Core M v.s. A8X in Geekbench 3

Page 2 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

Eug

Lifer
Mar 11, 2000
23,752
1,285
126
I like how Apple got around the whole "4 cores = bad" Chinese cultural thing by choosing three. Perhaps that was not the major motivator, as their cores are quite beefy and therefore expensive, but that should definitely help their sales over there.
Many Chinese are superstitious, but it seems to me that the mobile device geeks don't care. They just want m0ar speed!!!1 We're not talking about middle-aged housewives here.

Also, Apple didn't even bother advertising the number of cores. Triple-core is mentioned absolutely nowhere on their website. Furthermore, they have a ton of quad-core Macs on their site, and they've had a blowout quarter.
 

Eug

Lifer
Mar 11, 2000
23,752
1,285
126
I'm just impressed this thread even exists.

Never would I have thought people would have been seriously talking about an A8 series chip taking on Core M, regardless of how well this test suite works or doesn't work across platforms.
 

Bubbleawsome

Diamond Member
Apr 14, 2013
4,833
1,204
146
Huh. If it's really that comparable across platforms then an a8x at 2.0Ghz would be a beast. I could actually see them doing that. A 2.6-3Ghz modified a9 or a9x (so next gen) in a MacBook Air.
 

III-V

Senior member
Oct 12, 2014
678
1
41
Many Chinese are superstitious, but it seems to me that the mobile device geeks don't care. They just want m0ar speed!!!1 We're not talking about middle-aged housewives here.

Also, Apple didn't even bother advertising the number of cores. Triple-core is mentioned absolutely nowhere on their website. Furthermore, they have a ton of quad-core Macs on their site, and they've had a blowout quarter.
Well, people of all ages (even kids...) buy smartphones.

As far as the advertisement thing goes, that's a very good point. I think it's silly for them to not have made a point of the 2GB of RAM, though.
I'm just impressed this thread even exists.

Never would I have thought people would have been seriously talking about an A8 series chip taking on Core M, regardless of how well this test suite works or doesn't work across platforms.
It's definitely something we wouldn't have envisioned prior to the A7 coming out. The A8 series bores me though... and frankly it should, being a less radical change. I wonder where Apple will go from here.
 

jfpoole

Member
Jul 11, 2013
43
0
66
And there's a reason for that - http://www.primatelabs.com/blog/2013/12/developing-a-cross-platform-benchmark/

The problem with that approach being that it only applies to x86. (Not to mention, this appears to state that in some cases they're re-coding algorithms to avoid compiler optimizations while in others they're working around compiler shortcomings.) What happens for all of the ARM crowd though? Because there basically aren't any alternate OS options there to use for comparison - would be interesting to know if they even attempt to do the same level of leveling the playing field with respect to compiler optimizations or not.

Anthony's post only covered the automated system he developed for Geekbench 3 which (at the time) was limited to x86 platforms. It didn't cover any of the manual work we did during development.

What's great about the automated system is that it makes detecting cross-platform (or cross-compiler) issues really easy. What's also great is that even though the system runs on a subset of architectures, fixing an issue on one architecture can also fix it on other architectures (e.g., high-level optimizer issues where a compiler doesn't recognize that a loop can be unrolled or vectorized).

We do need to be careful for low-level code-generaiton issues on platforms and architectures not covered by the automated system. We can always fall back on manually examining the generated code, which is straightforward albiet time consuming. We try and use the same compilers for each minor update so this is work we only have to do once during development.

I hope this helps, and I'm happy to answer any questions.
 

ams23

Senior member
Feb 18, 2013
907
0
0
Engadget has Geekbench 3, Sunspider, and 3dmark Ice Storm Unlimited benchmarks listed for A8X in iPad Air 2: http://forums.anandtech.com/showpost.php?p=36831425&postcount=123



IIRC, Core M 5Y70 was demonstrated to have Sunspider scores close to 100ms and 3dmark Ice Storm Unlimited overall score close to 45,000, so it is a very very fast application processor in comparison (and hence why it is priced at a premium).
 
Last edited:

liahos1

Senior member
Aug 28, 2013
573
45
91
Engadget has Geekbench 3, Sunspider, and 3dmark Ice Storm Unlimited benchmarks listed for A8X in iPad Air 2: http://forums.anandtech.com/showpost.php?p=36831425&postcount=123



IIRC, Core M 5Y70 was demonstrated to have Sunspider scores close to 100ms and 3dmark Ice Storm Unlimited overall score close to 45,000, so it is a very very fast application processor in comparison (and hence why it is priced at a premium).

yes i guess while we compare stupid benchmarks, core m crushes in sunspider.

i dont know if 3dmark is considered a "bad" cross platform benchmark but the results are very good.
 

Khato

Golden Member
Jul 15, 2001
1,225
280
136
Anthony's post only covered the automated system he developed for Geekbench 3 which (at the time) was limited to x86 platforms. It didn't cover any of the manual work we did during development.

What's great about the automated system is that it makes detecting cross-platform (or cross-compiler) issues really easy. What's also great is that even though the system runs on a subset of architectures, fixing an issue on one architecture can also fix it on other architectures (e.g., high-level optimizer issues where a compiler doesn't recognize that a loop can be unrolled or vectorized).

We do need to be careful for low-level code-generaiton issues on platforms and architectures not covered by the automated system. We can always fall back on manually examining the generated code, which is straightforward albiet time consuming. We try and use the same compilers for each minor update so this is work we only have to do once during development.

I hope this helps, and I'm happy to answer any questions.

Thanks for the reply.

The automated system for x86 certainly makes sense from a development perspective, especially if the goal is to end up with the same performance (on x86) regardless of operating system. It should mostly remove the compiler from the equation so that you're comparing the execution capabilities of the hardware for a given piece of code. And, as you say, the effect of such will typically carry over to the non-x86 ISAs in terms of missing out on optimizations. (After all, I'd imagine much of that logic is more in the 'front end' of the compiler which wouldn't necessarily be ISA specific?)

Anyway, with respect to the validity of the subject of this thread - comparing x86 to ARM by means of Geekbench - what manner of checks are performed on the non-x86 versions of Geekbench to ensure that the compiled code is similarly optimized? Or put another way, to ensure that the results for such comparisons are based on the architecture rather than the compiler? Because when I see architectures of different ISAs with similar execution resources producing markedly different results my first thought is that it's due to how the algorithm ends up being compiled to each ISA.
 

MisterLilBig

Senior member
Apr 15, 2014
291
0
76
yes i guess while we compare stupid benchmarks, core m crushes in sunspider.

The core M will be good for web browsing.

"New for Broadwell-Y is a 3rd power state, PL3, which allows for even greater turbo boosting, but for only a very limited period of time – on the order of milliseconds."

Should kill most benchmarks left to right.
 

raghu78

Diamond Member
Aug 23, 2012
4,093
1,475
136
The A8X benchmarks point to a quad core device which is clocked at 1 Ghz for power limit reasons and which can turbo to 1.5 Ghz for single/dual threaded workloads. Three cores is very unlikely.

http://browser.primatelabs.com/geekbench3/1067761

http://www.youtube.com/watch?v=sBfvJn-fpnc

watch the video at 43:40 - 44:10 and 52:50 - 54:00 . They clearly state the A8X is twice as fast and the benchmarks are pretty close to it for multithreaded benchmarks.
 

Rakehellion

Lifer
Jan 15, 2013
12,182
35
91
The least believable thing about the new iPad is not its performance, but the number of cores it has. :hmm:
 

MisterLilBig

Senior member
Apr 15, 2014
291
0
76
I guess I'm not surprised with its performance, nor most people.

T300FA is comparable to Volantis. And Air 2 is slightly weaker in single core but more on multi core than both.

Now we just wait for the other new SoC's from the competition to see how they match up.
 

witeken

Diamond Member
Dec 25, 2013
3,899
193
106
Sunspider would probably be about 300, maybe a little less.

Which is still far away from Core M, which has 2x less transistors, 33% less cores, a 40% smaller die and 11 times less revisions (yay for apples-oranges comparison). Maybe they'll get there eventually, but they're chasing a moving target.
 

witeken

Diamond Member
Dec 25, 2013
3,899
193
106
I'm just impressed this thread even exists.

Never would I have thought people would have been seriously talking about an A8 series chip taking on Core M, regardless of how well this test suite works or doesn't work across platforms.

Why? Intel has to deal with the same laws of nature, they don't have magical power, they just have their many years of experience designing world class desktop chips and their bleeding edge node advantage. Apple has not done anything that hasn't been done before.
 

Idontcare

Elite Member
Oct 10, 1999
21,118
59
91
I like how Apple got around the whole "4 cores = bad" Chinese cultural thing by choosing three. Perhaps that was not the major motivator, as their cores are quite beefy and therefore expensive, but that should definitely help their sales over there.

I live in Taiwan (republic of china) and the whole "chinese people hate/avoid the number 4" thing is way overdone in terms of western impressions in my opinion.

Our buildings have 4th floors, we use coins and make change that tallies up to increments of 4's (4, 40, 44, 400, etc.) absolutely no one avoids the number "4" as if it were the plague, contrary to the way it is portrayed and captured in some westernized impressions of Asian culture.

We don't skip the 4th step when climbing stairs or ladders, we don't shun the 4th room on a given floor, etc. etc.

I can absolutely guarantee you not a single person here gives two shts whether their mobile phone has 3 cores or 4 cores or 5 cores. They care about price, features, and prestige (bling factor). That's it. In other words, no different than every other culture across the globe.
 

MisterLilBig

Senior member
Apr 15, 2014
291
0
76
Which is still far away from Core M, which has 2x less transistors, 33% less cores, a 40% smaller die and 11 times less revisions (yay for apples-oranges comparison). Maybe they'll get there eventually, but they're chasing a moving target.


Sunspider performance of the Core M is the least surprising and the most obvious improvement for it.

Core M will be fast for web browsing, browser gaming won't be as improved tho, only as much as IPC gains from last gen.
 

Enigmoid

Platinum Member
Sep 27, 2012
2,907
31
91
Which is still far away from Core M, which has 2x less transistors, 33% less cores, a 40% smaller die and 11 times less revisions (yay for apples-oranges comparison). Maybe they'll get there eventually, but they're chasing a moving target.

To be fair you have to count the PCH.
 

thunng8

Member
Jan 8, 2013
155
63
101
Engadget has Geekbench 3, Sunspider, and 3dmark Ice Storm Unlimited benchmarks listed for A8X in iPad Air 2: http://forums.anandtech.com/showpost.php?p=36831425&postcount=123



IIRC, Core M 5Y70 was demonstrated to have Sunspider scores close to 100ms and 3dmark Ice Storm Unlimited overall score close to 45,000, so it is a very very fast application processor in comparison (and hence why it is priced at a premium).

That was a reference platform with a massive heatsink.

First review shows severe throttling.

http://www.ultrabookreview.com/5486-lenovo-yoga-3-pro-review/

Eg. Stressing both cpu and gpu reduced the cpu frequency to 500mhz and gpu to 150mhz.
 

thunng8

Member
Jan 8, 2013
155
63
101
Sunspider performance of the Core M is the least surprising and the most obvious improvement for it.

Core M will be fast for web browsing, browser gaming won't be as improved tho, only as much as IPC gains from last gen.

IE11 is very good at Sunspider but not any good at any of the other major js benchmarks. In fact ie11 is close to 2x faster than Safari or Chrome on the same cpu but Chrome and Safari are more than 50% faster in more intensive js benchmarks like Octane2 and Kraken.

Make you wonder if the IE team spent a lot of time optimizing for Sunspider
 

witeken

Diamond Member
Dec 25, 2013
3,899
193
106
What else would you expect when trying to feed 24EUs worth of graphics performance and 2 desktop class CPU cores simultaneously in a 3.5W power envelope? Intel did a tremendous job with Broadwell, but you can't expect much more than how a 7W Haswell performed. I don't really understand why Lenovo didn't set the TDP at 6.5W, though.
 
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/    |