The Future Of Computer Technology

Frequently, here at CPF, questions are posed about what the future holds. Such curiosity is certainly understandable.

If learning more about the future of computer technology excites your interest, then you will be fascinated with the accomplishments and predictions of Ray Kurzweil, a gentleman of renowned engineering brilliance.

Some Accomplishments
Kurzweiltech.com
Spiritual Machines
Cybernetic Art
Kurzweil's Books
Kurzweil AI
Kurzweil Music Synthesizers
Exerpts from The Age Of Spiritual Machines
One of Ray's Many Honors Received

You may wish to check the book titles at the Barnes & Noble site above (or similar sites) and consider reading one or more of Ray Kurzweil's visionary books.

Take special note of Ray's predictions about the progressive growth of personal computers to processing power levels equalling that of the human brain and later equalling the power of all the human brains on earth. At some point, the machines will acquire consciousness. All at a price, in today's dollars, of $1,000. See 'The Age Of Spiritual Machines'.

Kurzweil's insights will give you a new appreciation of where our technology is going.

L3

I'm not sure of the future of computer tech., but I believe it will be a VERY long time before we see a computer come anywhere close to the human brain. Computers are simply to linear to ever achieve the level of computing of the human brain. They can only process what the are told and only understand one format at a time. Just try running Linux and windows on a machine at the same time(I don't mean a dual boot system). It can't happen. Nor does the computer have the ability to "learn" this operation by itself. Just look at voice recognition software, it's way better than it used to be, but it isn't nowhere close to the human brain. The human brain simply factors in to many elements to ever be controlled by man-made electronic processors.

Ultimately, I believe the possibility of building a linear processor or processors capable of the dimention of the human brain is still pretty far fetched at this point.

I think it's a pretty tough call.

I was in my teens when the original STAR TREK came out and I loved it, along with pretty much all other Sci-Fi. I think that the show was set about 400 years into the future. In the short span of less than 25 years many people were already using a microprocessor that could do everything that the original STAR-TREK ship's computer could do 400 years in the future. (INTEL 486 DX2 66Mhz)

I know that it was just a show and not a very scholarly effort, but to me it is a reminder of how truly difficult it is to look into the future.

Heck, we may be lucky if we have an electrical grid in the future. Who knows?

Hey I loved that movie. ILLLL BE BBAACKKK
Topper /ubbthreads/images/graemlins/smile.gif

the original star trek computer had better voice recognition than any voice recognition software running on a 486 dx2 66

[ QUOTE ]
tiktok 22 said:
Just try running Linux and windows on a machine at the same time(I don't mean a dual boot system). It can't happen.

[/ QUOTE ]

Actually it can, hang on a min...
I'm currently running linux as a native OS on my 1GHz Mobile PIII with 640M RAM. Running Windows XP under one vmware session, another instance of linux under another and Solaris x86 under another (allocated 128M to each virtual os and left the rest to the host os).
Anyway my point is modern systems can do more than what most people know or will ever use them for. In the late 80's/early90's MIT was doing neural network studies based on very simple multi processor systems on top of insect like robotic platforms. They had one 6 legged robot which could actually learn to walk. Problem is they never went any farther with it because of funding cuts (NASA funds). I started my own designs based on pic microprocessors, had a simple "neuron modules" and worked out a protocol by combining parts of the i2c but, ethernet, etc. I was building the robotic platform to test it out while in college, but graduated, got married, and have held various job supporting windows servers so I've never had the time or money to finish it. Basically the reason you don't see any real advances now is there isn't much research going on and our mainstream microprocessor systems are all setup wrong for the task.

Hi Eugene,

Understood, but wouldn't each OS still be somewhat independent of each other. Anyhow,that wasn't really the point I was trying to make.

I guess to sum it up, microprocessors can learn to walk & talk but again, this are linear type learning. Can a microprocessor learn emotion? No. Emotion and feelings are a large part of the way a brain processes information. Brains and microprocessors simply work differently. Not to mention HDD space. How many gigs is a human brain anyway? Like apples and oranges. Just my humble opinion.

[ QUOTE ]
tiktok 22 said:How many gigs is a human brain anyway?

[/ QUOTE ]

I don't know how many I've got, but I'd gladly trade a bunch of space for a better MTBF! /ubbthreads/images/graemlins/wink.gif

Sun has been muttering about asynchronous CPUs for a while now - some 20% of the circuits in the average CPU are related to clocking, and modern CPUs are really multiple processors on a single die (I think the Athlon XP has 7 distinct processing units).

Insect intelligence seems to be the threshhold of AI research these days. Their behavior can be simulated with modern programming technology.

There is a big difference between emulating a human brain (with all it's advantages and faults) and creating a system that can learn and (maybe) think. We don't know why the human brain works the way it does. We don't understand why most people have the same basic emotions nor do we understand why some things are instinctive but others are not.

We are already at the point where systems are controlled by very extensive lists of rules or knowledge based databases. It's not unrealistic to imagine a point where self modifying code (learning sofware) learns to have idle thoughts or day dreams.

A good portion of the brain appears to be memory. We don't really know how much. I can tell you that, by computer standards, it sucks. Memory recall is slow and error prone. Recent research has found that the more we think of a memory, the more we make up the details. The challange will not be duplicating the brain, but in realizing that something that flawed works.

So, Future of computers??? I think we will find a totally new way to get data into and out of them so that we can use them effortlessly in our daily lives. You don't need to see an address in order to have your computer guide you to it just as you don't need the UPC code for a box of cereal to find it on the shelf. I should not need to type in an address in order to find the nearest bank ATM. We need a way to ask the computer for info without having a screen and keyboard.

The computers we have now are extremely capable. The weakness is in the operating systems, most notably the popular resource hungry and buggy operating systems. At some point, the OS development will undergo a drastic change toward modularity, efficiency and robustness.

At some time in the future, optomized code will again be in vogue instead of the generalized routines that most programs currently use. This, by itself, will increase the computing power of systems two or three fold.


Well, that's my thoughts.

Daniel

The human brain's memory recall is both flawed and amazing. It can recall memories by association and does it at random sometimes - which can be useful when formulating new ideas.