[g2gcya]

Are the apps we running 32 bits or 64 bits?

[tickerguy]

32 I believe.

There is no reason to go to 64 unless your address space exceeds 4GB (including memory mapped I/O.) Then there is, but only for that reason; there is no inherent advantage to a 64-bit CPU over a 32-bit one, or 64-bit code over 32-bit.

[badiyee]

should be 32. the ARM architecture used is 32bit.

[Nharzhool]

Are the apps we running 32 bits or 64 bits?

Heads-up, it is 32 bit/64 bit. They are singular. Though as far as I know, the OS and all the apps are 32 bit. This makes sense because the STL100-1 uses TI OMAP 4470 which is 32 bit and the STL100-2-4 use the Snapdragon S4 which is also 32 bit so I suppose that answers your question...

I presume this is because of the recent announcement that the new iPhone is going to have a 64 bit SoC and OS which is quite cool I suppose...if they do it right. It just mustn't be ANYTHING like the 64 bit Windows XP which was just a massive heap of poopoo!

[app_Developer]

As others have said, the CPUs that Blackberry use are 32-bit.

[app_Developer]

There is no reason to go to 64 unless your address space exceeds 4GB (including memory mapped I/O.) Then there is, but only for that reason; there is no inherent advantage to a 64-bit CPU over a 32-bit one, or 64-bit code over 32-bit.

In theory, sort of, but in actual practice wouldn't you agree that historically most 64-bit CPUs, like the A7, have a richer set of instructions which compilers can take advantage. They also often have more registers, which compilers can take advantage of (Apple has confirmed this is true on the A7). And any compiler worth it's salt can at least take advantage of the extra register width with the proper instructions.

So in practice, wouldn't you agree that there are performance improvements to be had in most cases on ARM even with < 4GB? I'm interested to test this myself in a couple of weeks on the A7. I'm sure there will be lots of benchmarks out there, too

[Nharzhool]

In theory, sort of, but in actual practice wouldn't you agree that historically most 64-bit CPUs, like the A7, have a richer set of instructions which compilers can take advantage. They also often have more registers, which compilers can take advantage of (Apple has confirmed this is true on the A7). And any compiler worth it's salt can at least take advantage of the extra register width with the proper instructions.

So in practice, wouldn't you agree that there are performance improvements to be had in most cases on ARM even with < 4GB? I'm interested to test this myself in a couple of weeks on the A7. I'm sure there will be lots of benchmarks out there, too

Yeah, You're right. I'm not sure whether tickerguy knows exactly what the differences are between 64 and 32 bit are.

64 bit will perform better if the OS and apps are written correctly to work with the different architecture.

A short excerpt from the Wiki page:

--------------------------------------------------------------
A common misconception is that 64-bit architectures are no better than 32-bit architectures unless the computer has more than 4 GB of random access memory.[23] This is not entirely true:

Some operating systems and certain hardware configurations limit the physical memory space to 3 GB on IA-32 systems, due to much of the 3–4 GB region being reserved for hardware addressing; see 3 GB barrier; 64-bit architectures can address far more than 4 GB. However, IA-32 processors from the Pentium II onwards allow for a 36-bit physical memory address space, using Physical Address Extension (PAE), which gives a 64 GB physical address range, of which up to 62 GB may be used by main memory; operating systems that support PAE may not be limited to 4GB of physical memory, even on IA-32 processors. However, drivers and other kernel mode software, particularly older versions, may not be compatible with PAE.

Some operating systems reserve portions of process address space for OS use, effectively reducing the total address space available for mapping memory for user programs. For instance, 32-bit Windows reserves 1 or 2 GB (depending on the settings) of the total address space for the kernel, which leaves only 3 or 2 GB (respectively) of the address space available for user mode. This limit is much higher on 64-bit operating systems.

Memory-mapped files are becoming more difficult to implement in 32-bit architectures as files of over 4 GB become more common; such large files cannot be memory-mapped easily to 32-bit architectures—only part of the file can be mapped into the address space at a time, and to access such a file by memory mapping, the parts mapped must be swapped into and out of the address space as needed. This is a problem, as memory mapping, if properly implemented by the OS, is one of the most efficient disk-to-memory methods.

Some 64-bit programs, such as encoders, decoders and encryption software, can benefit greatly from 64-bit registers, while the performance of other programs, such as 3D graphics-oriented ones, remains unaffected when switching from a 32-bit to a 64-bit environment.

Some 64-bit architectures, such as x86-64, support more general-purpose registers than their 32-bit counterparts (although this is not due specifically to the word length). This leads to a significant speed increase for tight loops since the processor does not have to fetch data from the cache or main memory if the data can fit in the available registers.

Example in C:

int a, b, c, d, e;
for (a=0; a<100; a++)
{
b = a;
c = b;
d = c;
e = d;
}

If a processor only has the ability to keep two or three values or variables in registers it would need to move some values between memory and registers to be able to process variables d and e as well; this is a process that takes many CPU cycles. A processor that is capable of holding all values and variables in registers can loop through them without needing to move data between registers and memory for each iteration. This behavior can easily be compared with virtual memory, although any effects are contingent upon the compiler.
--------------------------------------------------------------

Though it helps to have more RAM available so that new iPhone better hope it has at least 4GB of RAM...though I believe in the mobile computing world, it is unnecessary to use 64 bit.

[eddy_berry]

64bit architecture will have little effect on the new iPhone5. It really is hype BUT it could force 64bit development across all manufacturers and subsequently cause Apple fans to gloat that they invented something first again. But in all seriousness, someone was bound to jump this and given Apples control over their own hardware and software and the cash they have for development I'm not surprised it was them who leaped first. Next it should be Windows phone and Android will find it hard unless Google can work closely with Hardware manufacturers to get a stable OS and they all have to offer Dev tools to switch current 32bit compatible apps to 64bit or risk losing their app catalogue all over this crazy obsession with the latest, albeit kinda useless, technological step that they can market like crazy to the general public who don't know any better. Wooh. And breath... But at least it drives innovation for the future of mobile computing. All these 64bit phones may do exactly what they do now but they will be more open to new technological advances and THAT is the cool part. Right now though don't worry about it. The real need for 64bit mobile phones is a long ways away and BlackBerry has time in that department for once. Right now it is marketing hype. Sorry folks, Apple beat everyone to it. In the future it may be the leap to the advancement of mobile technology. Thanks for listening.

[app_Developer]

Though it helps to have more RAM available so that new iPhone better hope it has at least 4GB of RAM...though I believe in the mobile computing world, it is unnecessary to use 64 bit.

I don't think we'll see 4GB this year. But I think we will see major improvements in image and video compression, which is something people do a lot of on phones. Things like taking a picture and sending it via iMessage should be a noticeably faster. I bet this is a big part of the claimed 2-3x performance improvement that Apple is making. It will be interesting to see.

[Nharzhool]

I don't think we'll see 4GB this year. But I think we will see major improvements in image and video compression, which is something people do a lot of on phones. Things like taking a picture and sending it via iMessage should be a noticeably faster. I bet this is a big part of the claimed 2-3x performance improvement that Apple is making. It will be interesting to see.

I'm not entirely sure how swapping to 64 bit will make a 2-3x performance increase...there is nowhere near this performance increase on Windows 7 32 and 64 bit versions.

As eddy_berry said, it is all just marketing hype. I'm pretty sure the new iPhone will be the same as the current one but with an even bigger price tag. I'm not saying that there is anything WRONG with the iPhone, I quite like it. I'm just saying that there is not much difference to be seen between them...I mean, their slogan is: "Indistinguishable Innovation" so what can one expect? :-P Cue the fan-boy hate...

[app_Developer]

I'm not entirely sure how swapping to 64 bit will make a 2-3x performance increase...there is nowhere near this performance increase on Windows 7 32 and 64 bit versions..

I don't think you can compare relative performance on desktop CPUs to what you would see on phones. I would think the additional register width on ARM would be relatively more valuable, for example.

but, again, we'll see when we get the devices. Third party benchmarks will be out there soon enough.

[Nharzhool]

I don't think you can compare relative performance on desktop CPUs to what you would see on phones. I would think the additional register width on ARM would be relatively more valuable, for example.

but, again, we'll see when we get the devices. Third party benchmarks will be out there soon enough.

Perhaps...honestly I can only speculate. Though Desktop comparisons are the only ones we have now so that is what I'm using. It MIGHT be better but it might also be nearly negligible. We will have to wait and see. Though, it will mostly likely perform well enough.

HOWEVER, Intel is releasing their 64 bit mobile SoC Atom Z3770 this quarter. So Apple wasn't there first...muhahahaha

[iN8ter]

32 I believe.

There is no reason to go to 64 unless your address space exceeds 4GB (including memory mapped I/O.) Then there is, but only for that reason; there is no inherent advantage to a 64-bit CPU over a 32-bit one, or 64-bit code over 32-bit.

Wrong.

Scientific and Math Code.
Media code (i.e. Photo and Video Editing on a Mobile device, Camera Software, etc.)
Anything that has to address large address spaces
etc.

Those all benefits a ton from 64-Bit. There are clear benefits for going 64-Bit, on a modern mobile device given what they are often used for, even if you're only putting 512 MB of RAM in the device.

Additionally, with the extra registers you can pass data in the registers instead of the stack, which is a performance gain.

A lot of the code in iOS is going to benefit a ton from 64-Bit, because Apple has a clue about how to develop software and the code will be properly optimized, not simply recompiled.

You have to actually optimize the software to take advantage of the benefits of a 64-Bit CPU/SoC. Simply recompiling is not going to gain you much.

For game development the 64-Bit CPU will be quite nice paired with the great GPU in that device. They probably gained a decent performance bump on the 5S because of the upgraded 64-Bit SoC for things like their camera software, image manipulation, media encoding/decoding, etc.

[tb146]

difference between 32bit and 64bit is depend on what kind of data and application do you want to use. With mobilephone, only communication what we need and i believe 32 bit more than enough.

If you want to do something heavier, small screen will not comfortable. Use a Laptop

[app_Developer]

difference between 32bit and 64bit is depend on what kind of data and application do you want to use. With mobilephone, only communication what we need and i believe 32 bit more than enough.

Have you ever, say, taken a video on your phone, edited it, and then uploaded it somewhere?

[iN8ter]

Some people aren't able to really see the difference unless they can do a comparison themselves. They'll just have to take our words for it.

Another thing I forgot to mention where a 64-bit processor may have an advantage is encrypting and decrypting data.

But since people here like Wikipedia, some snippets form it:

Some 64-bit programs, such as encoders, decoders and encryption software, can benefit greatly from 64-bit registers, while the performance of other programs, such as 3D graphics-oriented ones, remains unaffected when switching from a 32-bit to a 64-bit environment.

Lol. Wrote the above before I even seen this on the Wikipedia page. As for 3D oriented things, this WILL benefit from a better GPU, which usually comes with SoC upgrades since they are... SoCs...

This also touches on what the poster above stated. Simply taking a picture benefits from this, if the camera software is optimized, because encoding and compression code (i.e. raw data to Compressed JPEG image) benefits greatly from these things.

Some 64-bit architectures, such as x86-64, support more general-purpose registers than their 32-bit counterparts (although this is not due specifically to the word length). This leads to a significant speed increase for tight loops since the processor does not have to fetch data from the cache or main memory if the data can fit in the available registers.

As of June 2011, most proprietary x86 software is compiled into 32-bit code, with less being also compiled into 64-bit code (although the trend is rapidly equalizing[citation needed]), so most of that software does not take advantage of the larger 64-bit address space or wider 64-bit registers and data paths on x64 processors, or the additional general-purpose registers. However, users of most RISC platforms, and users of free or open source operating systems (where the source code is available for recompiling with a 64-bit compiler) have been able to use exclusive 64-bit computing environments for years. Not all such applications require a large address space or manipulate 64-bit data items, so these applications do not benefit from these features. The main advantage of 64-bit versions of such applications is the ability to access more registers in the x86-64 architecture.

From the Entry for "64-Bit Computing"

Going from 32-bit to 64-Bit Windows won't give a huge benefit because a ton of Windows applications (even in the OS) are still 32-Bit code and therefore do not really benefit much from the 64-Bit CPU. Those parts that are optimized, and coded to take advantage of it, however do, and of course you aren't going to access more than 4GB RAM in a data heavy app (like CAD, Video Editing, Photo Manipulation, etc.) with a 32-Bit OS. Some of those apps have completely dropped support for 32-Bit OSes.

[Sergeant_Im]

Until Apple and its app developers start writing apps in 64-bit I don't see any benefit of iPhones having 64-bit processors. Basically 0 performance gain.

Posted via CB10

[BBPandy]

I don't see BlackBerry going 64bit for a while. A couple years at least. Here's why.
The current QNX Neutrino RTOS (which is the heart of BB10) is still 32bit. Even if they were to release a new 64bit kernel next year, it would take another year for BlackBerry to implement it into their OS. Switching to 64bit is no simple thing.

QNX is due for another major update though as QNX has a new OS every 10 years.
1980's --- QNX 2
1990's --- QNX 4
2000's --- QNX 6
2010's --- QNX 8 (maybe it will be 64bit)

There was speculation last year that BB10 would be running QNX 8, but I don't know what truth there is to those rumours as I haven't heard anyone say that since BB10 was released

[thurask]

Even if BB10 were to implement 64-bit ARM, the OMAP and Snapdragon chips that are in BB10 devices past, present and future are ARMv7. If BBRY were to make the OS 64-bit capable, it would still be quite a while before the hardware gets upgraded to ARMv8; Qualcomm hasn't said anything openly about producing ARMv8 hardware.

[tlegend2012]

64 chip would be cool if you have a bigger screen,and right resolution.. just can't see the advantage on 5s

[u4ria]

I found the following article a pretty interesting read.

[digitalman101]

I found the following article a pretty interesting read.

Interesting indeed.

"iPhone 5S: 1GB DRAM (makes sense)" I think the author was being ironic

Posted via CB10

[tickerguy]

I have 30+ years of software development under my belt including OS-level and assembler work. I'm very-much aware of the differences between 32 and 64-bit architectures, considering that I started programming on architectures where 16 bits was the address space available (specifically, Z80s and similar, all in assembler.)

There is no material performance advantage until and unless you need the address space capability. Then there is, in a big way. Yes, there are small (very small) advantages that one gets from a "larger" register set but when it comes to things like encryption the big wins are from native instructions (e.g. the Intel AES-NI instruction set which is ~100x as fast as doing the same crypto in software.)

There is no free lunch. Manipulating more bits at a time sounds like a free lunch but it is not, and in particular it will impact power budget, although again probably not greatly.

In short the "advantages" of going to 64-bit on their own are tiny, but when it comes to address space considerations then the differences are gigantic. There are, incidentally, ways to "cheat" beyond the alleged maximum physical address space (I used to exploit these back in the 16-bit days quite-heavily out of necessity) but they're messy as hell and require hardware that's cooperative, which you often do not have.

What can be of course significant is a richer instruction set but that's not necessarily tied to the size of the address space.

[anon62607]

I have 30+ years of software development under my belt including OS-level and assembler work. I'm very-much aware of the differences between 32 and 64-bit architectures, considering that I started programming on architectures where 16 bits was the address space available (specifically, Z80s and similar, all in assembler.)

There is no material performance advantage until and unless you need the address space capability. Then there is, in a big way. Yes, there are small (very small) advantages that one gets from a "larger" register set but when it comes to things like encryption the big wins are from native instructions (e.g. the Intel AES-NI instruction set which is ~100x as fast as doing the same crypto in software.)

There is no free lunch. Manipulating more bits at a time sounds like a free lunch but it is not, and in particular it will impact power budget, although again probably not greatly.

In short the "advantages" of going to 64-bit on their own are tiny, but when it comes to address space considerations then the differences are gigantic. There are, incidentally, ways to "cheat" beyond the alleged maximum physical address space (I used to exploit these back in the 16-bit days quite-heavily out of necessity) but they're messy as hell and require hardware that's cooperative, which you often do not have.

What can be of course significant is a richer instruction set but that's not necessarily tied to the size of the address space.

I imagine 64 bit is really another way of saying "new revision of the instruction set which also happens to increase the word size to 64 bits". The performance advantages probably don't come so much from a larger native word size as it does from other revisions to the instruction set.

also, 64 bits seems like the "correct" size. the floating point width is 64 bits as well and thought this is both ISA and implementation dependent you could design the architecture such that a single register file and register alias table would handle both fixed and floating point, and devote all of the read/write ports that would have been independently partitioned to that same file.

also 2^32 is only 4 billion, its a small enough number that in real world situations you often need larger numbers than that. of course the compilers these days will handle all of that for you and it's something of an aesthetic desire but 64 bit integers seems "right".

[BBPandy]

I found the following article a pretty interesting read.

A very good article

Posted via CB10