• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

The Intel Atom Thread

Page 33 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
Airmont should be safe with a typical ~5% ''Intel tick'' IPC bump + higher base-clock/turbo clock but it definitely has a lot to improve on the GPU side. Hopefully Intel's new gfx architecture (Gen 8) and a lot more EUs than Bay Trail puts Airmont in a better position here next year. 🙂

Dude, even I will admit that it seems Intel needs to admit defeat to ARM. 1.3GHz dual core A7 with Smartphone class TDP is beating 4 core 1.4-2.4GHz Bay Trail with Tablet class TDP. And that's "28nm" vs 22nm.

The IPC of the A7 is on par with Ivy Bridge parts! Clock that sucker to 3GHz and it'll have a no worry replacement for the Macbook Air line and threaten even with x86 emulation.
 
I think Apple wants to reverse their market share losses to Samsung and other Android tablets manufacturers. The fact that A7 is the first 64 bit ARMv8 chip with 64 bit iOS and apps and shipping now is a clear indication of how seriously Apple views the threat. With A7 Apple is atleast a year ahead of other ARM SOC companies like Samsung, Qualcomm and Nvidia.
I don't think they designed the chip to counter Samsung or anyone else. Designing and validating a chip takes years so this chip is part of a long term strategy. In fact I wouldn't be surprised to see a new device between iPad and the MBA.
 
Dude, even I will admit that it seems Intel needs to admit defeat to ARM. 1.3GHz dual core A7 with Smartphone class TDP is beating 4 core 1.4-2.4GHz Bay Trail with Tablet class TDP. And that's "28nm" vs 22nm.

The IPC of the A7 is on par with Ivy Bridge parts! Clock that sucker to 3GHz and it'll have a no worry replacement for the Macbook Air line and threaten even with x86 emulation.
You can't just raise clocks this way, but I wonder how high A7 could clock without thermal limit of smaller form factors...
 
Dude, even I will admit that it seems Intel needs to admit defeat to ARM. 1.3GHz dual core A7 with Smartphone class TDP is beating 4 core 1.4-2.4GHz Bay Trail with Tablet class TDP. And that's "28nm" vs 22nm.

The IPC of the A7 is on par with Ivy Bridge parts! Clock that sucker to 3GHz and it'll have a no worry replacement for the Macbook Air line and threaten even with x86 emulation.

Can't just scale clocks up like that, but I agree. Apple's CPU teams have some serious talent on-board. A7 is an excellent SoC. Well balanced and with the best mobile CPU and GPU on the market.

But it's not defeat to ARM - ARM didn't design Cyclone - it's defeat to Apple. That's what money buys you, folks. It allows you to poach the best engineers and when you're making the kinds of margins on the devices you sell, you can splurge on your CPU design teams. The ROI is there.
 
You can't just raise clocks this way, but I wonder how high A7 could clock without thermal limit of smaller form factors...

2.3GHz for a tablet? Seriously apple run their chips at very low freq to have excellent battery life compared to others. Remember 4s...

But i have trouble accepting the a7 benchmarks and separating software from hardware here. Eg. Sunspider and memorybound bm.

But if the results is a reflection of reality then the days of x86 for mobile and desktop will end within 7 years. But i still dont trust the results...its hard to accept...
 
2.3GHz for a tablet? Seriously apple run their chips at very low freq to have excellent battery life compared to others. Remember 4s...

But i have trouble accepting the a7 benchmarks and separating software from hardware here. Eg. Sunspider and memorybound bm.

But if the results is a reflection of reality then the days of x86 for mobile and desktop will end within 7 years. But i still dont trust the results...its hard to accept...

Given Apple's maniacal focus on power efficiency and battery life the A7X or whatever is inside the upcoming ipad will be clocked conservatively at around 1.4 - 1.5 Ghz. remember the A6x was clocked at 1.4 Ghz and A6 at 1.3 Ghz. I expect the GPU to have more

http://www.anandtech.com/show/6426/ipad-4-gpu-performance-analyzed-powervr-sgx-554mp4-under-the-hood

"The A6X retains the 128-bit wide memory interface of the A5X (and it keeps the memory controller interface adjacent to the GPU cores and not the CPU cores as is the case in the A5/A6). It also integrates two of Apple's new Swift cores running at up to 1.4GHz (a slight increase from the 1.3GHz cores in the iPhone 5's A6). "

I expect the GPU to be a 6 cluster G6630 PowerVR Series 6 Rogue GPU. the iPad needs the extra GPU power as its running a massive 2048 x 1536 Retina display.

http://www.anandtech.com/show/7335/the-iphone-5s-review/7
 
Last edited:
2.3GHz for a tablet? Seriously apple run their chips at very low freq to have excellent battery life compared to others. Remember 4s...

But i have trouble accepting the a7 benchmarks and separating software from hardware here. Eg. Sunspider and memorybound bm.

But if the results is a reflection of reality then the days of x86 for mobile and desktop will end within 7 years. But i still dont trust the results...its hard to accept...

Moot point though, since iOS and Apple chips are both intertwined. One doesn't live without the another. If the resulting product runs faster than some combination of competitors it is no less a win.

But anyway, the A7 only proves the theory that Apple switching to full x86 is simply a load of bullocks by this point. More like the opposite is happening.
 
I'm kinda impressed by Apple.

I could foresee that the battle in 5 years will actually be more between Apple & Intel, than Intel & Qualcomm.


The question is what samsung will do - spend money on marketing your way out of worse performance - or jump ship to intel or invest in R&D themselves.
 
Moot point though, since iOS and Apple chips are both intertwined. One doesn't live without the another. If the resulting product runs faster than some combination of competitors it is no less a win.

But anyway, the A7 only proves the theory that Apple switching to full x86 is simply a load of bullocks by this point. More like the opposite is happening.

I agree. Apple has the resources to develop its own chips custom tailored to the workloads it wants to see. I don't see the Macs getting rid of Intel chips, but the truth is that the iPad will scale up to be the mainstream OS of choice and the Macs become a niche product.

Apple's got the money to build amazing chips (as we are seeing), and anybody who has followed their job boards knows that they've been hiring like mad for CPU teams.
 
But it's not defeat to ARM - ARM didn't design Cyclone - it's defeat to Apple. That's what money buys you, folks. It allows you to poach the best engineers and when you're making the kinds of margins on the devices you sell, you can splurge on your CPU design teams. The ROI is there.
Or you can turn it differently: no matter the huge margins Intel make on their chips, no matter the thousands of engineers they have, no matter the process advantage they have, no matter how serious they are attacking the mobile market, Intel still aren't ahead all of the competition; the return of the Conroe moment has yet to come.

I love that competition! I'm glad Intel and then Apple raised the bar, interesting times ahead for sure 🙂
 
I'm kinda impressed by Apple.

I could foresee that the battle in 5 years will actually be more between Apple & Intel, than Intel & Qualcomm.


The question is what samsung will do - spend money on marketing your way out of worse performance - or jump ship to intel or invest in R&D themselves.

Intel???
You tell me what Intel have to offer for phones right now?

Get real. The excitement for a win tablet cpu perf. lasted a week. And they dont even have a phone soc yet.
 
Or you can turn it differently: no matter the huge margins Intel make on their chips, no matter the thousands of engineers they have, no matter the process advantage they have, no matter how serious they are attacking the mobile market, Intel still aren't ahead all of the competition; the return of the Conroe moment has yet to come.

I love that competition! I'm glad Intel and then Apple raised the bar, interesting times ahead for sure 🙂

The Conroe moment will be harder to come by against the likes of Apple and Qualcomm - both with substantially better management & resources than AMD ever had.

Intel needs to do a lot better than this on both apps processors and connectivity/modems if it ever wants leadership in this space. Bay Trail will sell well, as I think the value prop of cheap, good Windows 8 systems (tabs, convertibles) is there, but smartphones is a whole other ballgame.

Merrifield will be 2C Silvermont + Rogue. I'm hoping it's much more competitive as an overall SOC than Bay Trail is against the Apple offerings.
 
Last edited:
Intel???
You tell me what Intel have to offer for phones right now?

Get real. The excitement for a win tablet cpu perf. lasted a week. And they dont even have a phone soc yet.

The irony here is the company that designed the A7 is also selling it to the target audience that is the most unlikely to care about SoC performance. This forum always seem to forget the SoC is merely only one part of the mobile success equation.
 
The irony here is the company that designed the A7 is also selling it to the target audience that is the most unlikely to care about SoC performance. This forum always seem to forget the SoC is merely only one part of the mobile success equation.

Class leading performance is part of the user experience. Your nonsensical proclamations across this forum about "good enough" are completely off-base. People love more performance as long as it enables new/better experiences.
 
"Intel Conroed the mobile market!"
"Apple will put an Intel chip on its phones!"
"Apple ditching Intel even for laptop/desktop? You bollocks?"
"Intel fab prowess is unmatched!"
"Samsung can't get its yields right"
"There will be BT Android tablets, doesn't matter that none have been announced"

Oh man, so much laugh in a single thread.
 
"Intel Conroed the mobile market!"
"Apple will put an Intel chip on its phones!"
"Apple ditching Intel even for laptop/desktop? You bollocks?"
"Intel fab prowess is unmatched!"
"Samsung can't get its yields right"
"There will be BT Android tablets, doesn't matter that none have been announced"

Oh man, so much laugh in a single thread.

Someone with only ~4 years of experience designing chips without fabs as one of their side projects versus somebody who lives and dies by bleeding edge fabs and selling CPUs over 30+ years. Yeah, I wonder who is Conroeing who. The ownage inflicted is just incredible.

BTW incoming A7X to wipe the tablet floor.
 
"Intel Conroed the mobile market!"
"Apple will put an Intel chip on its phones!"
"Apple ditching Intel even for laptop/desktop? You bollocks?"
"Intel fab prowess is unmatched!"
"Samsung can't get its yields right"
"There will be BT Android tablets, doesn't matter that none have been announced"

Oh man, so much laugh in a single thread.

you're crazy. this soc at best matches baytrail performance in cpu benchmarks and doesnt offer better battery life than the a6!

Lets take a step back and look at reality here. this is apples best attempt to improve on V8 architecture and AT BEST they match cpu performance for baytrail. Now lets look at the merchant apps processor market.

This January Qualcomm announced its krait 300 and krait 400s. We got performance previews in mid June and actual product is only starting to trickle out. If they follow that timing and announce a v8 architechture then 12 months from now we'll have to compare krait with v8 vs Airmont which will again be a bloodbath.

Looking at the IP side of the equation. ARM introduced V8 in mid 2012, 18 months later we have a7. What have they introduced since then? a12 which is laughable. They have nothing in the pipe they have announced that refines on v8. Baytrail itself will be competitive vs V8 ,
 
141142wc8bzwhzlyjdjod8.png

So, a 20NM A6 will be out at the end of next year! That will mean even more CPU performance and even more GPU performance,and that is assuming they do no architectural modifications too,and just increase frequencies and increase the amount of SIMDs for the GPU.
 
Last edited:
Looking at the IP side of the equation. ARM introduced V8 in mid 2012, 18 months later we have a7. What have they introduced since then? a12 which is laughable. They have nothing in the pipe they have announced that refines on v8. Baytrail itself will be competitive vs V8 ,
You're mixing architecture with CPU IP. You should read about ARM 😉
 
Dude, Bay Trail benchmark reign lasted barely 1 week and lost to an ARM custom SoC like the A7. A damn 2 core SoC and the one for the next iPad remains to be seen.

What was all this fuss about, seriously?

lol. The A7 wins exactly 1 benchmark and none of them are multi-threaded. Baytrail-T is a quad-core. I admit that single-threaded performance is more important but it's more or less a tie there while BT would destroy A7 in a multi-threaded benchmark since it has double the cores.

And we do not have any battery life numbers yet but due to OS these will also not be fully comparable.
 
lol. The A7 wins exactly 1 benchmark and none of them are multi-threaded. Baytrail-T is a quad-core. I admit that single-threaded performance is more important but it's more or less a tie there while BT would destroy A7 in a multi-threaded benchmark since it has double the cores.

And we do not have any battery life numbers yet but due to OS these will also not be fully comparable.

This. The fact that the A7 comes close to BT-T performance @ ST is impressive, but BT-T delivers that kind of ST with twice the number of cores, I doubt even a slightly higher-clocked A7X would come close to it in MT tasks. We need more benchmarks to compare both but I'm guessing a >2GHz dual-core Silvermont would still offer better single-thread performance (overall) than the A7, which means Merrifield (2C Silvermont + Rogue GPU) would be competitive in both CPU and GPU fronts.
 
Last edited:
lol. The A7 wins exactly 1 benchmark and none of them are multi-threaded. Baytrail-T is a quad-core. I admit that single-threaded performance is more important but it's more or less a tie there while BT would destroy A7 in a multi-threaded benchmark since it has double the cores.

And we do not have any battery life numbers yet but due to OS these will also not be fully comparable.

wow, are you sure about that? very funny how there is a "but" there once it is a result you disagree with.
 
Back
Top