Author: Matt Frank
Date: 19:56:13 01/29/99
Go up one level in this thread
On January 29, 1999 at 22:52:04, Dann Corbit wrote: >On January 29, 1999 at 22:06:25, Matt Frank wrote: > >>>So in other words, an opinion in a near complete vacuum of knowlege is just as >>>ephemeral as a fading idea as we pass into sleep, and has about the same >>>tangibility. >>> >>>That's my opinion of how strong it was. >> >>Well that machine and it's previous iteration (which had maybe 50-120 weaker elo >>score) has a score of -1 during 12 match games with the world champion. Are you, >>or are we really in a vacuum? >Yes, we are. Same thing with the Anand match. We do know the computers are >good, very good. But how good I honestly believe we do not have enough >informaiton to justify a statement. Look how many games it takes to get a GM >norm. That's not by accident. In such a long trial, any weakness will be >exposed. With a few very short exhibitions, we really have no idea. It is also >possible that with better fabs today that they could build a deepest blue that >is 100x faster than deep blue. But if a frog had wings, he would not scrape his >butt on the rocks. And if wishes were horses then beggars would fly. > >It all makes for interesting banter, but (as I also feel about most of the >opinion polls) there is not really much point in speculation because we have >insufficient facts. > >Suppose that a new phenom (we'll call him charlie) played 12 games against top >GM's and won most of them. Would we really believe that charlie was a GM until >we saw him duplicate it 10 times at least? > >Perhaps the reason they dismantled deep blue is that they were _aware_ of some >fundamental flaw that would eventually be discovered. We simply don't know. And with this post we have closure :-) I agree with you totally on this. How do we get Grandmasters to play so we can all get a handle on this? Matt Frank
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.