What ever happened to the 128bit Windows?

liquidsnake718

Distinguished
Jul 8, 2009
1,379
0
19,310
http://www.tomshardware.com/news/windows-7-128bit-next-generation,8827.html

They should actually try to release a 128 Win7 but I doubt it....

With the news of a new Windows 8 being developed we can assume that it will be 128bit... I wonder if they can truly release it by 2 years which imo will take more than 4 years for it to be marketed and sold... I think MS has a good thing with Win7 and they should mlik it as long as possible unlike what they did to Vista. I dont like it because I have Vista and am still using it for my Desktop but my netbook is win7...
 
Solution
I was aware of that... but I thought the OP might be getting confused. You hear about 128-bit registers all the time... but that doesn't make the processor capable of running a 128-bit OS.

From Wiki:

There are currently no mainstream general-purpose processors built to operate on 128-bit integers or addresses, though a number of processors do operate on 128-bit data. The System/370, made by IBM, could be considered the first rudimentary 128-bit computer as it used 128-bit floating point registers. Most modern CPUs feature SIMD instruction sets (SSE, AltiVec etc.) where 128-bit vector registers are used to store several smaller numbers, such as four 32-bit floating-point numbers, and a single instruction can operate on all these...
Considering there are currently no 128-bit processors on the market, it seems highly unlikely that MS would waste time developing for a platform that doesn't exist. Now if in the next year or so Intel or AMD creates a processor capable of processing 128-bit instructions... then we'll see a 128-bit version of Windows.
 
I don't see it. Not for Win8. No hardware. There's still a lot of 32 bit hardware out there. And for those of us with 64 bit machines, there's still a lot of 32 bit software.

For the near future, I think we need good multiprocessor software more than we need 128 bit software.
 
That's an old article reporting rumours that were spawned by some MS engineer's blog about working on support for the new AMD floating point instructions. AMD and Intel processors have had 128-bit registers in their floating point ("SSE") units for over a decade now, so that's nothing new.

128-bit computing in terms of using 128-bit numbers to address memory (which is the sense in which the term "64-bit computing" is generally used) is at least a few decades away. Going from 32-bit to 64-bit addressing doesn't just double the amount of available memory, it actually multiplies the amount of addressable memory by 4 BILLION.

To draw an analogy - consider 4GB, the maximum amount of memory that can be addressed by a 32-bit computer, to be like one sheet of paper. My 12GB system would hold the equivalent of 3 sheets of paper. A system that has all of the memory addressable by a 64-bit computer would be the equivalent of 4 BILLION sheets of paper - that's equivalent to a stack of paper 600 MILES HIGH.

Obviously, there are no computers anywhere in the world that are anywhere close to that now, nor will there be for at least a couple of decades. Right now, the maximum amount of memory that the biggest, baddest versions of Windows can support is 2TB - that's equivalent to 20 inches worth of those 4GB sheets. There's a long, loooonnng way to go from "20 inches" worth of memory to "600 miles" worth...
 

croc

Distinguished
BANNED
Sep 14, 2005
3,038
1
20,810


Did you not read the article that you linked, where it speculated on 128 bit extensions to the Itanium processor? (IA-64)

Personally I don't see it, nor the need for it, but then again an IBM exec once famously stated that there might be five customers world-wide that could use a main-frame...
 

hundredislandsboy

Distinguished
Nithing happened to it. Did it ever exist? What happened to it? For that to be answered, it had to have existed to begin with. Nothing happens to something when that something is a nothing and doesn't exist. So to answer your question, the answer is "Nothing". Lol.
 

croc

Distinguished
BANNED
Sep 14, 2005
3,038
1
20,810


Look at the charts at top500.org... Show me one processor that a well-heeled person could not afford, and one processor that is 128 bit. Must be a reason, eh?
 
General use CPUs capable of running Windows are NOT 128-bit... therefore there has never been a 128-bit version of Windows on the radar. Specialized processors may have 128 or even 256-bit registers... but that doesn't mean they are capable of running 128-bit operating systems.
 

wayneepalmer

Distinguished
Feb 23, 2009
228
0
18,690
Good Lord, you could run fission initiator compression wave calculations for pressure optimization to increase fission yields in Trident MARV warheads with a Bloomfield i7!!

What more processing power do we need? We aren't using a the current capabilities very efficiently now. We need software optimization far more than we need new hardware.
 
All current Intel and Intel-compatible x86 and x64 processors have eight 128-bit registers in their SSE units. See: http://en.wikipedia.org/wiki/Streaming_SIMD_Extensions

But in the traditional sense for Intel systems, the terms "32-bit" or "64-bit" do not refer to the size of the registers, but rather to the size of the addresses. As I posted above, current processors with a 64-bit address space have the ability to use far, far more memory than any current system (even the secret government ones!) can possibly get anywhere near using. There will be no need to go beyond a 64-bit address space for at least a couple of decades.

People erroneously think that going from 32- to 64-bit addressing merely doubles the address size, but it's way, way more than that. 64-bit addressing gives you the ability to use 4 billion times more memory than a 32-bit system can use.

As an analogy, think of a growing city in a Prairie town. The main street is getting longer, and there are now so many blocks that a 4-digit house number is no longer adequate. Going from 32- to 64-bit addressing is similar in concept to going from a 4-digit house number (maximum 9,999) to an 8-digit house number (maximum 99,999,999). You've only doubled the number of digits in the house number, but you've increased then number of possible addresses by a factor of 1000. It's going to take the city a *long* time before the main street becomes so long that it needs addresses beyond 100 million...
 
I was aware of that... but I thought the OP might be getting confused. You hear about 128-bit registers all the time... but that doesn't make the processor capable of running a 128-bit OS.

From Wiki:

There are currently no mainstream general-purpose processors built to operate on 128-bit integers or addresses, though a number of processors do operate on 128-bit data. The System/370, made by IBM, could be considered the first rudimentary 128-bit computer as it used 128-bit floating point registers. Most modern CPUs feature SIMD instruction sets (SSE, AltiVec etc.) where 128-bit vector registers are used to store several smaller numbers, such as four 32-bit floating-point numbers, and a single instruction can operate on all these values in parallel. These are 128-bit processors in the sense that they have 128-bit registers, but they do not operate on individual numbers that are 128 binary digits in length.

-128-bit processors could become prevalent when 16 exbibytes of addressable memory is no longer enough (128-bit processors would allow memory addressing for 340,282,366,920,938,463,463,374,607,431,768,211,456 bytes (~340.3 undecillion bytes or 281,474,976,710,656 yobibytes ). However, physical limits make such large amounts of memory currently impossible, given that amount greatly exceeds the total data stored on Earth today.

-Quadruple precision (128-bit) floating point number can store qword (64-bit) fixed point number or integer accurately without losing precision. Notice that since the 8087 (1980), x86 architecture supports 80bits float-points that store and process accurately 64bits integers

-Sony's Playstation 2 CPU Emotion Engine is advertised as a 128 bit processor. It has 128-bit SIMD registers, like many processors, but is only a 32-bit processor in the traditional sense as it can only use 32-bit memory addresses.

-The AS/400 virtual instruction set defines all pointers as 128-bit. This gets translated to the hardware's real instruction set as required, allowing the underlying hardware to change without needing to recompile the software. Past hardware was 32-bit CISC, while current hardware is 64-bit PowerPC. Because pointers are defined to be 128-bit, future hardware may be 128-bit without software incompatibility.

-Larger bit widths are also commonly seen on the memory interface of graphics processing units, including 128-bit, with some bus widths reaching 512-bits long.

So there you have it. No 128-bit processors means no 128-bit Windows. When we say 128-bit, we're of course referring to address space... not any other registers or bus widths. We will not see 128-bit Windows for quite a few years yet; MS might not even exist as a company by the time it becomes a reality.

 
Solution