Future of CPUs

Discussions on more advanced topics such as monolithic vs micro-kernels, transactional memory models, and paging vs segmentation should go here. Use this forum to expand and improve the wiki!
Post Reply
User avatar
paolodinhqd
Posts: 5
Joined: Mon Mar 01, 2010 2:29 am

Future of CPUs

Post by paolodinhqd »

I'm thinking of the future for OSes and so
i think of the future for CPUs.
CPUs are reaching their limit and breaking the Moore's law.
That's why Intel and other manufacturers trying to
add more and more cores to a single CPU.
The time for Moore's law is running out. [-X

We already have 64 bit CPUs,
and I believe in the next 100 years,
there won't be such thing called 128 bit, :roll:
except for servers those processing IP v6
and some other computer systems serving researches.

The limit for memory is enough in 64 bit.
It's twenty zeros behind number one. =D>

How do guys think? :)
redoktober
Member
Member
Posts: 109
Joined: Thu Feb 26, 2009 12:58 am
Location: Gurgaon/New Delhi, India
Contact:

Re: Future of CPUs

Post by redoktober »

it's possible.
Gordon Moore wrote:It can't continue forever. The nature of exponentials is that you push them out and eventually disaster happens. In terms of size [of transistors] you can see that we're approaching the size of atoms which is a fundamental barrier, but it'll be two or three generations before we get that far—but that's as far out as we've ever been able to see. We have another 10 to 20 years before we reach a fundamental limit. By then they'll be able to make bigger chips and have transistor budgets in the billions.


he's got a point.
but then, i guess that's where nanotechnology makes a grand entry.
i attended a conference on current trends in nanotechnology and its applications in computer science. it was hosted by a joint indo-french group of scientists, and they did have some cool ideas..
but the stuff they said is not really feasible to produce on a massive scale...at least, not for the next 10 years.
i guess chips may start growing in size....or they may apply layering, have, you know, different layers full of transistors. they might even look into quantum tunneling, though it seems intel already uses it.

some people say, that we'd reach a singularity point, where moore's law won't apply, and the progress we see in years would take place, like, in 3 seconds!

jeez....imagine downloading new compiler toolchains and manuals, and then reading them, and then seeing it become obsolete in a matter of hours :D
"Do you program in Assembly?" she asked. "NOP," he said.

"Intel Inside" is a Government Warning required by Law.
redoktober
Member
Member
Posts: 109
Joined: Thu Feb 26, 2009 12:58 am
Location: Gurgaon/New Delhi, India
Contact:

Re: Future of CPUs

Post by redoktober »

i forgot all about DNA and quantum computing!
if you think about it, moore's law can't hold true for DNA computers, or quantum computers..
"Do you program in Assembly?" she asked. "NOP," he said.

"Intel Inside" is a Government Warning required by Law.
User avatar
qw
Member
Member
Posts: 792
Joined: Mon Jan 26, 2009 2:48 am

Re: Future of CPUs

Post by qw »

I remember the times when 32 KiB seemed like a hell of a lot of memory. One thing is for sure: software is getting more and more bloated, and requires more and more memory and higher processor speed. Never underestimate human ingenuity. We'll come up with something.
Selenic
Member
Member
Posts: 123
Joined: Sat Jan 23, 2010 2:56 pm

Re: Future of CPUs

Post by Selenic »

Hobbes wrote:One thing is for sure: software is getting more and more bloated, and requires more and more memory and higher processor speed. Never underestimate human ingenuity.

In exchange for this, we get faster-to-write (and therefore somewhat cheaper) and more featureful programs, so overall I don't think it matters too much. Also, it means that now a large amount of the split between what's possible and what isn't is along theoretical, rather than practical, lines.

I remember reading something suggesting that computers in general are heading towards a more aSMP model. GPGPU (using graphics cards as highly parallel processors) is a step in that direction, as is the Cell processor. (PowerPC plus eight special SIMD-oriented cores)

@OP: IPv6 is deliberately overkill - the whole point is to make routing easier by making decent hierarchies possible, while for IPv4 this isn't possible as it's moving towards saturation.
pcmattman
Member
Member
Posts: 2566
Joined: Sun Jan 14, 2007 9:15 pm
Freenode IRC: miselin
Location: Sydney, Australia (I come from a land down under!)
Contact:

Re: Future of CPUs

Post by pcmattman »

The limit for memory is enough in 64 bit.

And 640 K was supposed to be enough for anybody ;)

I do think though that by the time we reach systems hitting the 64-bit memory limits, we'll no longer be using 64-bit systems and instead be working with a completely new concept of processing.
nedbrek
Member
Member
Posts: 44
Joined: Tue Dec 15, 2009 6:36 pm

Re: Future of CPUs

Post by nedbrek »

You can get a feel for things by running some numbers...

2^64 address space ~= 1.844e19 bytes

It's hard to imagine any technology that will allow you to store more than one bit per atom... (or add an order of magnitude (or two!), later if you like)

One mol of matter contains 6.02e23 atoms. So, ~1e4 (10,000) bits per mol. Say 10 bits per byte (to make the math easier), gives you 1,000 times the 64 bit address space per mol.

1 mol of C12 is 12 grams.

So, there may be some demand for 128 bit computers... (I am ignoring power consumption!)
User avatar
Love4Boobies
Member
Member
Posts: 2111
Joined: Fri Mar 07, 2008 5:36 pm
Location: Bucharest, Romania
Contact:

Re: Future of CPUs

Post by Love4Boobies »

Quantum computers will be here before we need 128-bit CPUs. I totally agree with pcmattman.
"Computers in the future may weigh no more than 1.5 tons.", Popular Mechanics (1949)
[ Project UDI ]
User avatar
Brendan
Member
Member
Posts: 8561
Joined: Sat Jan 15, 2005 12:00 am
Location: At his keyboard!
Contact:

Re: Future of CPUs

Post by Brendan »

Hi,

My predictions are:
  • more cores (with little improvement in the performance of each core)
  • wider SIMD (e.g. AVX)
  • less power consumption per core
  • more integration (Intel and AMD have already shifted the memory controller onboard, and have CPUs with built-in GPU). In the next 10 years I can imagine ethernet, disk controllers and HPET being shifted into the CPU, and eventually the RAM chips too (which would improve RAM bandwidth and latency, and allow for smaller caches).

As more is included in the chip itself, Intel and AMD will become more flexible (e.g. allow for a wide variety of "cores + gpus + other devices" combinations); and 80x86 will be used in more and more mobile/portable devices. I can imagine internet over WiFi killing the mobile phone network; and eventually Intel will push out millions of embedded CPUs with very little markup (maybe even selling them at a loss) and kill off ARM and/or MIPS.

For servers, I think that 80x86 will gain more reliability and fault tolerance features for high-end servers and kill Itanium.

VIA and Nvidia will be OK for about 5 years, then they'll start struggling. Eventually VIA will buy Nvidia or Nvidia will buy VIA, and we'll end up with three "80x86 CPU+GPU+extras" manufacturers and not much else.

For quantum computers, it'll take another 5 years before they're viable, and 5 years and 1 week before people realise they're mostly useless (outside of a few very specialised fields). You might see a "quantum co-processor" if you're lucky (just like you might see a "physics processor" today if you're lucky), but most people will probably never see either of these things. :)


Cheers,

Brendan
For all things; perfection is, and will always remain, impossible to achieve in practice. However; by striving for perfection we create things that are as perfect as practically possible. Let the pursuit of perfection be our guide.
User avatar
Solar
Member
Member
Posts: 7615
Joined: Thu Nov 16, 2006 12:01 pm
Location: Germany
Contact:

Re: Future of CPUs

Post by Solar »

Well... 128 bit might actually be ahead. But I did that "one bit per atom" calculation once and found that, with 128 bit, we're already talking more atoms than the entire earth (IIRC), which makes it pretty much a definite limit. ;-)
Every good solution is obvious once you've found it.
User avatar
paolodinhqd
Posts: 5
Joined: Mon Mar 01, 2010 2:29 am

Re: Future of CPUs

Post by paolodinhqd »

Solar wrote:Well... 128 bit might actually be ahead. But I did that "one bit per atom" calculation once and found that, with 128 bit, we're already talking more atoms than the entire earth (IIRC), which makes it pretty much a definite limit. ;-)

Solar [Taking more atoms than the entire earth] <-- you must be great at physics in calculating this, you're sure about this calculation? =D>

Solor [128 bit might actually be ahead] <-- how many years you think it would come? My point is 100 years+, this is out of my life and I just shouldn't care :wink:
User avatar
Love4Boobies
Member
Member
Posts: 2111
Joined: Fri Mar 07, 2008 5:36 pm
Location: Bucharest, Romania
Contact:

Re: Future of CPUs

Post by Love4Boobies »

Not sure, maybe soon we'll all hold 3D hologram data files of entire landscapes in really high resolution on our disks and will want to cache them in memory (be that RAM or IRAM or something else). We might make the leap sooner than you think :)

paolodinhqd wrote:Solar [Taking more atoms than the entire earth] <-- you must be great at physics in calculating this, you're sure about this calculation? =D>


Sounds to me like elementary math with big numbers.
"Computers in the future may weigh no more than 1.5 tons.", Popular Mechanics (1949)
[ Project UDI ]
User avatar
01000101
Member
Member
Posts: 1598
Joined: Fri Jun 22, 2007 12:47 pm
Location: New Hampshire, USA
Contact:

Re: Future of CPUs

Post by 01000101 »

I predict that 128-bit processors will be announced sometime next year (or two) and be available within 5 years.
User avatar
Love4Boobies
Member
Member
Posts: 2111
Joined: Fri Mar 07, 2008 5:36 pm
Location: Bucharest, Romania
Contact:

Re: Future of CPUs

Post by Love4Boobies »

32-bit CPUs are pretty mainstream today, we haven't even moved to 64-bit completely. How many years were we stuck with 16-bit systems and how many years were we stuck with 32-bit systems? Do you think 64-bit systems will only last a few years considering that we don't even use them yet at full power? (Canonical addresses anyone?)
"Computers in the future may weigh no more than 1.5 tons.", Popular Mechanics (1949)
[ Project UDI ]
User avatar
Owen
Member
Member
Posts: 1700
Joined: Fri Jun 13, 2008 3:21 pm
Location: Cambridge, United Kingdom
Contact:

Re: Future of CPUs

Post by Owen »

I doubt it. Current CPUs are really close to the limits of performance working on 64-bit numbers; remember that for many operations doubling the word size doubles the time. For example, Phenom IIs overclock better in 32-bit than 64-bit mode.

Additionally, no processor in production has a 64-bit physical or virtual address space and we are nowhere near needing one.

As for Intel killing ARM, I don't see that happening. You cannot make an x86 of comparable performance with a comparable silicon budget. x86 is a very ugly architecture; ARM is not perfect but a very clean one. Secondly, Intel and ARM do not directly compete: ARM licenses its cores to companies which have a combined revenue ~4x Intels. Thirdly, Intel is right on the EU's anti-trust watch list, and any attempt to do anything anti-competitive against ARM (Such as the aforementioned selling at a loss) is going to land them more large fines.

You have to note that the ARM cores used in phones contain lots of peripherals, from the simple (SPI & I²C/SMBus) to communicate with external devices, to the specialized (H.264 decoders), to the application specific (Wi-Fi, UMTS and GSM basebands - often on one chip!). With Intel's current Atom platform, you require at least 4 chips to do the same (CPU + Northbridge + Southbridge + Radios); with their next, they're shrinking it to 3, but still each of those chips requires more power than an ARM which can do the same.

Intel will continue to grow, but don't expect them to get into the mobile phone market. They simply cannot compete there, and they have over half of the industry competing against them.

Oh, and the performance of ARM cores is increasing faster than that of the Atom. And this from a company with 1/1000th the revenue to do R&D with.
Post Reply