When are we going 128-bit? How it will be? What gets better? [closed]

Question

With the launch of new 64-bit OS and processors, we can say we already are in the 64-bit age. So, what's the next step? 128-bit.

What are the 128-bit processors already available? Which operating systems plan to move to 128-bit? What advantages will we have when the 128-bit age comes? What will change from now?

Answer 1

There really is no hurry to switch to 128-bit any time soon. Switching from 32 to 64-bit processing didn't double the capabilities, it raised them to a power of two. We have no need for mainstream desktop 128-bit computing any time soon.

Answer 2

I don't think we'll be heading to the 128-bit realm any time soon. At least in general use cases (I'm sure there are some scientific uses that could benefit). Wikipedia has an interesting little tidbit on 128-bit computing.

My favorite part from the article:

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); about 340,282,366,920,938.5 yottabytes; exactly 281,474,976,710,656 yobibytes; exactly 256 tebi-yobibytes; or exactly 256 exbi-exbibytes of memory). However, physical limits make such large amounts of memory currently impossible, given that amount greatly exceeds the total data stored on Earth.

So yeah, no rush!

Answer 3

The main reason for the 64bit change is the RAM blockage. 64bit has an upper limit of, I believe, 4 petabytes. It'll be a while before we hit that :)

edit: It's 16 exabytes. To put that in perspective, 1000 petabytes is one exabyte. One petabyte is 1000 terabytes. One terabyte is 1000 gigabytes. Average today is 2 gigabytes of RAM, I believe.

It'll be a while.

Answer 4

The big advantage of 64 bits is smooth addressing of more memory. 32-bit addresses can cover 4 gigabytes, while 64 bits can cover about 16 billion gigabytes, which is currently good enough for pretty much any purpose.

This will not necessarily be enough forever, but it will take some major breakthroughs to use up 64 bits of address space.

The other potential advantage is that 64-bit computers deal with larger chunks of data at a time, and therefore can do things faster with such larger chunks. Unfortunately, this doesn't seem to be a big improvement for most purposes, since most data values people use tend to be relatively small.

The other real advantage of the x64 architecture over the old 32-bit is that it's a lot cleaner. It's more than a set of extensions maintaining compatibility with a chip design over 25 years old that was in a way compatible with the first microprocessor ever marketed. The x64 architecture isn't ideal, but it's much better than the 32-bit. I don't see moving to even the best possible 128-bit architecture as being worth it just for that.

Answer 5

64 bit really only refers to the standard word/bus/address size of a system. The reason the size needed to change was because we were starting to deal with volumes of data that were impossible to address in 32 bits without some ugly hacks.

An easy example would be Windows' old 16 bit FAT16 file system; it worked "fine" until hard drives passed 4 gigs, at which point you had to repartition all your drives into 4 gig partitions, or upgrade to a 32 bit file system. Obviously a bit of a problem when your 40 gig drive is split into 10, 4 gig partitions.

The same thing happens with memory and bus architecture. The RAM limit for a 32 bit OS is around 4 gigs (RAM is addressed in smaller blocks than harddrive space), and once that limit is reached you have to start using ugly hacks like PAE, or you need to move to a 64 bit address space, with it's upper RAM limit of 16.8 exabytes (16.8 million trillion terabytes).

So, when the day comes where you can buy 2 exabyte RAM crystals at the hypermart (say, 2025), then we'll look at making it 128 bits. Until then, barring some obscene bus technology leap, there's just no need.

Answer 6

Two points:

• 128 bits on address bus is not a need in my opinion.

• 128 bits or more on data bus may open more interesting architectures, especially with multiple core CPU and multiple CPU machines (each core remaining 64 bits). At the end, this leads to massively parallel computers. I only hope that future OSes and future computing university degrees will be able to handle that.

Answer 7

Funny that only one week later Ars Technica quotes Robert Morgan at Microsoft:

Working in high security department for research and development involving strategic planning for medium and longterm projects. Research & Development projects including 128bit architecture compatibility with the Windows 8 kernel and Windows 9 project plan. Forming relationships with major partners: Intel, AMD, HP, and IBM.

Robert Morgan is working to get IA-128 working backwards with full binary compatibility on the existing IA-64 instructions in the hardware simulation to work for Windows 8 and definitely Windows 9.

Edit: one day later people are all telling us (like many other answers have said as well):

There is no IA-128. The Abominable Snowman and the Easter Bunny could hold a press conference about the lost sex tapes of The Tooth Fairy and it would have more credibility than this rumor. In fact, calling it a rumor lends too much credence to the story. There may be a Robert Morgan, there certainly is a Windows 8, but there is no IA-128.

and

IA-64 was Intel's joint venture with HP to create a next generation 64 bit server processor. Finally released far behind schedule, the Itanium line has been beset with problems. Poor performance, a lagging upgrade path, and poor vendor uptake have all conspired to create what John Dvorak recently called, "one of the great fiascoes of the last 50 years""

Answer 8

I think everyone's missed the point. How many bits wide a CPU is considered to be is about how wide the registers are in the CPU, not about how wide the data or address busses are.

Motorola's 68000 series processors were all 32 bit, despite have either 8 bit or 16 bit wide databases and 16 or 24 bit wide address busses (68000 had a 24 bit address bus, 16 bit data bus, 68008 was the same core but only had an 8 bit external data bus - the 68000 would simply make two memory reads to fill a 32 bit register, but all opcodes acted on 32 bit operands in 32 bit registers).

The value of widening the registers is that you can manipulate bigger, or more precise numbers in fewer steps. I guess if we're dealing with numeric values, then the value is in the fact that you can represent a real (floating point) number much more precisely with more bits, which means numeric applications that require very tight tolerances can do their sums in fewer steps, meaning much faster processing.

Number crunching is not the only important thing though. We often use bitfields, and if you have to shift left, shift right, xor, and etc loads of bits, then obviously the more we can fit in a register in one go, then the fewer instructions are needed to complete the task and therefore the faster it happens.

All that said, I doubt there'll be a hurry to get 128bit to the mass user desktop market. Instead we'll continue for a while on the trend of going multicore. For general use the need for the kind of precision afforded by 128 bit registers is so infrequent that we can live with it happening over multiple steps. The more pressing matter is the fact that we multitask much more than we used to, whether we realise it (eg loads of apps running at once) or not (loads of services and multithreaded apps). Going multicore gives fills that need, because a single core has the added overhead of task switching, so more cores are the way to more power for now.

As for multithreaded programming being hard, it's not really, its just different. Us older ones come from a background of procedural programming, so its a bit of a change in thought process to go multithreaded. The younger folks are lumbered with the cumulative knowledge and experience of the older folks, so there's still not that much good info out there about multithreaded programming, but we're getting there.

Answer 9

The 64-bit age really just started.

Programmers are just now learning how to use this power. Up until recently many drivers and other software would not work properly on 64-bit OS's. Don't expect to see 128-bit for a long time.

Answer 10

128 bit computers would have the power to address more RAM than there are atoms in the universe. I can't really see a need for that - to paraphrase a quote misattributed to Bill Gates, "one universe should be enough for anybody".

(Ok, that's a slight exaggeration. There are approximately 4x10^79 hydrogen atoms in the universe according to the first reference I found online, while 2^128 is closer to 10^38.)

Answer 11

Never, or we've already had 1024 bit cpus, where have you been?

There have been VLIW (very long instruction word) cpus, but they haven't worked out that great (the Transmeta cpus were, for instance). VLIW would be the 1024 bit.

Never as 64 bits ought to be enough for longer than we can make sane guesses. Now for file systems, that's another matter, as Sun figured that 64bits would be hit in a decade or two (a decade from now) so they made ZFS 128 bit, so it should be good until it is obsolete for other reasons.

The last thing is you are just talking about instruction width, there is addressing and data. So to stick with the PC, the first PC had a 8 bit data bus, but a 20 bit address bus. From the Pentium Pro on, they have ahd the ability to do 36 bit addressing, and the various sorts of SIMD allow for data at 64-128 bits.

Itanium (ia64) is 128 bit, but that is really 3 40 bit instructions bundled together.

And with IPv6 (128 bit addresses) I can see specialized router cpus being more than just 128 bit for data, but who knows?

Answer 12

I am sure it will happen one day, but not anytime soon. Besides, all the operating systems seem to bloat and grow right along with computer speeds. ;)

It will probably be seen in super computing and scientific research before it ever sees the light of day on your average computer desk. I also agree that application programmers are just now beginning to even design for 64 bit, let alone use it to its full potential. Not everything benefits from 64 bit either and that will probably not change with 128bit.

Answer 13

Funny you should all mention it... but check out this link on Windows 8

Research & Development projects including 128-bit architecture compatibility  
with the Windows 8 kernel and Windows 9 project plan

Here's the story

Answer 14

Don't know if it really matters but to correct an earlier statement of the first desktop computer being an 8 bit. It was a 16 bit (8086 INTEL made by AMD for IBM). The 8088 came out later (8 bit). Probably the earlier version Zilog chips were Lesser bits; not necessarly Desktops! 64 bit was dropped because, programmers refused to program in 64 bit. Still a problem today.