laptop.html

I take all of these benchmarks with a pinch of salt. You really need a battery of tests to have any kind of confidence, doubly so when you lack a common platform.
I don't really have any personal opinion on the matter, but here is another article from the same site (PCWorld) about BAPCo and their benchmarking software:

http://www.pcworld.com/article/3023373/hardware/amd-accuses-bapco-and-intel-of-cheating-with-sysmark-benchmarks.html
 

IntelUser2000

Elite Member
Oct 14, 2003
7,250
1,839
136
Man, these benchmarks don't even matter in reality anyway.

No matter how superior Apple chips seem to be compared to Intel, or vice versa, they have both walled themselves in their own garden, and no one can penetrate it. One argument given by Intel investor in one investor forum when someone pointed out that Apple chips are faster than Atom ones is that "the OS is different, you'll never see Atom on iOS or Ax chips on Android!".

And even though their chip might be better, Android seems to be a good enough, or even a better alternative. And you'd still want x86 PC for some applications. Just a big ploy to make us spend more. Whether you buy Wintel, Android, or iOS.
 

stingerman

Member
Feb 8, 2005
100
11
76
In this way, Apple and Intel are polar opposites of each other. The former knows how to captivate consumers, but little of servers. The latter has almost no clue about consumers, but knows quite a bit about servers.
Your statement is a bit too general in my opinion. Apple runs some of the largest server farms and growing. If anyone understands their requirements for server processors, Apple would be one of them. They would be their own biggest customer.

And, in the businesses I've been involved in, we're looking at highly threaded small transactions for server processors. Of course the science, weather, and related industries are looking at vector operations. All of these are looking for high transactional throughput for the least amount of power/heat. So, you see where I am going? Apple could enter the Server business in a big way and maybe their Services partners will encourage them to do so...

It may seem impressive for manufacturers like Apple to sell $600 phones constantly, but very impressive too that Intel can sell just a piece of silicon for $300 to quite a few as well.
Intel's strength is currently their biggest weakness. They can't maintain their premium pricing when Apple could produce a similar performing processor for the form factors Apple develops for $20 a piece. When you think of Apple's needs, MacBook, Air, iMac - they are all mobile spec'ed components. And with Intel no longer able to improve performance at the pace they used to, I can't imagine Apple staying with Intel maybe just for the high-end Pro devices until a transition is complete.

And, I think, Apple's transitional Mac processor will accommodate an x86 emulator in a VM that's hardware accelerated.
 

Mopetar

Diamond Member
Jan 31, 2011
5,432
2,013
136
Apple already tried the server market, and then closed up shop on them many years ago. . . . The server market isn't really a good fit for the currently very secretive Apple who prides itself on aesthetically pleasing designs that they release at big press events. Apple didn't really want to create the massive support matrix either.
Your statement is a bit too general in my opinion. Apple runs some of the largest server farms and growing. If anyone understands their requirements for server processors, Apple would be one of them. They would be their own biggest customer.
I have to agree with this. Apple clearly wants to make moves to get into the kind of spaces (e.g. cloud services) that requires massive server farms which typically require using expensive server chips. If Apple could use their own chips, that could translate into some large cost savings.

It might also enable them to reenter the market as a provider as well because if they price comparable to other offerings from Intel and IBM, it would be Apple that's soaking up the nice profit margins, whereas before there wasn't as much margin room for them because they can't tack on their usual markup on top of the existing markup to use someone else's server-grade processors and expect to move a lot of hardware.
 

Eug

Lifer
Mar 11, 2000
23,015
536
126
I have to agree with this. Apple clearly wants to make moves to get into the kind of spaces (e.g. cloud services) that requires massive server farms which typically require using expensive server chips. If Apple could use their own chips, that could translate into some large cost savings.

It might also enable them to reenter the market as a provider as well because if they price comparable to other offerings from Intel and IBM, it would be Apple that's soaking up the nice profit margins, whereas before there wasn't as much margin room for them because they can't tack on their usual markup on top of the existing markup to use someone else's server-grade processors and expect to move a lot of hardware.
I'm not an expert here, because I don't do this for a living, but two of the biggest criticisms I have heard about Apple's foray into the server space were:

1) Apple had mediocre support. Sure, it was fine for small and some medium sized businesses, but it was useless for enterprise. The support wasn't as reliable, and the highest end support tiers that big customers would expect simply didn't exist.
2) Apple never would provide any sort of roadmap. What you see is what you get.

Sure, they could deploy these chips for in-house use, but if they want to sell to third parties, they're really going to have to up their game where it really counts. It's not all just about the chip performance. Does Apple want to compete head to head with the likes of IBM for server sales AND server support contracts with enterprise?
 

Mopetar

Diamond Member
Jan 31, 2011
5,432
2,013
136
Apple doesn't need to get into the support business just because they sell the hardware, and I suspect even if there's money to be made there, it's a business Apple doesn't need or want to be in.

Not all server sales are big iron that comes with extensive support contracts.
 

witeken

Diamond Member
Dec 25, 2013
3,876
154
106
Seeking Alpha makes a good point: QCOM has been the exclusive modem supplier for iPhone for a loooong time, which has now changed. I think the rumors from earlier, that Apple wants to integrate modem on die, have come a little closer to reality.

Edit: I wonder why INTC stock has reacted poor to the news. I hope we hear some things about it in the earnings call.

Edit2: Why else would Apple introduce versions of their phone that do not work with some providers?
 
Last edited:

Eug

Lifer
Mar 11, 2000
23,015
536
126
Apple doesn't need to get into the support business just because they sell the hardware, and I suspect even if there's money to be made there, it's a business Apple doesn't need or want to be in.

Not all server sales are big iron that comes with extensive support contracts.
Indeed. That's the market Apple already tried, and then quit.
 

jpiniero

Diamond Member
Oct 1, 2010
8,878
1,628
126
Edit: I wonder why INTC stock has reacted poor to the news. I hope we hear some things about it in the earnings call.
Edit2: Why else would Apple introduce versions of their phone that do not work with some providers?
CDMA is not really necessary if your provider doesn't use it (ie: AT&T, Sprint). Intel's modem is probably cheaper. So cheap that Intel's probably not making much on it.
 

scannall

Golden Member
Jan 1, 2012
1,764
1,202
136
Indeed. That's the market Apple already tried, and then quit.
The Xserve's were actually pretty decent. And a *little* better than what was out there at the time. BUT, a little better isn't enough of a reason to try and overcome inertia. The G5 in particular, while hot was a beast compared to the competition at the time. But, still not enough to warrant such a large change.

If you want to break into a market like that, you really have to come in way over the top, or just stay home. It won't fly any other way.
 

ASK THE COMMUNITY