Fudzilla: The Future of Computing Performance, Game over or next level?

[cbn]

http://fudzilla.com/home/item/21255-technology-growth-to-come-to-a-screaming-halt

A new report from the US National Research Council has warned that technology growth will grind to a halt unless a new generation of hardware and software is developed. The catchy-titled “The Future of Computing Performance: Game over or next level?” is packed full of bad news for the industry.

While microprocessors improved in speed by a factor of 10,000 during the 1980s and 1990s, two obstacles could mean computing power hitting a wall in the next decade, the report said.

While transistors have become ever smaller and more tightly packed, the speed at which microchips are clocked has levelled off, reaching around 3 gigahertz in 2005. That's because such fast chips generate too much heat to be used in smartphones and personal computers.

This plateau will kill Moore's law. While manufacturers have been fabricating two, four or eight microprocessor cores on a single chip to get around this hurdle, it is not enough the report warns. This is because the power efficiency of present transistors cannot be improved much more, and performance "will become limited by power consumption within a decade". It needs a yet-to-be-invented transistor architecture to save it.

Software is still behind hardware developments with multicore chips still not properly factored into designs. Software has to be designed to execute multiple tasks in parallel, rather than serially and few programs are up to this challenge.

But the report warns that converting the vast majority of software, written for serial execution, to work efficiently in parallel mode will be exceedingly difficult. It needs new software-engineering processes and tools. Programmers will need to be retrained, it warned.

 

[IntelUser2000]

I'm gonna call that report BS.

The biggest roadblock in achieving such science fiction like dreams is the possibility of mankind destroying themselves before the dream ever occurs. Newer methods to overcome previous problems will likely continue to exist.

 

[steve wilson]

Some genius somewhere will come up with a solution and computers will continue to progress. I'd bet my house on it.

 

[BD231]

Douchebaggary

 

[soccerballtux]

yawn, keep making smaller transistors and we can keep packing more into a chip, Moore's law lives on.

When we hit the ~11nm node, however, I will call it to be over around then.

 

[Bateluer]

yawn, keep making smaller transistors and we can keep packing more into a chip, Moore's law lives on.

When we hit the ~11nm node, however, I will call it to be over around then.

Who's to say it'll still use transistors, or electric impulses?

Worry not, SkyNet will still destroy humanity on Judgment Day.

 

[smangular]

yawn, keep making smaller transistors and we can keep packing more into a chip, Moore's law lives on.

When we hit the ~11nm node, however, I will call it to be over around then.

Perhaps without future advancements which have been happening for decades.

Yes of course getting programmers to leverage multi-core computing is and will continue to be a challenge. Technologies like Intel Turbo-Boost will continue to increase the speed of single threaded tasks.

 

[Bryf50]

So wait then why is 1 core in my 3ghz i7 multiple times faster then a 3ghz p4?

 

[aigomorla]

Who's to say it'll still use transistors, or electric impulses?

Worry not, SkyNet will still destroy humanity on Judgment Day.

no im putting money on giant asteriod smashing into the earth b4 computers take over the world.

 

[Lonyo]

So wait then why is 1 core in my 3ghz i7 multiple times faster then a 3ghz p4?

Exactly. Talking about a GHz wall is the most stupid thing ever when you do more per clock.
The only reason that we got to 3.8GHz with the P4 was because they designed it to scale to 5 or 10GHz and failed horribly.
Now they design differently and get more work done.

And the next line makes no sense. I don't even know what it's trying to say. We've levelled off at 3GHz because any faster and chips won't work in mobile phones? What?

 

[StinkyPinky]

Yet another person that thinks CPU speed is all about the GHZ

I get that at work. "Why am I only getting 3GHZ?? That's the same speed as my current PC". Yeah sure, you keep that P4 then. This i5 can go on my desk.

 

[God Mode]

The summary I got from this is that with exceptions, programmers suck shit. Hardware is getting better but a lot of the programs I see is still windows 95 era in efficiency.

Bloat and a lot of calling home and other behind the scene BS but no better in terms of operation or speed. These days it simply wants to hijack your computer and install crap all over the system.

 

[Jacky60]

Who's to say it'll still use transistors, or electric impulses?

Worry not, SkyNet will still destroy humanity on Judgment Day.

But surely a new hero will emerge from the smouldering ashes of our civilisation and allow us to fight back against the machines.

 

[Anomaly1964]

But surely a new hero will emerge from the smouldering ashes of our civilisation and allow us to fight back against the machines.

The Matrix has you...

(Tron, The Matrix, Terminator, all related)

 

[smangular]

The summary I got from this is that with exceptions, programmers suck shit. Hardware is getting better but a lot of the programs I see is still windows 95 era in efficiency.

Bloat and a lot of calling home and other behind the scene BS but no better in terms of operation or speed. These days it simply wants to hijack your computer and install crap all over the system.

What do you mean by no programing efficiency advances in 15 years? That linked article certainly does not support that conclusion and furthermore seams based on the misconception that GHz is the primary measure of performance. Do you have additional evidence?

 

[Phynaz]

I request a feature be added to the BBS software.

Fudzilla links result in an automatic temp ban.

Please?

 

[Fenixgoon]

Yet another person that thinks CPU speed is all about the GHZ

I get that at work. "Why am I only getting 3GHZ?? That's the same speed as my current PC". Yeah sure, you keep that P4 then. This i5 can go on my desk.

exactly what i thought. clock speed may have hit a practical wall, but overall processor performance has still been increasing steadily.

 

[Markfw]

I request a feature be added to the BBS software.

Fudzilla links result in an automatic temp ban.

Please?

Nope, sorry, never going to happen.

HOWEVER, it would be nice if we could all have an understanding in this forum. If you are going to quote fudzilla, make it in the title, like:


"fudzilla:The Future of Computing Performance: Game over or next level?"

Then you can hit the ignore button on your PC. The OP even did it with another thread, quoting xbit labs, on the first page here. Why not this thread ?

 

[cbn]

This post over at XS really caught my attention:

has anyone noticed where this doom and gloom article comes from ???

While I cannot say I have ever heard of the organization responsible for publishing the report, I can say with certainty that many of the committee members involved are what you'd considered experts in the field. Both sides, academia (William Dally & Mark Hill are those I'm familiar with) and industry (Intel/AMD/Nvidia/Google/etc.), are represented in the group.

Despite it appearing all doom and gloom to you, the issues at hand have been known for quite some time. The report is just summarizing them and recommending what could probably be defined as a national course of action, something targeted at policy makers. Mainly it comes down to advocating a better coordinated paradigm shift to (thread) parallelism, ideally through better education/research into algorithms, hardware and (power-)efficiency at all levels. Support is mentioned for specialized/heterogeneous computing resources as well.

Like some of the others have stated, nothing new here really.

 

[Davidh373]

The point is though, that there was no point to the article or reason for concern if it was old news. Not to mention, as we've mentioned, the 3GHz cap isn't even something to worry about (lol, we seem to be passing it in January, with 3.1GHz+ i5/i7 with turbo mode up to 3.8Ghz). As long as processor tech and programming continues to advance (I agree with God_Mode in that programming lags behind) we'll be ok. Another thought is, what is the matter with technology capping off? What would happen if Processors lasted 10-20 years without advancement? I think it would be a good thing to not have to spend $500 on your computer every 2 years to keep it up to date.

 

[OBLAMA2009]

im thinking computing is pretty close to hitting a wall as far as clock speed but fortunately computers are fast enough.

 

[StinkyPinky]

I wonder how fast they could get a single core CPU with a 32nm process?

 

[Idontcare]

Who's to say it'll still use transistors, or electric impulses?

Worry not, SkyNet will still destroy humanity on Judgment Day.

Skynet exists but hasn't destroyed us yet, we should have more faith in our benevolent machine overlord.

I'm gonna call that report BS.

The biggest roadblock in achieving such science fiction like dreams is the possibility of mankind destroying themselves before the dream ever occurs. Newer methods to overcome previous problems will likely continue to exist.

I had no appreciation for why these types of reports are drafted nor the reasoning for the agency existing in the first place until I got involved with the ITRS, R&D alliances, and NFS grants first-hand.

It is all posturing, but very necessary posturing for the sake of the industry at the behest of shifting federal priorities and all these hands jockeying for a piece of Uncle Sam's R&D budgeting.

If you can't lay claim to impending doom and gloom in your particular sector then you as a federal agency chartered with managing that sector will cease to exist in short order, and the industry itself will loose its access to internal representation at the granting agency level.

It is a necessary political game to be played.

I wonder how fast they could get a single core CPU with a 32nm process?

What ISA? Simple circuits could probably be clocked at around 25-30GHz, maybe higher.

 

[deimos3428]

There is no doubt we'll face and conquer technological hurdles in the future. That said, there's also no guarantee that we'll clear said hurdles in a timely fashion without pause. We've been fortunate in the last half-century or so in that respect. History tends to favor technological leaps over constant incremental improvement.

 

[Edrick]

Meh, we will have Quantum Computers on our desks in 10-15 years.