mersenneforum.org  

Go Back   mersenneforum.org > Extra Stuff > Blogorrhea > jasong

Reply
 
Thread Tools
Old 2012-09-09, 08:57   #1
jasong
 
jasong's Avatar
 
"Jason Goatcher"
Mar 2005

66638 Posts
Default Is Moore's Law going to die soon?

Because of the attainment and reapplication of knowledge, things tend to get cheaper over time. Barring some sort of disaster, with a very loose definition of disaster, things just tend to get better as time passes. Food production gets more efficient, so people can suddenly afford to pay extra for quality control, which they wouldn't have tolerated a 100 years ago because food was already expensive.

In the case of electronics, improvements have occurred at a rate that is unrivaled by any industry that I'm aware of. The problem of course, is that it's not really possible for things to improve infinitely. Even if we assumed things would continue to double at least once every two years, that would mean that the entire known universe would be saturated with information within the next thousand years.(2^500)

The question is when will Moore's Law die and how will it happen? Will improvements come to a screeching halt, where one year it's business as usual, and the next prices are crashing because we've reached the limits of atoms? Atoms aren't a true limit, in the sense that they're not literally solid blocks of "stuff," but our breakneck technological improvement rate can't go on forever. In a world where gadgets tend to get twice as fast every couple years or so, for the same price, how will people respond to technology if the improvements were reduced to, say, 10% a year? Obviously, upgrading would happen a lot less. Tech support would probably be a bigger chunk of the pie simply because there would be less desire to actually be sold gear. The gaming industry would probably have to change their model since they'd find it harder to get people to buy stuff based on awesome graphics since games would have about the same graphics for years.

I personally think it will be a good thing for actual productivity, there'd be less bit rot, so people would have tons of time to actually deal with bugs in software. The environment would be better off because people wouldn't be throwing out nearly as much gear each year because of upgrades. Other technologies could continue to get better, I'd imagine tvs would still tend to become cheaper based on new fabrication methods that might be in development right now.

So how do you think it will happen, and what do you think people will do differently when it does?
jasong is offline   Reply With Quote
Old 2012-09-09, 09:49   #2
ET_
Banned
 
ET_'s Avatar
 
"Luigi"
Aug 2002
Team Italia

10010110100112 Posts
Default

AFAIK, Moore's "law" has to do with the number of transistors packed in a defined area...

Luigi
ET_ is offline   Reply With Quote
Old 2012-09-09, 09:50   #3
xilman
Bamboozled!
 
xilman's Avatar
 
"π’‰Ίπ’ŒŒπ’‡·π’†·π’€­"
May 2003
Down not across

2×5,393 Posts
Default

I suspect that your model of the future is too simplistic. No offence intended because I suspect that everyone's model, mine included, is too simplistic.

Here are some possibilities which may completely invalidate your economic model.

What if smart (note, not necessarily intelligent) matter becomes ubiquitous? Some of the infamous blue/grey goo scenarios posit that essentially anything material can be made as easily as any other object. Other than perhaps for sentimental reasons there would be no reason to regard anything as more or less valuable than anything else. Note that non-material resources may continue to be more or less valuable Such resources may include cultural constructions such as art works, scientific research and conversations; they may include genuinely inter-personal services (as distinct from similar services provided by machines) such massage, hair-styling, sexual activities, food& drink preparation, and companionship.

What if the advances in IT continue to improve long enough for genuinely intelligent machines to become commonplace? Would they continue to adhere to our present ideas of how to run an economy? What would humanity's role be in such a society? How about the roles of machine intelligences less intelligent than us? Machines of comparable intelligence to us, or smarter than we are but way behind the most intelligent?

It is far from obvious to me that either or both of those scenarios (and several others for that matter) won't occur well before Moore's law comes to a significant slow-down.
xilman is offline   Reply With Quote
Old 2012-09-09, 19:28   #4
petrw1
1976 Toyota Corona years forever!
 
petrw1's Avatar
 
"Wayne"
Nov 2006
Saskatchewan, Canada

10010010101112 Posts
Default contrast

Let me contrast the above long well thought comments with: "No"
petrw1 is offline   Reply With Quote
Old 2012-09-10, 07:25   #5
retina
Undefined
 
retina's Avatar
 
"The unspeakable one"
Jun 2006
My evil lair

22·32·173 Posts
Default

Quote:
Originally Posted by jasong View Post
So how do you think it will happen, and what do you think people will do differently when it does?
It will play out in much the same way that the car has. People will start to look for styling and personal lifestyle fit. If you can't differentiate by performance or efficiency then there is little left to make a decision other than the intangible things.

Does it come in pink? I prefer the one with the charger socket on the left side.
retina is online now   Reply With Quote
Old 2012-09-10, 23:25   #6
jasong
 
jasong's Avatar
 
"Jason Goatcher"
Mar 2005

1101101100112 Posts
Default

I'm going to go back and read the thread after I write this, but when I say Moore's Law, and I know this contradicts the official definition, I'm talking about the rampant growth in computational ability. I'm basically considering the possibility that things could rapidly and quite suddenly start progressing at a much slower rate, a rate more akin to medical knowledge growth or the ability to grow food.
jasong is offline   Reply With Quote
Old 2012-09-10, 23:38   #7
jasong
 
jasong's Avatar
 
"Jason Goatcher"
Mar 2005

3×7×167 Posts
Default

Quote:
Originally Posted by xilman View Post
I suspect that your model of the future is too simplistic. No offence intended because I suspect that everyone's model, mine included, is too simplistic.

Here are some possibilities which may completely invalidate your economic model.

What if smart (note, not necessarily intelligent) matter becomes ubiquitous? Some of the infamous blue/grey goo scenarios posit that essentially anything material can be made as easily as any other object. Other than perhaps for sentimental reasons there would be no reason to regard anything as more or less valuable than anything else. Note that non-material resources may continue to be more or less valuable Such resources may include cultural constructions such as art works, scientific research and conversations; they may include genuinely inter-personal services (as distinct from similar services provided by machines) such massage, hair-styling, sexual activities, food& drink preparation, and companionship.

What if the advances in IT continue to improve long enough for genuinely intelligent machines to become commonplace? Would they continue to adhere to our present ideas of how to run an economy? What would humanity's role be in such a society? How about the roles of machine intelligences less intelligent than us? Machines of comparable intelligence to us, or smarter than we are but way behind the most intelligent?

It is far from obvious to me that either or both of those scenarios (and several others for that matter) won't occur well before Moore's law comes to a significant slow-down.
That is a definite possibility, the idea that machines could advance beyond us in intelligence. When Gordon Moore and other men-in-the-know made their predictions about the continuance of Moore's Law, the common theme seems to be "it will continue for another decade or so." Obviously, it's continued for 50 years and the nay-sayers have been wrong for all those 50 years. It could easily grind to a halt in the next decade, but it's impossible to know about a discovery until after someone makes it. I read an article about spooky action at a distance(I can't for the life of me remember the proper term) and scientists were detecting conditions without changing properties, and doing it at level where a couple years ago it was believed that you HAD to be willing to change the properties in order to detect things.(I butchered that explanation, sorry)

So if we discover a way to seemingly instantaneously communicate with hardware that is anywhere from a few feet to thousands of miles away, then that in and of itself opens up a huge number of opportunities, from low lag gaming to unheard of computational ability. Plus, of course, the Mars Rover Curiousity could climb that damn mountain in hours instead of months.

It sort of gets away from the subject of Moore's Law, but it keeps with the topic of rampant computational growth.
jasong is offline   Reply With Quote
Reply

Thread Tools


Similar Threads
Thread Thread Starter Forum Replies Last Post
Moore's law dead? Or just in need of an update. jasong Lounge 17 2018-08-08 13:03
The status of Moore's law: It's complicated jasong Lounge 25 2016-06-22 02:05
Moore's law again, I'm nothing if not predictable ;) jasong jasong 8 2013-02-20 10:40
Moore's Law? jasong Lounge 41 2007-12-16 06:25
The myth of Moore's law, why is it prevalent? jasong Hardware 15 2005-09-13 22:18

All times are UTC. The time now is 23:23.


Fri Aug 6 23:23:05 UTC 2021 up 14 days, 17:52, 1 user, load averages: 3.64, 3.93, 3.99

Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2021, Jelsoft Enterprises Ltd.

This forum has received and complied with 0 (zero) government requests for information.

Permission is granted to copy, distribute and/or modify this document under the terms of the GNU Free Documentation License, Version 1.2 or any later version published by the Free Software Foundation.
A copy of the license is included in the FAQ.