Apple A8 Benchmarked

Page 20 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

Khato

Golden Member
Jul 15, 2001
1,225
281
136
One has to be careful with SPEC results: Intel is known to tune icc with it, while I doubt LLVM is being tuned with SPEC, so there's some bias (other companies such as Sun did/do the same). Also between the i3 and Core Duo there are more than 7 years of compiler tuning

OTOH I have little doubt that Haswell IPC is higher for bigger workloads than A8, as I think the uncore performance is much higher. But as a reminder, A8 gains more on SPEC than on Geekbench, so this might hint that Apple is starting to push uncore (wish them the best to catch up with Intel who have a huge lead there :biggrin.

Oh don't get me wrong, SPEC has plenty of failings as a benchmark as well. It's just fun to point out how on SPEC Intel has a massive lead over Apple's A8 while Geekbench claims that they're on par. Neither is a 'fair' comparison.

Interesting to compare memory access latencies. Note that the tests (Sandra and whatever AT is using are different).

Note that L1 latencies are roughly equal at ~4ns. L2 is roughly equal at ~14-17 ns.

However Haswell's L3 is only a tad bit slower than L2 and significantly faster than Apple's L3. Edram, despite being located on a separate die is faster (55 vs. 75 ns). Access to system Ram is nearly twice as fast on Haswell.

Not sure what A9 will bring.

Is the above meant to be a comparison between Haswell and A8 cache latencies? If so, you missed the point of the Haswell chart being measured in number of clock cycles rather than ns. If Anandtech's iPhone 5s commentary is correct and A8 is the same 3 clocks for L1, then it'd be roughly 2ns compared to Haswell's 1ns. It's difficult to guess where the A8 L2 latency is, going by the graph it'd be something around 14ns... compared to Haswell's 3ns. And once you get to L3 it's something like 75ns versus 5ns. But that's what's necessary in order to feed more execution resources running at almost 3x the frequency.
 

Nothingness

Platinum Member
Jul 3, 2013
2,769
1,429
136
Oh don't get me wrong, SPEC has plenty of failings as a benchmark as well. It's just fun to point out how on SPEC Intel has a massive lead over Apple's A8 while Geekbench claims that they're on par. Neither is a 'fair' comparison.
I definitely agree! And that's why many benchmarks have to be run to get a better understanding, and I'm glad someone finally ran SPEC to show that yes A8 is a good chip, but not at the same level as Intel higher-end ones.

PS - Just to put some weight to my claim about icc and SPEC here is an example:
SPEC CINT2006 Dell Precision T3400 QX9650 Apr-2008
SPEC CINT2006 Dell Precision T3400 QX9650 Nov-2008
Same machine, icc 10.1 vs 11.0, gain 10%. Note that 403.gcc doesn't gain a lot as this one is quite hard to tune by a compiler (branch heavy in particular, no SIMD possible, etc.).
 

TreVader

Platinum Member
Oct 28, 2013
2,057
2
0
I definitely agree! And that's why many benchmarks have to be run to get a better understanding, and I'm glad someone finally ran SPEC to show that yes A8 is a good chip, but not at the same level as Intel higher-end ones.

PS - Just to put some weight to my claim about icc and SPEC here is an example:
SPEC CINT2006 Dell Precision T3400 QX9650 Apr-2008
SPEC CINT2006 Dell Precision T3400 QX9650 Nov-2008
Same machine, icc 10.1 vs 11.0, gain 10%. Note that 403.gcc doesn't gain a lot as this one is quite hard to tune by a compiler (branch heavy in particular, no SIMD possible, etc.).
Intel can't even win a single smartphone. What makes you say they are "better" than Apple? If anything, they aren't even competition! If apple wanted to beat intel at x86, they would. On the other hand intel has wanted to *touch* apple in mobile since 2007 and hasn't come close to anything more than a powerpoint presentation saying they will.


But please, continue the Core M circle jerk in an thread about a smartphone SoC. Core could not be put into a "tablet" form factor until Ivy bridge, and that was a 12" joke with 3hrs of battery life. Even haswell can't fit into anything close to even an original iPad chassis! If you put core M into an iPhone, would it even fit? What would the battery life be like, maybe an hour?


Core M is the first intel chip thats useful in a tablet, hopefully, but its nothing short of comical that Apple's smartphones are beating intels x86 chips. You guys pitch 1.6x in SPECint as if it's impressive. Intel should be beating Apple by 10x with that so called "huge" process advantage and 30 years more experience and way more R&D and a way larger thermal envelope. But they can't
 
Last edited:

Nothingness

Platinum Member
Jul 3, 2013
2,769
1,429
136
Intel can't even win a single smartphone. What makes you say they are "better" than Apple? If anything, they aren't even competition! If apple wanted to beat intel at x86, they would. On the other hand intel has wanted to *touch* apple in mobile since 2007 and hasn't come close to anything more than a powerpoint presentation saying they will.


But please, continue the Core M circle jerk in an thread about a smartphone SoC. Core could not be put into a "tablet" form factor until Ivy bridge, and that was a 12" joke with 3hrs of battery life. Even haswell can't fit into anything close to even an original iPad chassis! If you put core M into an iPhone, would it even fit? What would the battery life be like, maybe an hour?


Core M is the first intel chip thats useful in a tablet, hopefully, but its nothing short of comical that Apple's smartphones are beating intels x86 chips. You guys pitch 1.6x in SPECint as if it's impressive. Intel should be beating Apple by 10x with that so called "huge" process advantage and 30 years more experience and way more R&D and a way larger thermal envelope. But they can't
Your rant was fun. I'm far from an Intel fanboi, and I'm the first to claim Intel has demonstrated its inability to design a competitive Soc for smartphones. But people claiming Apple A8 can be compared to Haswell performance-wise are just smoking crack.
 

TreVader

Platinum Member
Oct 28, 2013
2,057
2
0
I'm far from an Intel fanboy.


Why have to explain this to me? Is it because you think it yourself?


And the benchmarks speak for themselves, Intel is just barely keeping pace with apples smartphone chips. Deal with it.
 

Nothingness

Platinum Member
Jul 3, 2013
2,769
1,429
136
Why have to explain this to me? Is it because you think it yourself?
Because you cite me and then rant how people are praising Intel, saying "continue the Core M circle jerk", and so on. So just say you didn't mean to imply I'm an Intel fanboi and we'll be fine

And the benchmarks speak for themselves, Intel is just barely keeping pace with apples smartphone chips. Deal with it.
You mean Apple is at Haswell level? Oh so you're one of those...
 

asendra

Member
Nov 4, 2012
156
12
81
There's a rumor about Apple using Samsung 14nm process to manufacturate the A9.

http://9to5mac.com/2014/10/02/samsu...s-for-ios-devices-using-14-nanometer-process/

I was under the impression apple moved to 20nm while focusing only in small arquitectural improvements and efficiency (equivalent to a Intel 'tick'), and next year they would do a 'tock', focusing more in performance, while still using 20nm.

I personally find a little crazy that they would go from 28nm to 20nm to 14nm in just 2 years. You guys think is possible?
 

Nothingness

Platinum Member
Jul 3, 2013
2,769
1,429
136
There's a rumor about Apple using Samsung 14nm process to manufacturate the A9.

http://9to5mac.com/2014/10/02/samsu...s-for-ios-devices-using-14-nanometer-process/

I was under the impression apple moved to 20nm while focusing only in small arquitectural improvements and efficiency (equivalent to a Intel 'tick'), and next year they would do a 'tock', focusing more in performance, while still using 20nm.

I personally find a little crazy that they would go from 28nm to 20nm to 14nm in just 2 years. You guys think is possible?
I think it's unlikely unless they have two CPU design teams working in parallel : one designed A8 while the other was working on an improved micro-arch; the latter could already have ES in 20nm and start working on 14/16nm. That still would be crazy...
 

NTMBK

Lifer
Nov 14, 2011
10,269
5,134
136
I think it's unlikely unless they have two CPU design teams working in parallel : one designed A8 while the other was working on an improved micro-arch; the latter could already have ES in 20nm and start working on 14/16nm. That still would be crazy...

They released the A6 and A7 in the space of a year, with a completely revised CPU architecture and on a shrunk process node (32nm->28nm). I think it's pretty safe to assume that they have two teams working in parallel.
 

Enigmoid

Platinum Member
Sep 27, 2012
2,907
31
91
Is the above meant to be a comparison between Haswell and A8 cache latencies? If so, you missed the point of the Haswell chart being measured in number of clock cycles rather than ns. If Anandtech's iPhone 5s commentary is correct and A8 is the same 3 clocks for L1, then it'd be roughly 2ns compared to Haswell's 1ns. It's difficult to guess where the A8 L2 latency is, going by the graph it'd be something around 14ns... compared to Haswell's 3ns. And once you get to L3 it's something like 75ns versus 5ns. But that's what's necessary in order to feed more execution resources running at almost 3x the frequency.

Yeah thanks, can't believe I missed that.
 

TreVader

Platinum Member
Oct 28, 2013
2,057
2
0
Because you cite me and then rant how people are praising Intel, saying "continue the Core M circle jerk", and so on. So just say you didn't mean to imply I'm an Intel fanboi and we'll be fine

I was only pointing out that you and a few others have spent a good page waxing lyrical about why this specific benchmark means Intel isn't worried about A7/A8 and why Core is so awesome, when core isn't even in the same form factor. Why do you think that is? Or were you confused and mixed silvermont (A7/8 competitor) with Core (a desktop/server x86 CPU part) :hmm:


You mean Apple is at Haswell level? Oh so you're one of those...

Yes, I'm one of them. So is the guy this site is named after. And anybody else that isn't so stuck in the PC desktop mindset of 2003.
 

dawheat

Diamond Member
Sep 14, 2000
3,132
93
91
I was only pointing out that you and a few others have spent a good page waxing lyrical about why this specific benchmark means Intel isn't worried about A7/A8 and why Core is so awesome, when core isn't even in the same form factor. Why do you think that is? Or were you confused and mixed silvermont (A7/8 competitor) with Core (a desktop/server x86 CPU part) :hmm:

Yes, I'm one of them. So is the guy this site is named after. And anybody else that isn't so stuck in the PC desktop mindset of 2003.

I hope everyone who contributed to the IPC discussion ignores this guy. As an outsider, I found it very interesting to try to compare the IPC between the two arches, even if they are intended for different purposes.

Good thing this is a CPU forum and not the mobile forum.

If you have a problem with another poster take it to PMs. Your problems are not our problem and do not belong in discussion threads
-ViRGE
 
Last edited by a moderator:

TreVader

Platinum Member
Oct 28, 2013
2,057
2
0
I hope everyone who contributed to the IPC discussion ignores this guy. As an outsider, I found it very interesting to try to compare the IPC between the two arches, even if they are intended for different purposes.

Good thing this is a CPU forum and not the mobile forum.

Everybody just ignore this guy. He followed me from the mobile forum to fight for android. I guess?



I'm not even gonna report you, lol. And you obviously have no idea what a CPU is if you think that any CPU competes with any other. Does IBM compete with Huawei?

Tre, you're being cranky again. Go get a Snickers and try to enjoy your vacation.
-ViRGE
 
Last edited by a moderator:

Nothingness

Platinum Member
Jul 3, 2013
2,769
1,429
136
I was only pointing out that you and a few others have spent a good page waxing lyrical about why this specific benchmark means Intel isn't worried about A7/A8 and why Core is so awesome, when core isn't even in the same form factor. Why do you think that is? Or were you confused and mixed silvermont (A7/8 competitor) with Core (a desktop/server x86 CPU part) :hmm:
Silvermont is a joke, so comparing A8 against it is no fun. Given that Intel has two lines of CPU, then the logical next step is to compare A8 against Core architecture, especially when people want Apple Ax processors to get up the processing ladder into laptops and above, don't you think?

You should also have noticed I spent a fair amount of time explaining why SPEC should be taken with care due to Intel icc being heavily tuned for it.

Yes, I'm one of them. So is the guy this site is named after. And anybody else that isn't so stuck in the PC desktop mindset of 2003.
I work for a company that designs mobile CPU, so yes please explain me how much I am "stuck in the PC desktop mindset of 2003" and how much I'm an Intel lover, that will be a lot of fun :biggrin:
 

TreVader

Platinum Member
Oct 28, 2013
2,057
2
0
Silvermont is a joke, so comparing A8 against it is no fun. Given that Intel has two lines of CPU, then the logical next step is to compare A8 against Core architecture, especially when people want Apple Ax processors to get up the processing ladder into laptops and above, don't you think?

You should also have noticed I spent a fair amount of time explaining why SPEC should be taken with care due to Intel icc being heavily tuned for it.

I work for a company that designs mobile CPU, so yes please explain me how much I am "stuck in the PC desktop mindset of 2003" and how much I'm an Intel lover, that will be a lot of fun :biggrin:

Fair enough, I agree that it's no fun to compare silvermont and A7/A8.





I've stated before that Core is a great design and (IMO) the only decent thing intel has put out since the Pentium III. That doesn't mean it handily beats A7 or A8.


Core has the headroom to run higher clocks and definitely more performance on tap, but that's not the name of the game. Power consumption and thermals are a huge factor, and I think A7/8 are (as yet) on another level to Core.
 
Last edited:

dawheat

Diamond Member
Sep 14, 2000
3,132
93
91
Everybody just ignore this guy. He followed me from the mobile forum to fight for android. I guess?



I'm not even gonna report you, lol. And you obviously have no idea what a CPU is if you think that any CPU competes with any other. Does IBM compete with Huawei?

Follow you? I primarily read the Mobile, CPU, and OT forums and read through the most recent threads throughout the day. It just happens that you posted nonsense in some of the most recent thread in a couple forums.

I read through the discussion on the last page as some quite knowledgeable folks trying to compare the IPC of the A8 arch vs the newer Core arch, ignoring clock speeds and power usage. Most weren't taking it as statements that one is better, since it's common sense to not compare a mobile CPU vs. a desktop CPU. But as a knowledge exercise, it was certainly interesting.

Then you had to come and herp derp it up.
 

witeken

Diamond Member
Dec 25, 2013
3,899
193
106
Yes, I'm one of them. So is the guy this site is named after. And anybody else that isn't so stuck in the PC desktop mindset of 2003.
Where can I buy a 4GHz+ Cyclone CPU with Iris Pro level graphics?
 

TreVader

Platinum Member
Oct 28, 2013
2,057
2
0
Follow you? I primarily read the Mobile, CPU, and OT forums and read through the most recent threads throughout the day. It just happens that you posted nonsense in some of the most recent thread in a couple forums.

I read through the discussion on the last page as some quite knowledgeable folks trying to compare the IPC of the A8 arch vs the newer Core arch, ignoring clock speeds and power usage. Most weren't taking it as statements that one is better, since it's common sense to not compare a mobile CPU vs. a desktop CPU. But as a knowledge exercise, it was certainly interesting.

Then you had to come and herp derp it up.

Lol i've been in this thread since the start. You obviously have zero reading comprehension, or are lying, or both.




Do you have anything to offer besides your claim of being "knowledgable" and calling me names?
 

dawheat

Diamond Member
Sep 14, 2000
3,132
93
91
Lol i've been in this thread since the start. You obviously have zero reading comprehension, or are lying, or both.

Do you have anything to offer besides your claim of being "knowledgable" and calling me names?

Well I think we can both agree this back and forth is pointless. But please read my posts more carefully - I stated that as an outsider, I found the posts by *other* knowledgeable posters both interesting and informative. Unfortunately since I don't have a hardware background, there's not much I can contribute knowledge-wise to threads in this forum, especially when comparing architectures.

But it doesn't seem you do either, yet it doesn't keep you from posts like the below.

Originally Posted by TreVader
Intel can't even win a single smartphone. What makes you say they are "better" than Apple? If anything, they aren't even competition! If apple wanted to beat intel at x86, they would. On the other hand intel has wanted to *touch* apple in mobile since 2007 and hasn't come close to anything more than a powerpoint presentation saying they will.


But please, continue the Core M circle jerk in an thread about a smartphone SoC. Core could not be put into a "tablet" form factor until Ivy bridge, and that was a 12" joke with 3hrs of battery life. Even haswell can't fit into anything close to even an original iPad chassis! If you put core M into an iPhone, would it even fit? What would the battery life be like, maybe an hour?


Core M is the first intel chip thats useful in a tablet, hopefully, but its nothing short of comical that Apple's smartphones are beating intels x86 chips. You guys pitch 1.6x in SPECint as if it's impressive. Intel should be beating Apple by 10x with that so called "huge" process advantage and 30 years more experience and way more R&D and a way larger thermal envelope. But they can't
 

TreVader

Platinum Member
Oct 28, 2013
2,057
2
0
Well I think we can both agree this back and forth is pointless. But please read my posts more carefully - I stated that as an outsider, I found the posts by *other* knowledgeable posters both interesting and informative. Unfortunately since I don't have a hardware background, there's not much I can contribute knowledge-wise to threads in this forum, especially when comparing architectures.



But it doesn't seem you do either, yet it doesn't keep you from posts like the below.


There is more to a discussion than hardware, and I obviously know substantially more than you. I made a concise argument, you resort to name calling and lying about being knowledgeable and reading this thread.


Next time you want to go on a crusade for android in the CPU forum, at least take the time to actually figure out what thread you are posting in.
 

Enigmoid

Platinum Member
Sep 27, 2012
2,907
31
91
I was only pointing out that you and a few others have spent a good page waxing lyrical about why this specific benchmark means Intel isn't worried about A7/A8 and why Core is so awesome, when core isn't even in the same form factor. Why do you think that is? Or were you confused and mixed silvermont (A7/8 competitor) with Core (a desktop/server x86 CPU part) :hmm:

Intel probably loves the A7/A8/A9... from a business perspective (so far). Apple isn't directly competing with them with their core/atom line and its keeps competitors in the smartphone/tablet space behind in fabrication technology as Apple chews up TSMC's newest node capacity.

If Apple starts replacing core in their laptops however they may get concerned but that is at least several years away.
 
sale-70-410-exam    | Exam-200-125-pdf    | we-sale-70-410-exam    | hot-sale-70-410-exam    | Latest-exam-700-603-Dumps    | Dumps-98-363-exams-date    | Certs-200-125-date    | Dumps-300-075-exams-date    | hot-sale-book-C8010-726-book    | Hot-Sale-200-310-Exam    | Exam-Description-200-310-dumps?    | hot-sale-book-200-125-book    | Latest-Updated-300-209-Exam    | Dumps-210-260-exams-date    | Download-200-125-Exam-PDF    | Exam-Description-300-101-dumps    | Certs-300-101-date    | Hot-Sale-300-075-Exam    | Latest-exam-200-125-Dumps    | Exam-Description-200-125-dumps    | Latest-Updated-300-075-Exam    | hot-sale-book-210-260-book    | Dumps-200-901-exams-date    | Certs-200-901-date    | Latest-exam-1Z0-062-Dumps    | Hot-Sale-1Z0-062-Exam    | Certs-CSSLP-date    | 100%-Pass-70-383-Exams    | Latest-JN0-360-real-exam-questions    | 100%-Pass-4A0-100-Real-Exam-Questions    | Dumps-300-135-exams-date    | Passed-200-105-Tech-Exams    | Latest-Updated-200-310-Exam    | Download-300-070-Exam-PDF    | Hot-Sale-JN0-360-Exam    | 100%-Pass-JN0-360-Exams    | 100%-Pass-JN0-360-Real-Exam-Questions    | Dumps-JN0-360-exams-date    | Exam-Description-1Z0-876-dumps    | Latest-exam-1Z0-876-Dumps    | Dumps-HPE0-Y53-exams-date    | 2017-Latest-HPE0-Y53-Exam    | 100%-Pass-HPE0-Y53-Real-Exam-Questions    | Pass-4A0-100-Exam    | Latest-4A0-100-Questions    | Dumps-98-365-exams-date    | 2017-Latest-98-365-Exam    | 100%-Pass-VCS-254-Exams    | 2017-Latest-VCS-273-Exam    | Dumps-200-355-exams-date    | 2017-Latest-300-320-Exam    | Pass-300-101-Exam    | 100%-Pass-300-115-Exams    |
http://www.portvapes.co.uk/    | http://www.portvapes.co.uk/    |