I really interested to understand, how a 64bit process speedup the application unless it has the sufficient registry locations[RAM] provided. after all the apple phone has 1GB RAM. A 32bit process is morethan enough to handle this. by the way how much u got )
I think many critics of A7 were jaded by Microsoft's slow and awkward transition from 32 to 64 bit, as well Intel's insistence of a separate product architecture for the 64-bit processor initially. Many people also don't understand the difference between 64-bit address space (memory) and 64-bit instruction set (code).
Either they don't understand or they are working for the marketing units of Samsung, Motorola, MS or whomever. Some of the posts seen already in this thread are so full of ignorance it is hard to believe people are that stupid. That is why I suspect marketing efforts from the competition.
As noted elsewhere the 64 bit address space offers Apple the chance to do things with pointers leading to much higher efficiencies in execution for certain tasks. So even if the RAM isn't in the device, the addressing capabilities can be put to use in ways that enhance performance. So one can't realistically knock the address space expansion.
It is the combination of instruction set and greatly enhanced architecture here though that is making the biggest contribution to the performance gains.
Because Apple controls the processor architecture as well the OS and the App Store ecosystem, only Apple can pull this off and establish it in what seems like a short time frame.
That is part of it but if you look at the failure of MicroSoft over the years a lot of their short comings have been due to poor management. The management team at Apple seems to have an ability to get everybody on the same page, working efficiently towards a common goal. Let's face it both MS and Apple have extremely smart people working for them, so why does the software from MS suck so much? All those items you listed above are important but leveraging them takes real skill.
Someone with true vision will look at an advance in processor power and think "what new feature could I implement now that I have this much power at my disposal."
While others will go "meh, my phone is already fast enough - it's a waste having that much power in a phone."
Funny but I've seen those exact words in several different forums, at least the "my phone is already fast enough" part.
Which of these two people would make a good App developer?
Which of those people even have a chance in hell of creating something, new or innovative. One is the mentality of a burger flipper and the other an innovator or creator.
I'm going iPad 5 all the way, and my expectation is A7X, 2GB RAM, and 128GB storage. It will be a beast as tablets go, but lack 802.11 ac, 4K screen and TB2 out.
Next year, A8X will cover 802.11ac and 4K screen. A9X brings TB2 to mobile and we will probably see an iOS Air.
"As a result, Apple's iPhone 5s delivers performance equal or better performance to Samsung's latest large phablet, despite the Note 3 being equipped with twice the system RAM, a system clock running twice as fast and a battery over twice as large (3,200 mAh vs 1570 mAh in the iPhone 5s). It's not only faster (above), but vastly more efficient, allowing iPhone 5s to beat the Note 3 in battery life when browsing the web over LTE (below). "
Without taking the massive screen difference into consideration, this is just pure fanboy speak.
Not really because we have no idea what the average power draw is on that screen compared to Apples screen. Further if the tester really wanted too each device could be set to drive the screen at the same power level.
As an example the IP5s has a 5wh battery, the iPad 4 has a 42wh battery. They both (on LTE) last about the same time on web browsing test. Does this mean that the IP5 is vastly more efficient than the iPad 4?
As an overall system it most certainly is. This should be obvious really. From an efficiency standpoint most of the Phablets are a joke which really explains why the iPhone 5s does so well when stacked up against them.
It does but you're comparing apples to oranges. The screen is a huge differentiator.
The fact that the screen uses more power only serves to lower overall efficiency which is why Apple is feeding so much money to display developers to do a better screen. If Apple cuts the iPad screens power usage by 50% that frees up a massive amount of power to drive enhanced GPUs and CPU cores. In the end it is overall efficiency that drives the market
This "vastly more efficient" device gets an extra 20 minutes after 8 hrs of use.
You don't consider that a huge accomplishment considering everything delivered in iPhone 5s? I really don't see your argument here.
64 bit is a bit of a gimmick, what really matters is the ARMv8 support.... But try selling that to customers....
Only problem I see here is fragmentation...If the difference is this large, won't we get 5s only apps ?
Some of the gains are from the 128-bit (Edit: corrected from 64-bit) wide registers, some from the ARMv8 instruction set, and some from improving the processor's architecture in other ways. For (lots) more technical detail, read http://www.mikeash.com/pyblog/friday-qa-2013-09-27-arm64-and-you.html
As far as fragmentation goes, the 64-bit chip will still run 32-bit apps natively. However apps that actually require the A7's full power, such as these audio processing apps, of course may not be usable on older hardware.
You don't consider that a huge accomplishment considering everything delivered in iPhone 5s? I really don't see your argument here.
I never said it wasn't a huge accomplishment, that's not my point. I'm arguing that the comment that "5s is vastly more efficient than note 3" is not quantifiable based on the benchmarks/battery sizes the author showed, without considering the considerable power consumption of the much larger and higher resolution note screen.
Further if the tester really wanted too each device could be set to drive the screen at the same power level.
Brightness level maybe, but explain to me how a black box tester can make a screen of different size and technology use the same power as a screen 1/2 the area with 1.3 million less pixels?
This is old but illustrative... IP4 vs IP5 vs Galaxy S3
The Galaxy S3s larger and differing technology (OLED vs LCD) result in over 2x power efficiency.
I never said it wasn't a huge accomplishment, that's not my point. I'm arguing that the comment that "5s is vastly more efficient than note 3" is not quantifiable based on the benchmarks/battery sizes the author showed, without considering the considerable power consumption of the much larger and higher resolution note screen.
People can argue about the overall phone, but there's no question the A7 is vastly more efficient than the Snapdragon 800 in the Note 3.
Qualcomm's chief marketing officer Anand Chandrasekher, for example, recently told the media, "there's a lot of noise because Apple did [64-bit] on their A7. I think they are doing a marketing gimmick
..the sad part is that guy has a CS and Engineering degrees
He started his career at Intel in 1988 after graduating from Cornell University with a bachelor%u2019s degree in Computer Science and a Master%u2019s Degree in Engineering.
guess he drank marketing kool-aid and forgot his technical roots.
This article leaves me wondering if it might be possible to put an A7 chip into the next generation of MBAs and get a battery life in excess of twenty hours. That would burn the competition socks off.
This article leaves me wondering if it might be possible to put an A7 chip into the next generation of MBAs and get a battery life in excess of twenty hours. That would burn the competition socks off.
I don't think the processor is the dominant power eater in the MBA. Likely to get the same performance the A7 would need to run at a faster clock speed which might negate any power savings.
Interestingly Anandtech compares the A7 cpu power with the first macbook Air. The first MBA had battery life of around 4 hours compared to the new ones with 11 hrs, but there's a lot of variables here.
Not sure I buy all the Note 3 battery 2x times IP5s etc etc. Pretty much everyone knows that the display is the biggest drain on battery any you can be pretty sure that the note 3's 5.7in screen is a major juice sucker.
Well you can decide whether you want to compare the power & energy performance of iPhone 5s against a phone like the Galaxy S4 (slower, but also not Samsung’s fastest) or the Note 3 (bigger, huge battery, fastest chip). In performance and battery life, Apple wins either way.
You’re doing the Windows/Android comparison game, where you pick the best specs available and the best price obtainable and compare them to Apple’s product, so you can either say “Cheaper!” or “Faster!” Well on mobile devices, you can add in “Lasts Longer!”
The thing about engineering, though, is that there are tradeoffs. That huge semi-tablet screen certainly does impact battery life. It also affords it the RV room for a huge battery. But as a package, it’s not faster nor longer lasting, nor is it compact and nice looking. But there’s no question that despite all the cores and clock speed and battery that it’s simply not faster nor as capable, and it doesn’t last longer.
If you’re saying a big screen is more important to you personally than a fast capable device with battery life, that’s fine. But don’t confuse your ability to imagine an opinion with the overall desires of the market. The Note 3 isn’t outselling iPhone 5s. Samsung isn’t even ever going to say how many it sells, because sales wouldn’t be a flattering comparison either.
Well you can decide whether you want to compare the power & energy performance of iPhone 5s against a phone like the Galaxy S4 (slower, but also not Samsung’s fastest) or the Note 3 (bigger, huge battery, fastest chip). In performance and battery life, Apple wins either way.
You’re doing the Windows/Android comparison game, where you pick the best specs available and the best price obtainable and compare them to Apple’s product, so you can either say “Cheaper!” or “Faster!” Well on mobile devices, you can add in “Lasts Longer!”
Keeping it all Apple then. The iPad 4 has a battery with 9 times, yes 9X more capacity than the 5s yet has the same (approximate on LTE) battery life. Does it mean the A7 is 9 times more efficient than the A6X.
Brightness level maybe, but explain to me how a black box tester can make a screen of different size and technology use the same power as a screen 1/2 the area with 1.3 million less pixels?
This is old but illustrative... IP4 vs IP5 vs Galaxy S3
The Galaxy S3s larger and differing technology (OLED vs LCD) result in over 2x power efficiency.
The Galaxy S III has an inferior display on top of chowing down three times the power of iPhone 5. Your source also says GS3 has half the brightness, suffers PenTile's Red/Blue phony-resolution, and what DisplayMate calls "poorer image contrast and screen readability in high ambient."
"Samsung has not bothered to calibrate the Color Gamut on any of its OLED displays, so they are wildly inaccurate and produce inaccurate and over saturated colors."
The Galaxy S III has an inferior display on top of chowing down three times the power of iPhone 5. Your source also says GS3 has half the brightness, suffers PenTile's Red/Blue phony-resolution, and what DisplayMate calls "poorer image contrast and screen readability in high ambient."
"Samsung has not bothered to calibrate the Color Gamut on any of its OLED displays, so they are wildly inaccurate and produce inaccurate and over saturated colors."
Perhaps you don’t get it: iPad isn’t a phone. Note 3 & GS 4 are phones.
You think comparing Apple’s fastest phone against Samsung’s is unfair due to Samsung using a larger screen in a desperate attempt to differentiate, then try to pit the iPhone against the iPad? Why not just compare a Mac Pro to an iPod shuffle?
You have no point, you’re just arguing to keep the subject changed so you don’t have to admit all your complaints are absurd.
Perhaps you don’t get it: iPad isn’t a phone. Note 3 & GS 4 are phones.
You think comparing Apple’s fastest phone against Samsung’s is unfair due to Samsung using a larger screen in a desperate attempt to differentiate, then try to pit the iPhone against the iPad? Why not just compare a Mac Pro to an iPod shuffle?
You have no point, you’re just arguing to keep the subject changed so you don’t have to admit all your complaints are absurd.
What planet do you live on. An iPad LTE is a lot more like an iPhone than any other comparison you care to pull out of your hat.
And don't put words in my mouth. I didn't say comparing 5s to note was unfair. I said your assertion that comparing the 5s efficiency to a note 3 where the battery life and performance are similar (5s being slightly ahead) is somehow magical because 5s has half the battery capacity is useless because the huge difference in screen is a major factor in battery consumption. Got it now?
64 bit is a bit of a gimmick, what really matters is the ARMv8 support.... But try selling that to customers....
Only problem I see here is fragmentation...If the difference is this large, won't we get 5s only apps ?
No fragmentation wont be an issue. Because of fat binaries in X code. Also because of ARMv8's new ISA, the A7 supports AArch 32bit and AArch 64 bit directly. Which means to the end programmer and user 32 bit and 64 bit code can be run separately or both side by side in the A7.
Comments
)
As noted elsewhere the 64 bit address space offers Apple the chance to do things with pointers leading to much higher efficiencies in execution for certain tasks. So even if the RAM isn't in the device, the addressing capabilities can be put to use in ways that enhance performance. So one can't realistically knock the address space expansion.
It is the combination of instruction set and greatly enhanced architecture here though that is making the biggest contribution to the performance gains.
That is part of it but if you look at the failure of MicroSoft over the years a lot of their short comings have been due to poor management. The management team at Apple seems to have an ability to get everybody on the same page, working efficiently towards a common goal. Let's face it both MS and Apple have extremely smart people working for them, so why does the software from MS suck so much? All those items you listed above are important but leveraging them takes real skill.
Which of those people even have a chance in hell of creating something, new or innovative. One is the mentality of a burger flipper and the other an innovator or creator.
I'm going iPad 5 all the way, and my expectation is A7X, 2GB RAM, and 128GB storage. It will be a beast as tablets go, but lack 802.11 ac, 4K screen and TB2 out.
Next year, A8X will cover 802.11ac and 4K screen. A9X brings TB2 to mobile and we will probably see an iOS Air.
I can dream, can't I.
Not really because we have no idea what the average power draw is on that screen compared to Apples screen. Further if the tester really wanted too each device could be set to drive the screen at the same power level. As an overall system it most certainly is. This should be obvious really. From an efficiency standpoint most of the Phablets are a joke which really explains why the iPhone 5s does so well when stacked up against them. The fact that the screen uses more power only serves to lower overall efficiency which is why Apple is feeding so much money to display developers to do a better screen. If Apple cuts the iPad screens power usage by 50% that frees up a massive amount of power to drive enhanced GPUs and CPU cores. In the end it is overall efficiency that drives the market You don't consider that a huge accomplishment considering everything delivered in iPhone 5s? I really don't see your argument here.
If hate to see an AIR with the current limited versions of iOS installed. I'd rather see an ARM based AIR with an open version of Mac OS installed.
64 bit is a bit of a gimmick, what really matters is the ARMv8 support.... But try selling that to customers....
Some of the gains are from the 128-bit (Edit: corrected from 64-bit) wide registers, some from the ARMv8 instruction set, and some from improving the processor's architecture in other ways. For (lots) more technical detail, read http://www.mikeash.com/pyblog/friday-qa-2013-09-27-arm64-and-you.html
As far as fragmentation goes, the 64-bit chip will still run 32-bit apps natively. However apps that actually require the A7's full power, such as these audio processing apps, of course may not be usable on older hardware.
You don't consider that a huge accomplishment considering everything delivered in iPhone 5s? I really don't see your argument here.
I never said it wasn't a huge accomplishment, that's not my point. I'm arguing that the comment that "5s is vastly more efficient than note 3" is not quantifiable based on the benchmarks/battery sizes the author showed, without considering the considerable power consumption of the much larger and higher resolution note screen.
Even this didn't really bother me.
Further if the tester really wanted too each device could be set to drive the screen at the same power level.
Brightness level maybe, but explain to me how a black box tester can make a screen of different size and technology use the same power as a screen 1/2 the area with 1.3 million less pixels?
This is old but illustrative... IP4 vs IP5 vs Galaxy S3
The Galaxy S3s larger and differing technology (OLED vs LCD) result in over 2x power efficiency.
http://www.displaymate.com/Smartphone_ShootOut_2.htm
Display Relative Power Efficiency
same Peak Luminance 556 cd/m2
same 4.0 inch screen size area
0.51 watts
0.74 watts
2.2 watts
This compares the Relative Power Efficiency
by looking at the same screen brightness and
screen area.
..the sad part is that guy has a CS and Engineering degrees
http://en.wikipedia.org/wiki/Anand_Chandrasekher
He started his career at Intel in 1988 after graduating from Cornell University with a bachelor%u2019s degree in Computer Science and a Master%u2019s Degree in Engineering.
guess he drank marketing kool-aid and forgot his technical roots.
This article leaves me wondering if it might be possible to put an A7 chip into the next generation of MBAs and get a battery life in excess of twenty hours. That would burn the competition socks off.
I don't think the processor is the dominant power eater in the MBA. Likely to get the same performance the A7 would need to run at a faster clock speed which might negate any power savings.
Interestingly Anandtech compares the A7 cpu power with the first macbook Air. The first MBA had battery life of around 4 hours compared to the new ones with 11 hrs, but there's a lot of variables here.
Not sure I buy all the Note 3 battery 2x times IP5s etc etc. Pretty much everyone knows that the display is the biggest drain on battery any you can be pretty sure that the note 3's 5.7in screen is a major juice sucker.
Well you can decide whether you want to compare the power & energy performance of iPhone 5s against a phone like the Galaxy S4 (slower, but also not Samsung’s fastest) or the Note 3 (bigger, huge battery, fastest chip). In performance and battery life, Apple wins either way.
You’re doing the Windows/Android comparison game, where you pick the best specs available and the best price obtainable and compare them to Apple’s product, so you can either say “Cheaper!” or “Faster!” Well on mobile devices, you can add in “Lasts Longer!”
The thing about engineering, though, is that there are tradeoffs. That huge semi-tablet screen certainly does impact battery life. It also affords it the RV room for a huge battery. But as a package, it’s not faster nor longer lasting, nor is it compact and nice looking. But there’s no question that despite all the cores and clock speed and battery that it’s simply not faster nor as capable, and it doesn’t last longer.
If you’re saying a big screen is more important to you personally than a fast capable device with battery life, that’s fine. But don’t confuse your ability to imagine an opinion with the overall desires of the market. The Note 3 isn’t outselling iPhone 5s. Samsung isn’t even ever going to say how many it sells, because sales wouldn’t be a flattering comparison either.
Well you can decide whether you want to compare the power & energy performance of iPhone 5s against a phone like the Galaxy S4 (slower, but also not Samsung’s fastest) or the Note 3 (bigger, huge battery, fastest chip). In performance and battery life, Apple wins either way.
You’re doing the Windows/Android comparison game, where you pick the best specs available and the best price obtainable and compare them to Apple’s product, so you can either say “Cheaper!” or “Faster!” Well on mobile devices, you can add in “Lasts Longer!”
Keeping it all Apple then. The iPad 4 has a battery with 9 times, yes 9X more capacity than the 5s yet has the same (approximate on LTE) battery life. Does it mean the A7 is 9 times more efficient than the A6X.
No.
Brightness level maybe, but explain to me how a black box tester can make a screen of different size and technology use the same power as a screen 1/2 the area with 1.3 million less pixels?
This is old but illustrative... IP4 vs IP5 vs Galaxy S3
The Galaxy S3s larger and differing technology (OLED vs LCD) result in over 2x power efficiency.
http://www.displaymate.com/Smartphone_ShootOut_2.htm
Display Relative Power Efficiency
same Peak Luminance 556 cd/m2
same 4.0 inch screen size area
0.51 watts
0.74 watts
2.2 watts
This compares the Relative Power Efficiency
by looking at the same screen brightness and
screen area.
The Galaxy S III has an inferior display on top of chowing down three times the power of iPhone 5. Your source also says GS3 has half the brightness, suffers PenTile's Red/Blue phony-resolution, and what DisplayMate calls "poorer image contrast and screen readability in high ambient."
"Samsung has not bothered to calibrate the Color Gamut on any of its OLED displays, so they are wildly inaccurate and produce inaccurate and over saturated colors."
The Galaxy S III has an inferior display on top of chowing down three times the power of iPhone 5. Your source also says GS3 has half the brightness, suffers PenTile's Red/Blue phony-resolution, and what DisplayMate calls "poorer image contrast and screen readability in high ambient."
"Samsung has not bothered to calibrate the Color Gamut on any of its OLED displays, so they are wildly inaccurate and produce inaccurate and over saturated colors."
What's that got to do with the price of fish?
Perhaps you don’t get it: iPad isn’t a phone. Note 3 & GS 4 are phones.
You think comparing Apple’s fastest phone against Samsung’s is unfair due to Samsung using a larger screen in a desperate attempt to differentiate, then try to pit the iPhone against the iPad? Why not just compare a Mac Pro to an iPod shuffle?
You have no point, you’re just arguing to keep the subject changed so you don’t have to admit all your complaints are absurd.
Perhaps you don’t get it: iPad isn’t a phone. Note 3 & GS 4 are phones.
You think comparing Apple’s fastest phone against Samsung’s is unfair due to Samsung using a larger screen in a desperate attempt to differentiate, then try to pit the iPhone against the iPad? Why not just compare a Mac Pro to an iPod shuffle?
You have no point, you’re just arguing to keep the subject changed so you don’t have to admit all your complaints are absurd.
What planet do you live on. An iPad LTE is a lot more like an iPhone than any other comparison you care to pull out of your hat.
And don't put words in my mouth. I didn't say comparing 5s to note was unfair. I said your assertion that comparing the 5s efficiency to a note 3 where the battery life and performance are similar (5s being slightly ahead) is somehow magical because 5s has half the battery capacity is useless because the huge difference in screen is a major factor in battery consumption. Got it now?
I enjoyed the rest of the article, though. :-)64 bit is a bit of a gimmick, what really matters is the ARMv8 support.... But try selling that to customers....
Only problem I see here is fragmentation...If the difference is this large, won't we get 5s only apps ?
No fragmentation wont be an issue. Because of fat binaries in X code. Also because of ARMv8's new ISA, the A7 supports AArch 32bit and AArch 64 bit directly. Which means to the end programmer and user 32 bit and 64 bit code can be run separately or both side by side in the A7.
Good explanations here, and very technical here.