I don't think Samsung was surprised by Apple's 64 bit architecture (lol, they built it). What caught them off guard was the media hype Apple was able to get out of it. Many tech oriented people questioned why you'd even need it with the state of current phones' hardware (there will be a need in the future).
The 32 bit note is faster than the 64 bit flagship Apple 5s.
Except for when Samsung cheats- then it is a lot faster.
Take a hike bribed troll!
Samsung is digging its own grave.. Through fraud and trolls like you!
Well, technically, they aren't cheating. They are handicapping all non benchmark testing apps. The hardware is, obviously, capable of these speeds in some manner.
But to make the user experience better, they put on these governors. Without them, we get the "cheat" scores...
...ok, I have to go shower now after writing all that...
Sure it is cheating. At best those speeds in continued processing would result in abhorrent battery life, and at worst—and not terribly unlikely over some short span of time—it could result in hardware failure. Benchmarks are only meaningful if they work under the same conditions that any other app would.
Well, technically, they aren't cheating. They are handicapping all non benchmark testing apps. The hardware is, obviously, capable of these speeds in some manner.
But to make the user experience better, they put on these governors. Without them, we get the "cheat" scores...
...ok, I have to go shower now after writing all that...
Actually it is cheating. Benchmarks are meant to test real world performance... that is performance that all other apps are capable of achieving. Clearly that's not the case when Samsung singles out specific apps and sets the CPU into boost mode just for those apps, especially when those apps are mostly benchmarking apps. They can claim that a few other apps make use of this, but those are specialized cases and not indicative of how all other apps will perform.
I don't think Samsung was surprised by Apple's 64 bit architecture (lol, they built it). What caught them off guard was the media hype Apple was able to get out of it. Many tech oriented people questioned why you'd even need it with the state of current phones' hardware (there will be a need in the future).
That's some mighty fine apologetics. See how long you can keep reality at bay by telling yourself that 64-bit doesn't matter. You only need to do that until Samsung/Google can come around with their own 64-bit phone, then you can flip-flop and extol the virtues of 64-bit phones. Just remember who got their first.
Base on which metric? The dual core A7 @ 1.3Ghz beats the Samsung Exynos 8-cores @ 1.8Ghz on both efficiency and performance hands down.
According to the chart in the article, you're wrong. The non-cheating Note scores higher than the 5S in all but one metric. If you look at performance per core or per clock, then the 5S pulls ahead.
The 32 bit note is faster than the 64 bit flagship Apple 5s.
Except for when Samsung cheats- then it is a lot faster.
It clearly scores higher, but that's not the point here.
The A7 is a dual core SoC running at 1.3GHz and it can definitely hold its own against the "real" Note 3 performance, which is a 4 core SoC running at 2.3GHz.
Throw a 4 core A7 running at 2.3GHz up there and see just how "fast" the Note 3 is. An A7 beefed up that much would approach the power of the CPU in the MacBook Air.
In mobile devices, efficiency takes precedence over performance and it is clear that the A7 is a much, much more efficient chip than the Snapdragon 800.
Well, technically, they aren't cheating. They are handicapping all non benchmark testing apps. The hardware is, obviously, capable of these speeds in some manner.
But to make the user experience better, they put on these governors. Without them, we get the "cheat" scores...
...ok, I have to go shower now after writing all that...
Problem is with raising the frequency rapidly decrease the efficiency as it's greatly increase the power consumption and heat. So every mobile SoC has do be throttle to maximize efficiency and current sweat spot is under 1.5 Ghz. With the A7, Apple has made clear choices with a lower frequency and lower cores counts but with greater performance and efficiency then its competitor.
Shameful behavior. Samsung I can believe as that sham of a company decides every time that no level is too low to stoop too.
Fandroids and iHaters show exceptional hypocrisy and shameful denial. Had Apple did this, they would be screaming at the top of their lungs and demanding lawsuits. These jackasses are actually defending Samsung's reasonings. Unfrickenbelievable.
I don't think Samsung was surprised by Apple's 64 bit architecture (lol, they built it). What caught them off guard was the media hype Apple was able to get out of it. Many tech oriented people questioned why you'd even need it with the state of current phones' hardware (there will be a need in the future).
The 32 bit note is faster than the 64 bit flagship Apple 5s.
Except for when Samsung cheats- then it is a lot faster.
I hope it's faster. It's twice as large, the battery is twice as large, it has twice the cores, at twice the clock speed, three times the memory, a fraction of the battery life and it still BARELY beats out the iPhone 5s. All in a package the size of a book.
If the point of the test is to measure *normal* processing power, then any conspicuous throttle-up by one of entrants above and beyond the norm, when it shouldn't be happening under normal operation, *is* artificially inflating the power of that processor.
If it's not cheating, it's using a feature of your processor that will operate in a certain manner under conditions that don't warrant it, for the purpose of getting a higher benchmark.
It's as synonymous with cheating as you can possibly get.
You expect doping scandals in sports but who ever thought they would also show up in smart phone benchmarks.
I hope this story gains a lot of traction beyond just a few Apple sites like this one and makes it into mainstream news outlets as well. Samsung can not get away with lying about how their phone will perform with these amazing benchmark scores and then when people buy and actually use them they fall far short of that.
Talk about a great ad campaign for HTC, LG, or even Microsoft attacking this fraud by Samsung. Apple is above that but those other companies should run with it.
That's some mighty fine apologetics. See how long you can keep reality at bay by telling yourself that 64-bit doesn't matter. You only need to do that until Samsung/Google can come around with their own 64-bit phone, then you can flip-flop and extol the virtues of 64-bit phones. Just remember who got their first.
More precisely, ARM64 programs derive much of their performance boost from the new instructions available only to A64 builds. If ARM had decided instead to take the Intel route of making the new instructions available to 32-bit programs as well, the phrase "64-bit" wouldn't be getting this much attention. That's why the transition to 64-bit computing on the desktop happened without fanfare.
More precisely, ARM64 programs derive much of their performance boost from the new instructions available only to A64 builds. That's why the improved performance is associated with the phrase "64-bit", which technically describes merely the register width. If ARM had decided instead to take the Intel route of making the new instructions available to 32-bit programs as well, the phrase "64-bit" wouldn't be getting this much attention.
This is a non-issue,
There is a lot goings on with the new AArch64 and AArch32 ISA for ARM SoC. here is a good in-depth Q&A
Very informative article, thanks. This summarizes things nicely.
Quote:
The "64-bit" A7 is not just a marketing gimmick, but neither is it an amazing breakthrough that enables a new class of applications. The truth, as happens often, lies in between.
The simple fact of moving to 64-bit does little. It makes for slightly faster computations in some cases, somewhat higher memory usage for most programs, and makes certain programming techniques more viable. Overall, it's not hugely significant.
The ARM architecture changed a bunch of other things in its transition to 64-bit. An increased number of registers and a revised, streamlined instruction set make for a nice performance gain over 32-bit ARM.
If I were developing apps for the Galaxy devices, I would want to enable that speed boost myself. I wonder what happens if it is enabled for all apps all the time? Obviously it saps your battery life but not when the app is idle right? What is the down side to having it on all the time? Does it melt the CPU or something?
Did you even read the benchmarking article? The 5S is much faster than the Note. The Note is sweating to beat the LG. Reading is fundamental.
What is more fundamental than reading is comprehension, or even being able to draw your own conclusions based on data. This is AI and usually is pretty good at spinning the facts to support their team. Yes, if you 'read' the article it makes it sound like the 5s would really like to be faster- so I'm assuming that's what you're making your 'The 5S is much faster than the Note' claim despite the article showing the data that proves otherwise?
If you look at the actual data provided..... The Note 3 was faster in 3 out of the 4 tests, without cheating (the cheating tests are no contest, and without question dubious on Samsung's part). The article references this "essentially tied' but is quick to point out that 'Apple won' in the one metric that it actually did slightly beat the others. Its pretty good kindergarden spin when 'winning is winning, but losing is a tie'
The author is quick to make the tests seem skewed or almost unfair because Apple ran them with 'half the clock speed and half the RAM' and then nicely schmoozes the fact that the test was run on a native 64 bit benchmark instead of the running the same test as the other phones. In my book both tests are perfectly valid- the Apple 5s ships with half the clock speed and half the RAM... and the other phones don't support 64 bit so there you go. But even running the native 64 bit benchmark it was still slower in 3 of the 4 tests as unpopular a result as that might be on this site.
According to the chart in the article, you're wrong. The non-cheating Note scores higher than the 5S in all but one metric. If you look at performance per core or per clock, then the 5S pulls ahead.
lolwut? per core or per clock???? That would be taking spin to new levels. That would be like saying Android devices are twice as fast 'per bus bit.' Both would be true statements... but pointless.
If Apple had done this it would be on CNN, all the national evening news shows, and all over the web within minutes. Has this story gained any traction at all?
Comments
Samsung is digging its own grave.. Through fraud and trolls like you!
The 32 bit note is faster than the 64 bit flagship Apple 5s.
Base on which metric? The dual core A7 @ 1.3Ghz beats the Samsung Exynos 8-cores @ 1.8Ghz on both efficiency and performance hands down.
Sure it is cheating. At best those speeds in continued processing would result in abhorrent battery life, and at worst—and not terribly unlikely over some short span of time—it could result in hardware failure. Benchmarks are only meaningful if they work under the same conditions that any other app would.
Well, technically, they aren't cheating. They are handicapping all non benchmark testing apps. The hardware is, obviously, capable of these speeds in some manner.
But to make the user experience better, they put on these governors. Without them, we get the "cheat" scores...
...ok, I have to go shower now after writing all that...
Actually it is cheating. Benchmarks are meant to test real world performance... that is performance that all other apps are capable of achieving. Clearly that's not the case when Samsung singles out specific apps and sets the CPU into boost mode just for those apps, especially when those apps are mostly benchmarking apps. They can claim that a few other apps make use of this, but those are specialized cases and not indicative of how all other apps will perform.
That's some mighty fine apologetics. See how long you can keep reality at bay by telling yourself that 64-bit doesn't matter. You only need to do that until Samsung/Google can come around with their own 64-bit phone, then you can flip-flop and extol the virtues of 64-bit phones. Just remember who got their first.
According to the chart in the article, you're wrong. The non-cheating Note scores higher than the 5S in all but one metric. If you look at performance per core or per clock, then the 5S pulls ahead.
The 32 bit note is faster than the 64 bit flagship Apple 5s.
Except for when Samsung cheats- then it is a lot faster.
It clearly scores higher, but that's not the point here.
The A7 is a dual core SoC running at 1.3GHz and it can definitely hold its own against the "real" Note 3 performance, which is a 4 core SoC running at 2.3GHz.
Throw a 4 core A7 running at 2.3GHz up there and see just how "fast" the Note 3 is. An A7 beefed up that much would approach the power of the CPU in the MacBook Air.
In mobile devices, efficiency takes precedence over performance and it is clear that the A7 is a much, much more efficient chip than the Snapdragon 800.
Well, technically, they aren't cheating. They are handicapping all non benchmark testing apps. The hardware is, obviously, capable of these speeds in some manner.
But to make the user experience better, they put on these governors. Without them, we get the "cheat" scores...
...ok, I have to go shower now after writing all that...
Problem is with raising the frequency rapidly decrease the efficiency as it's greatly increase the power consumption and heat. So every mobile SoC has do be throttle to maximize efficiency and current sweat spot is under 1.5 Ghz. With the A7, Apple has made clear choices with a lower frequency and lower cores counts but with greater performance and efficiency then its competitor.
Shameful behavior. Samsung I can believe as that sham of a company decides every time that no level is too low to stoop too.
Fandroids and iHaters show exceptional hypocrisy and shameful denial. Had Apple did this, they would be screaming at the top of their lungs and demanding lawsuits. These jackasses are actually defending Samsung's reasonings. Unfrickenbelievable.
I don't think Samsung was surprised by Apple's 64 bit architecture (lol, they built it). What caught them off guard was the media hype Apple was able to get out of it. Many tech oriented people questioned why you'd even need it with the state of current phones' hardware (there will be a need in the future).
The 32 bit note is faster than the 64 bit flagship Apple 5s.
Except for when Samsung cheats- then it is a lot faster.
I hope it's faster. It's twice as large, the battery is twice as large, it has twice the cores, at twice the clock speed, three times the memory, a fraction of the battery life and it still BARELY beats out the iPhone 5s. All in a package the size of a book.
Truly, a marvel.
If the point of the test is to measure *normal* processing power, then any conspicuous throttle-up by one of entrants above and beyond the norm, when it shouldn't be happening under normal operation, *is* artificially inflating the power of that processor.
If it's not cheating, it's using a feature of your processor that will operate in a certain manner under conditions that don't warrant it, for the purpose of getting a higher benchmark.
It's as synonymous with cheating as you can possibly get.
And oh how very Samsung of Samsung.
You expect doping scandals in sports but who ever thought they would also show up in smart phone benchmarks.
I hope this story gains a lot of traction beyond just a few Apple sites like this one and makes it into mainstream news outlets as well. Samsung can not get away with lying about how their phone will perform with these amazing benchmark scores and then when people buy and actually use them they fall far short of that.
Talk about a great ad campaign for HTC, LG, or even Microsoft attacking this fraud by Samsung. Apple is above that but those other companies should run with it.
That's some mighty fine apologetics. See how long you can keep reality at bay by telling yourself that 64-bit doesn't matter. You only need to do that until Samsung/Google can come around with their own 64-bit phone, then you can flip-flop and extol the virtues of 64-bit phones. Just remember who got their first.
More precisely, ARM64 programs derive much of their performance boost from the new instructions available only to A64 builds. If ARM had decided instead to take the Intel route of making the new instructions available to 32-bit programs as well, the phrase "64-bit" wouldn't be getting this much attention. That's why the transition to 64-bit computing on the desktop happened without fanfare.
More precisely, ARM64 programs derive much of their performance boost from the new instructions available only to A64 builds. That's why the improved performance is associated with the phrase "64-bit", which technically describes merely the register width. If ARM had decided instead to take the Intel route of making the new instructions available to 32-bit programs as well, the phrase "64-bit" wouldn't be getting this much attention.
This is a non-issue,
There is a lot goings on with the new AArch64 and AArch32 ISA for ARM SoC. here is a good in-depth Q&A
http://www.mikeash.com/pyblog/friday-qa-2013-09-27-arm64-and-you.html
There is a lot goings on with the new AArch64 and AArch32 ISA for ARM SoC. here is a good in-depth Q&A
http://www.mikeash.com/pyblog/friday-qa-2013-09-27-arm64-and-you.html
Very informative article, thanks. This summarizes things nicely.
The "64-bit" A7 is not just a marketing gimmick, but neither is it an amazing breakthrough that enables a new class of applications. The truth, as happens often, lies in between.
The simple fact of moving to 64-bit does little. It makes for slightly faster computations in some cases, somewhat higher memory usage for most programs, and makes certain programming techniques more viable. Overall, it's not hugely significant.
The ARM architecture changed a bunch of other things in its transition to 64-bit. An increased number of registers and a revised, streamlined instruction set make for a nice performance gain over 32-bit ARM.
If I were developing apps for the Galaxy devices, I would want to enable that speed boost myself. I wonder what happens if it is enabled for all apps all the time? Obviously it saps your battery life but not when the app is idle right? What is the down side to having it on all the time? Does it melt the CPU or something?
Did you even read the benchmarking article? The 5S is much faster than the Note. The Note is sweating to beat the LG. Reading is fundamental.
What is more fundamental than reading is comprehension, or even being able to draw your own conclusions based on data. This is AI and usually is pretty good at spinning the facts to support their team. Yes, if you 'read' the article it makes it sound like the 5s would really like to be faster- so I'm assuming that's what you're making your 'The 5S is much faster than the Note' claim despite the article showing the data that proves otherwise?
If you look at the actual data provided..... The Note 3 was faster in 3 out of the 4 tests, without cheating (the cheating tests are no contest, and without question dubious on Samsung's part). The article references this "essentially tied' but is quick to point out that 'Apple won' in the one metric that it actually did slightly beat the others. Its pretty good kindergarden spin when 'winning is winning, but losing is a tie'
The author is quick to make the tests seem skewed or almost unfair because Apple ran them with 'half the clock speed and half the RAM' and then nicely schmoozes the fact that the test was run on a native 64 bit benchmark instead of the running the same test as the other phones. In my book both tests are perfectly valid- the Apple 5s ships with half the clock speed and half the RAM... and the other phones don't support 64 bit so there you go. But even running the native 64 bit benchmark it was still slower in 3 of the 4 tests as unpopular a result as that might be on this site.
According to the chart in the article, you're wrong. The non-cheating Note scores higher than the 5S in all but one metric. If you look at performance per core or per clock, then the 5S pulls ahead.
lolwut? per core or per clock???? That would be taking spin to new levels. That would be like saying Android devices are twice as fast 'per bus bit.' Both would be true statements... but pointless.
If Apple had done this it would be on CNN, all the national evening news shows, and all over the web within minutes. Has this story gained any traction at all?