Enjoy while it last because the rumour said that Kyro would be the last custom core from Qualcomm. If ARM were to release a new arch to replace v8-a or the newer cortex were to much better than Kyro, then we're back to many cores madness.
Enjoy while it last because the rumour said that Kyro would be the last custom core from Qualcomm. If ARM were to release a new arch to replace v8-a or the newer cortex were to much better than Kyro, then we're back to many cores madness.
So you're telling me that the 950 is faster in single-core and web browsing performance than the A9? I understand that Twister is much larger than A72, but there's no catching it in terms of IPC. When the A10 drops with improved IPC and higher clocks, there's no way A73 should be able to beat it.
Surely you get that Apple has their own closed architecture that they totally control. They control everything from the hardware, to the drivers, to the API's and even the benchmarking apps. Open architecture is entirely different especially when benchmarking. Doing it Apple's way is "a way" to go, but it benefits Apple and no-one else. Apples stuff is fast, good CPU's and they "benchmark" extremely fast for sure (in the closed architecture)... But this is a discussion about the tech involved and the advantages of it not an "Apple is best because I say so" argument. Its closed and therefore cant be benchmarked in the same manor. All of todays high end chips, the A9, SD820, Exynos 8890, Kirin 950, even Mediatek x20 are all insanely fast (Girls you are ALL pretty, now stop arguing).
Retrospooty, you are a fool. Apple's approach benefits anyone who buys an iOS device. The excessively multi-core solution benefits almost no one who buys an android phone, except for people who like running multithreaded benchmarks.
Apple's way of using a few beefy cores isn't better just because they say so, or even because they have tight control over both software and hardware. Some things just don't parallelize well, and many of those that do are better run on the GPU. That's not to say that having lots of cores on a relatively inexpensive SoC isn't useful for some people, it is, though generally not in a phone.
Most of the android market is SoCs that integrate licensed ARM IP for CPU cores, not original designs. ARMs designs have tended not to push the edge of performance, because doing so would undermine their architecture licensees. The other reason is that one of their architecture licensees, Apple, invested much more aggressively than the others (particularly Qualcomm) and caught the others off guard in-terms of both performance, and the transition to 64-bit.
Now, though, Qualcomm is back in the game, Samsung has joined it, and it seems that ARM might be a bit more aggressive than they have in the past. Why are more companies developing big cores for ARM? It's not because of doing it Apple's way is "a way" to go, its because it is THE way to go, because, as explained, a lot of things don't parallelize well, and many of those that do are better run on the GPU.
You are missing what I am saying. The discussion is on CPU's and benchmarking cross platform isnt really accurate. The speed benefits of tight integration are more from the tight integration, not from the CPU itself. Its a good CPU, made to run much better based on the tight integration and control. The end result is very fast, but as I mentioned they are all fast in real world use.There are drawbacks to it as well. The extreme lack of options being the biggest.
Agreed, and also an 8 core ARM cpu isnt really running 8 cores in a standard sense. It uses big.little and the 4 high power ones are used when speed is needed and the low power ones are used for less important ops and when the device is idle. People that say "we dont need 8 cores on a phone" simply dont understand the tech involved and how it works.
In real life usage it was very hard to see any diffidence between the kirin 650 and s801AC (2.46GHz). Antutu score was pretty close at 56k vr 61k as well.
Hmmm I think A9 is still appreciably faster than anything else, though of course it's hard to compare iOS and Android meaningfully. But the iPhones seem to get a lot done, quickly, with smaller batteries.
Apple has their own closed architecture that they totally control. They control everything from the hardware, to the drivers, to the API's and even the benchmarking apps. Open architecture is entirely different especially when benchmarking. Doing it Apple's way is "a way" to go and it benefits... Well, it only benefits Apple and no-one else. Apples stuff "benchmarks" fast at single core ops for sure... But calling it "appreciably faster" is a bit of a stretch. It benchmarks very well.
I love this viewpoint so much that I'm going to completely disagree with it and ask why on earth you posted it?
The model SoC is still the OMAP4470, and the fact that no one has imitated their strategy is a crying shame. A small number of comparatively-big cores, and two ultra-lightweight processors for select common, lightweight tasks.
Cores above four on a phone is just spec masturbation. There's no reason why a goddamn phone would need up to two times the amount of cores you need on a computer for good performance. It's no coincidence Qualcomm went back to four cores after the 810 disaster. Speaking of which, according to you the 820 should underperform AND consume more battery compared to the 810 (or the 808, or, hell, any hexa/octa/deca (!) core SoC), which is clearly not the case
And what exactly is that "most powerful, most power efficient ARM processor"? Kirin 950? A9 is faster than that. Exynos 8890? Until power measurements are posted, I'm pretty sure Kirin 950 is still more power efficient.
What about the efficiency? SD820 lacks the efficiency of Exynos 8890. Having used both versions of S7, it was clear as day and night that Exynos version was superior in every way where it mattered in day to day use. All that theoretical superiority in some benchmarks didnt result in better experience when using the phone. In fact it was the other way around. At the end of the day, SD82x is vast improvement over 810 but having used multiple devices powered by SD820 (Mi 5, S7 and OnePlus 3) I can safely say that all the media love that chip received was simply hogwash for end user. Exynos lasts longers, was more snappy on S7 and was great at multitasking. Much better than SD820 version of S7.
What I don't understand is why Samsung are back to dual sourcing. I figured after the SD810 disaster that Samsung would only use Exynos from here on out.
I'd rather have a Snapdragon device running AOSP/CM than an Exynos device running Touchwiz... I'd hazard a guess that the former will be faster than the latter in real world usage, regardless of the Exynos being a technically superior part, too.
So 10% performance improvement for the CPU and 5% for the GPU? Meh. I would've preferred if they used that to "convert" it into 15% less power consumption for the CPU and 10% for the GPU, at the same performance levels.
There's little point in trying to make high-end mobile chips more powerful now. For on-device AI, it's better to have some kind of accelerator anyway. For VR, I guess more powerful GPUs would be nice, but efficiency is just as important, as you don't want your phone to only last 15 mins in VR before it burns.
Depends how its achieved, that increase in clock speed in the new revision could be achieved in the same power budget as the slower original. Bit more performance and a bit more efficiency with the faster race to sleep and lower intermediate voltages.
I don't see how the link shows anything about race to sleep. He just shows that a faster processor uses more power when doing things faster. Unless the battery benchmarks have a set amount of things to do, this is not an article about race to sleep. If you use a 4 core 4GHz 130W CPU full tilt, you will use more power than a 4 core 3GHz 95W CPU. But if you use them for the exact same thing (i.e. encoding a video), the 4c/4G CPU might finish that task faster and use overall less power than the 4c/3G CPU. Not necessarily, but it might.
Lets see how it melts, hard to trust them at this point.
BTW you guys keep promising a proper review for SD820, somehow you've managed to delay it until it was irrelevant.The most value is on day one, delaying it for months and months renders it pointless, you need to do better even if it means splitting it in many relevant bits.
Assuming the 2 versions for SD821 that top at 2.2GHz and 2.4GHz. Note the "variant 2" here "ARM implementer 81 architecture 8 variant 2 part 517 revision 0".
Intel and ARM are approaching the same point from two opposite directions. Intel is lowering the TDP and ARM increasing the TDP. And at that point Intel will be the winner. It could even be AMD. All Intel has to do is to remove some instructions from its x86 set and use the margin somewhere else or given the competition, nowhere at all. These ARM new revisions with no tangible improvement in performance are just aimed at keeping the prices at their current levels. That is all. The mobile market has peaked and the mobile devices can only improve on their communication speed and volume, which is dependent on a lot other than the chip in the mobile device.
Quad core CPU is just enough for smartphones. These Hexa & Octa cores are power hogs. It would also be nice if they can come up with dual core units with 2x Kryo 2.4 GHz.
Retrospooty, you are a fool. Apple's approach benefits anyone who buys an iOS device. The excessively multi-core solution benefits almost no one who buys an android phone, except for people who like running multithreaded benchmarks.
Apple's way of using a few beefy cores isn't better just because they say so, or even because they have tight control over both software and hardware. Some things just don't parallelize well, and many of those that do are better run on the GPU. That's not to say that having lots of cores on a relatively inexpensive SoC isn't useful for some people, it is, though generally not in a phone.
Most of the android market is SoCs that integrate licensed ARM IP for CPU cores, not original designs. ARMs designs have tended not to push the edge of performance, because doing so would undermine their architecture licensees. The other reason is that one of their architecture licensees, Apple, invested much more aggressively than the others (particularly Qualcomm) and caught the others off guard in-terms of both performance, and the transition to 64-bit.
We’ve updated our terms. By continuing to use the site and/or by logging into your account, you agree to the Site’s updated Terms of Use and Privacy Policy.
78 Comments
Back to Article
Chaitanya - Monday, July 11, 2016 - link
would like to see some benchmarks to see how it compares against the chip its upgrading from.piroroadkill - Monday, July 11, 2016 - link
I'm liking the return to quad core CPUs. This 6 and 8 core madness in a goddamn /telephone/ was getting out of hand.Spectrophobic - Monday, July 11, 2016 - link
I'd love a dual core 1.5 GHz Kryo + Adreno 530 @ ~500 MHz.blackice85 - Monday, July 11, 2016 - link
Yeah, it was starting to remind me of the bit wars from console gaming back in the '90s.WorldWithoutMadness - Monday, July 11, 2016 - link
Soon, it will be overtaken by 10 cores madness.Enjoy while it last because the rumour said that Kyro would be the last custom core from Qualcomm. If ARM were to release a new arch to replace v8-a or the newer cortex were to much better than Kyro, then we're back to many cores madness.
Andrei Frumusanu - Monday, July 11, 2016 - link
There's absolutely nothing wrong to 6 and 8-core SoCs. They're faster, more power efficient, actually smaller than for example a Kryo 4-core design.It's unfortunate that readers still have this misinformed and misleading perception of things.
osxandwindows - Monday, July 11, 2016 - link
Faster at what?Multi threaded operations?
retrospooty - Monday, July 11, 2016 - link
YesAndrei Frumusanu - Monday, July 11, 2016 - link
No. Faster in everything.Amandtec - Monday, July 11, 2016 - link
Yes. Scaling on phones is very different to scaling on PC.easp - Monday, July 11, 2016 - link
What amazing luck! Can we use phones as PCs to get this magical scaling benefit?hapeid - Friday, September 30, 2016 - link
Enjoy while it last because the rumour said that Kyro would be the last custom core from Qualcomm. If ARM were to release a new arch to replace v8-a or the newer cortex were to much better than Kyro, then we're back to many cores madness.http://hapeid.com
http://iteote.com
CloudWiz - Monday, July 11, 2016 - link
So you're telling me that the 950 is faster in single-core and web browsing performance than the A9? I understand that Twister is much larger than A72, but there's no catching it in terms of IPC. When the A10 drops with improved IPC and higher clocks, there's no way A73 should be able to beat it.Geranium - Tuesday, July 12, 2016 - link
Apple SoC do good in running Apple optimized benchmark.osxandwindows - Monday, July 11, 2016 - link
Lol, ok.How is it that a dual core cpu (apple a9) equals or exceeds 4 and 8 core ARM cpus and almost everything.
retrospooty - Monday, July 11, 2016 - link
Surely you get that Apple has their own closed architecture that they totally control. They control everything from the hardware, to the drivers, to the API's and even the benchmarking apps. Open architecture is entirely different especially when benchmarking. Doing it Apple's way is "a way" to go, but it benefits Apple and no-one else. Apples stuff is fast, good CPU's and they "benchmark" extremely fast for sure (in the closed architecture)... But this is a discussion about the tech involved and the advantages of it not an "Apple is best because I say so" argument. Its closed and therefore cant be benchmarked in the same manor. All of todays high end chips, the A9, SD820, Exynos 8890, Kirin 950, even Mediatek x20 are all insanely fast (Girls you are ALL pretty, now stop arguing).easp - Monday, July 11, 2016 - link
Retrospooty, you are a fool. Apple's approach benefits anyone who buys an iOS device. The excessively multi-core solution benefits almost no one who buys an android phone, except for people who like running multithreaded benchmarks.Apple's way of using a few beefy cores isn't better just because they say so, or even because they have tight control over both software and hardware. Some things just don't parallelize well, and many of those that do are better run on the GPU. That's not to say that having lots of cores on a relatively inexpensive SoC isn't useful for some people, it is, though generally not in a phone.
Most of the android market is SoCs that integrate licensed ARM IP for CPU cores, not original designs. ARMs designs have tended not to push the edge of performance, because doing so would undermine their architecture licensees. The other reason is that one of their architecture licensees, Apple, invested much more aggressively than the others (particularly Qualcomm) and caught the others off guard in-terms of both performance, and the transition to 64-bit.
Now, though, Qualcomm is back in the game, Samsung has joined it, and it seems that ARM might be a bit more aggressive than they have in the past. Why are more companies developing big cores for ARM? It's not because of doing it Apple's way is "a way" to go, its because it is THE way to go, because, as explained, a lot of things don't parallelize well, and many of those that do are better run on the GPU.
retrospooty - Tuesday, July 12, 2016 - link
You are missing what I am saying. The discussion is on CPU's and benchmarking cross platform isnt really accurate. The speed benefits of tight integration are more from the tight integration, not from the CPU itself. Its a good CPU, made to run much better based on the tight integration and control. The end result is very fast, but as I mentioned they are all fast in real world use.There are drawbacks to it as well. The extreme lack of options being the biggest.tuxRoller - Tuesday, July 12, 2016 - link
In what way are the 2-3 wide cores "big"?Geranium - Thursday, July 14, 2016 - link
Apple has no nown architechure, they just refine ARM's work.tipoo - Tuesday, July 19, 2016 - link
Apple controls the benchmarking apps? Lol. Here, have some tinfoil for your hat.patrickjp93 - Monday, July 11, 2016 - link
No, just multithreaded operations. Apple is still destroying the competition in single-threaded everything.Geranium - Tuesday, July 12, 2016 - link
Nokia S40 like operating will score very high even running on Cortex-A53.retrospooty - Monday, July 11, 2016 - link
Agreed, and also an 8 core ARM cpu isnt really running 8 cores in a standard sense. It uses big.little and the 4 high power ones are used when speed is needed and the low power ones are used for less important ops and when the device is idle. People that say "we dont need 8 cores on a phone" simply dont understand the tech involved and how it works.Ariknowsbest - Monday, July 11, 2016 - link
In real life usage it was very hard to see any diffidence between the kirin 650 and s801AC (2.46GHz). Antutu score was pretty close at 56k vr 61k as well.But on the other hand i prefer big custom cores.
retrospooty - Monday, July 11, 2016 - link
Yup, they are all plenty fast. The real life difference between the Kirin, SD820, Exynos 8890, and Apple A9 is basically nothing.Meteor2 - Monday, July 11, 2016 - link
Hmmm I think A9 is still appreciably faster than anything else, though of course it's hard to compare iOS and Android meaningfully. But the iPhones seem to get a lot done, quickly, with smaller batteries.retrospooty - Monday, July 11, 2016 - link
Apple has their own closed architecture that they totally control. They control everything from the hardware, to the drivers, to the API's and even the benchmarking apps. Open architecture is entirely different especially when benchmarking. Doing it Apple's way is "a way" to go and it benefits... Well, it only benefits Apple and no-one else. Apples stuff "benchmarks" fast at single core ops for sure... But calling it "appreciably faster" is a bit of a stretch. It benchmarks very well.TheinsanegamerN - Thursday, July 14, 2016 - link
It also benefits developers, who have a more consistent market to work with, and the consumer, who get a higher quality end product.easp - Monday, July 11, 2016 - link
You demonstrate again that the one who doesn't understand the tech involved and how it works is you.jerrylzy - Tuesday, July 12, 2016 - link
And you clearly don't know how difficult scheduling is on a big.LITTLE platform.more-or-less - Monday, July 11, 2016 - link
http://browser.primatelabs.com/geekbench3/7255309Any details on Exynos 8893 yet?
GXCoder - Monday, July 11, 2016 - link
It's Exynos 8890Eden-K121D - Tuesday, July 12, 2016 - link
A higher clocked versionlmcd - Monday, July 11, 2016 - link
I love this viewpoint so much that I'm going to completely disagree with it and ask why on earth you posted it?The model SoC is still the OMAP4470, and the fact that no one has imitated their strategy is a crying shame. A small number of comparatively-big cores, and two ultra-lightweight processors for select common, lightweight tasks.
tuxRoller - Monday, July 11, 2016 - link
They're smaller b/c, to a chip, they use absurdly underpowered GPUs.alvareo - Wednesday, July 13, 2016 - link
Cores above four on a phone is just spec masturbation. There's no reason why a goddamn phone would need up to two times the amount of cores you need on a computer for good performance. It's no coincidence Qualcomm went back to four cores after the 810 disaster. Speaking of which, according to you the 820 should underperform AND consume more battery compared to the 810 (or the 808, or, hell, any hexa/octa/deca (!) core SoC), which is clearly not the casefanofanand - Monday, July 11, 2016 - link
You mean like the already-released Helio X20 decacore processor?lilmoe - Monday, July 11, 2016 - link
ANDDDDD the most powerful, most power efficient ARM processor happens to be an eight-core "mad" monster.Keep drinking the cool-aid guys. It's good for you.
TheinsanegamerN - Thursday, July 14, 2016 - link
So why is qualcomm now making a quad core, and why does apple rule the roost with a dual core design?CloudWiz - Saturday, July 16, 2016 - link
And what exactly is that "most powerful, most power efficient ARM processor"? Kirin 950? A9 is faster than that. Exynos 8890? Until power measurements are posted, I'm pretty sure Kirin 950 is still more power efficient.jeanforster - Monday, October 31, 2016 - link
Why not just make the MSM8996 Pro capable of individual page erases instead of blocks as problem solved?SydneyBlue120d - Monday, July 11, 2016 - link
Maybe HEVC encoding enabled?Eden-K121D - Monday, July 11, 2016 - link
So basically a higher clocked SD 820 with some tweaks and tuning to make it better.I hope they improve the ISPNallaikkumaran - Monday, July 11, 2016 - link
Snapdragon SucksTheMysteryMan11 - Monday, July 11, 2016 - link
What about the efficiency?SD820 lacks the efficiency of Exynos 8890. Having used both versions of S7, it was clear as day and night that Exynos version was superior in every way where it mattered in day to day use. All that theoretical superiority in some benchmarks didnt result in better experience when using the phone. In fact it was the other way around.
At the end of the day, SD82x is vast improvement over 810 but having used multiple devices powered by SD820 (Mi 5, S7 and OnePlus 3) I can safely say that all the media love that chip received was simply hogwash for end user. Exynos lasts longers, was more snappy on S7 and was great at multitasking. Much better than SD820 version of S7.
jimjamjamie - Monday, July 11, 2016 - link
Is it really that surprising though that the Samsung SoC performs better in a Samsung phone?syxbit - Monday, July 11, 2016 - link
What I don't understand is why Samsung are back to dual sourcing. I figured after the SD810 disaster that Samsung would only use Exynos from here on out.shabby - Tuesday, July 12, 2016 - link
Maybe because qualcomm has a better cdma modem.TheinsanegamerN - Thursday, July 14, 2016 - link
verizon and sprint ruining it for us againAzurael - Tuesday, July 12, 2016 - link
I'd rather have a Snapdragon device running AOSP/CM than an Exynos device running Touchwiz... I'd hazard a guess that the former will be faster than the latter in real world usage, regardless of the Exynos being a technically superior part, too.TheinsanegamerN - Thursday, July 14, 2016 - link
This is true. now if samsung could stop making 28 versions of the same phone.tuxRoller - Tuesday, July 12, 2016 - link
It's not snapdragon, it's samsung:)Look at the one plus 3.
GXCoder - Monday, July 11, 2016 - link
The frequency is the same as Apple A10, double hurricane cores,up to 2.4GHz.Krysto - Monday, July 11, 2016 - link
So 10% performance improvement for the CPU and 5% for the GPU? Meh. I would've preferred if they used that to "convert" it into 15% less power consumption for the CPU and 10% for the GPU, at the same performance levels.There's little point in trying to make high-end mobile chips more powerful now. For on-device AI, it's better to have some kind of accelerator anyway. For VR, I guess more powerful GPUs would be nice, but efficiency is just as important, as you don't want your phone to only last 15 mins in VR before it burns.
frostyfiredude - Monday, July 11, 2016 - link
Depends how its achieved, that increase in clock speed in the new revision could be achieved in the same power budget as the slower original. Bit more performance and a bit more efficiency with the faster race to sleep and lower intermediate voltages.ikjadoon - Monday, July 11, 2016 - link
"Race to sleep" - I thought this had been debunked, by no one other than Anand himself:http://www.anandtech.com/show/7113/2013-macbook-ai...
Death666Angel - Tuesday, July 12, 2016 - link
I don't see how the link shows anything about race to sleep. He just shows that a faster processor uses more power when doing things faster. Unless the battery benchmarks have a set amount of things to do, this is not an article about race to sleep. If you use a 4 core 4GHz 130W CPU full tilt, you will use more power than a 4 core 3GHz 95W CPU. But if you use them for the exact same thing (i.e. encoding a video), the 4c/4G CPU might finish that task faster and use overall less power than the 4c/3G CPU. Not necessarily, but it might.tuxRoller - Tuesday, July 12, 2016 - link
If he actually debunked race to idle he'd have won an acm award.Well, a minor one:)
Jigolo - Monday, July 11, 2016 - link
Hopefully it's more power efficienthttp://www.phonearena.com/news/Galaxy-S7-battery-l...
http://www.phonearena.com/news/Galaxy-S7-battery-l...
UltraWide - Monday, July 11, 2016 - link
LOL clickbaitAlexey291 - Monday, July 11, 2016 - link
Yup business as usual these days...colinisation - Monday, July 11, 2016 - link
Joshua / Andrei,Were there not rumours of another revision - Snapdragon 823? Or is that the Samsung manufactured variant.
Also what happened to the L3 Cache, heard any whispers of it being enabled or removed entirely?
Thanks
JoshHo - Monday, July 11, 2016 - link
L3 on the S820 was not shipped due to cost. Hard to say whether Snapdragon 821 will ship with an L3.jjj - Monday, July 11, 2016 - link
Lets see how it melts, hard to trust them at this point.BTW you guys keep promising a proper review for SD820, somehow you've managed to delay it until it was irrelevant.The most value is on day one, delaying it for months and months renders it pointless, you need to do better even if it means splitting it in many relevant bits.
kingmustard - Monday, July 11, 2016 - link
The 821 is a higher-clocked 820, that's all. Same GPU (Adreno 530), same DSP (Hexagon 680), same memory bandwidth etc.It's the 830 (with its Adreno 540) that I'm looking forward to getting into my devices.
Stochastic - Monday, July 11, 2016 - link
Any guesses as to when 821 and 830 SoCs will begin shipping in devices? I'm assuming we'll have to wait until 2017 for mainstream phones in the US.jjj - Monday, July 11, 2016 - link
Looks like the small cores might be at 1996MHz and 2188MHz https://browser.primatelabs.com/geekbench3/search?...https://browser.primatelabs.com/geekbench3/search?...
Assuming the 2 versions for SD821 that top at 2.2GHz and 2.4GHz.
Note the "variant 2" here "ARM implementer 81 architecture 8 variant 2 part 517 revision 0".
HideOut - Monday, July 11, 2016 - link
Well now we know what cpu the Note 7 will have this fall.versesuvius - Tuesday, July 12, 2016 - link
Intel and ARM are approaching the same point from two opposite directions. Intel is lowering the TDP and ARM increasing the TDP. And at that point Intel will be the winner. It could even be AMD. All Intel has to do is to remove some instructions from its x86 set and use the margin somewhere else or given the competition, nowhere at all. These ARM new revisions with no tangible improvement in performance are just aimed at keeping the prices at their current levels. That is all. The mobile market has peaked and the mobile devices can only improve on their communication speed and volume, which is dependent on a lot other than the chip in the mobile device.Death666Angel - Tuesday, July 12, 2016 - link
ARM margins are not Intel margins and probably never will be. They will continue to co-exist.tipoo - Tuesday, July 19, 2016 - link
"All Intel has to do is to remove some instructions from its x86 set"Sounds so simple. Look at how big the ucode and x86 decode are.
http://i.imgur.com/OKs8Qs2.jpg
speculatrix - Tuesday, July 12, 2016 - link
I'm sure Qualcomm use the name kyro not kyro. The article mostly used the latter. Commenters using eitherhp79 - Tuesday, July 12, 2016 - link
Cry-YoJoshHo - Wednesday, July 13, 2016 - link
https://www.qualcomm.com/news/snapdragon/2015/09/0...It's definitely Kryo.
jedih - Tuesday, July 12, 2016 - link
Quad core CPU is just enough for smartphones. These Hexa & Octa cores are power hogs. It would also be nice if they can come up with dual core units with 2x Kryo 2.4 GHz.7beauties - Friday, August 5, 2016 - link
Friends, beware of the promoted stories section at the bottom of AnandTech reviews. You may encounter ransomware like I did.hapeid - Monday, November 21, 2016 - link
Retrospooty, you are a fool. Apple's approach benefits anyone who buys an iOS device. The excessively multi-core solution benefits almost no one who buys an android phone, except for people who like running multithreaded benchmarks.Apple's way of using a few beefy cores isn't better just because they say so, or even because they have tight control over both software and hardware. Some things just don't parallelize well, and many of those that do are better run on the GPU. That's not to say that having lots of cores on a relatively inexpensive SoC isn't useful for some people, it is, though generally not in a phone.
Most of the android market is SoCs that integrate licensed ARM IP for CPU cores, not original designs. ARMs designs have tended not to push the edge of performance, because doing so would undermine their architecture licensees. The other reason is that one of their architecture licensees, Apple, invested much more aggressively than the others (particularly Qualcomm) and caught the others off guard in-terms of both performance, and the transition to 64-bit.
http://www.apkxmod.info - http://gadgetgaul.info