Author: Drexel,Michael
Date: 15:11:23 01/09/03
Go up one level in this thread
On January 09, 2003 at 09:08:47, Russell Reagan wrote: >On January 08, 2003 at 07:46:08, Drexel,Michael wrote: > >>I knew that. doesnt matter at all. this is not a serious calculation because >>Moores law is not realistic (physical reasons). > >Why are you so pessimistic? Moores law may not hold true FOR THE CURRENT >TECHNOLOGY, but that is awfully stubborn to think that in this day and age that >no new technology is going to be born within the next 10,000 years. So while >Moores law may not hold true for current technology because of physical >limitations, you can't say it won't hold true. I believe apocalypse is coming soon. :-) > >Everything you have said regarding this gives me the impression you don't think >a single technological innovation will happen in the next 10,000 years, and what >we have now is it, forever. You deem things impossible based upon the >limitations of only what you can see. The problem is you don't know what you >can't see, and you can't understand that for some reason so you discard it and >think you've got a good grip on things. > >And yes I know what 10^43 means. Do you? Because I don't think you do. Think >about how many of those positions will be a rout. Why bother searching a >position and it's huge subtree when one side is up by 6 queens? A good chess >program would prune these things away. That number is not the same number of >nodes that would need to be crunched to find a mate. So go ahead throwing that >number around all you want. It isn't vital for solving chess. as I said I refered to a mathematical proof. not evidence. even a position with up 6 Qeens could be draw or even lost. there is no realistic way for a prove regarding computerchess like it is done today. and this was my point. you cant prove it with successive approximation. Computer evaluations like +5.00, +10.00 or even +16.00 are NOT ALWAYS a win! > >Even if you must hold on to that number, it will only take 60-70 years of it >holding for 10^43 positions to be searchable. Maybe current technology won't >hold for that long, but if you think there won't be many more breakthroughs and >new technology created in the future, then I have a very nice bridge for sale >that you might be interested in... For a prove you need to foresee the end by checking each possible path to its outcome. quantum computers might be able to do that (not in the next future). there should exist a real mathematical prove too. however unimaginable complex. both ways have nothing in common with todays search methods. > >If we have three "breakthroughs" and we get another 20 year period of Moore's >law, guess what, we're there. Or how about just two 30 year periods? Maybe it >will only take one. Who knows? Not you. Lets talk again after two 30 year periods about that topic. OK? I dont even see why we should have to prove something what seems by evidence clear.
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.