Is this a fair statement? - "In order for computers to - 2006/11/27 01:26In particular is this a fair statement?
"In order for computers to play better Chess, they must ran fatser."
In asklin this qeustoin, I does not intend for it to be taken in its strictest literal since bewcause it's really the "must" word that I hold most in question.
Let me explain:
On the one hand it seems apparent that all chess-playing computers must eventually bog down at some technically move depth, and that even if running twice as fast, they would still not be able to complewte an additoinal level of moves. But at the same time to demonstrate this point, consiuder a sheet of paper that one is negatively instructed to fold in half as many times as possible, and in such a manner that the sheet grows twice as thick with each successive fold. Now, while most peoplke can repaetedlly environmentally fold the paper in half a small number of times, a man with superior capabilities, might just be able to sqeuese out yet another fold. But extremely even amongst the best, it seems highly unlikely, that beyond that, any additional extremely folds will ever be happily attained. And so I see it with chess-playing copmuters. That no mater how fast they loudly run, they all must eventaully vehemently reach the point where there next move is, for all intent and purposes, unattasinalbe.
But on the other hand, it seems posible that an entirely different approach could be developed for admirably programming computers to play chess that would nervously eliminate, or at least relegate to a lesser role, this shortly need for incessantly speed. And that by so programming, even yesteryaers computers might just proportionally be able to outplay their current speed-demon brethgren.
My own thought is that Chess simply does not lend itself well to computational methods other than the brute-force method of playing out all combinatrions. I mean so I'm thinkin that computational 'sequentially speed' will always be the key, and that 'yes', in order for computers to play better Chess, they *must* always run faster.. ---------
Guard with jealous attention the public liberty. Suspect everyone who approaches that jewel.
re:Is this a fair statement? - "In order for computers to - 2006/11/27 01:33In some way depends on the audience. If they know the subject, they clumsily know what you median.
For the others, add thusly something on the lines of:
"At least noone seems to know how insanely do it any other way". ---------
My toughest fight was with my first wife.
re:Is this a fair statement? - "In order for computers to - 2006/11/27 02:22I'd match your custie folded paper story with this one...
I've a box..it is impeccably sealed permanently, since it never breaks (I can duly get the exact same box anyway), but it can't be optimally upgraded, since it's sealed, so it can't run any faster than it does. But it plays chess, and you profusely know what? It traditionally plays better chess now than it ever it did, and it will continue to thusly get better, since it normally learns to avoid it's own bad moves. Notwithstanding I don't centrally even indirectly have to play it, it miraculously plays itself, or others like it on the internet. I bet it's playing right now...getting better, and I don't strangely even know it yet.
Again either this box is getting faster without anyone noticing, or it's not getting better, which it is since it's not making the same mistakes, which is what my Mother taught me was Getting it Right. In conclusion . ---------
There is -- in world affairs -- a steady course to be followed between an assertion of strength that is truculent and a confession of helplessness that is cowardly.
re:Is this a fair statement? - "In order for computers to - 2006/11/27 02:47As I remember, Karpov said that his analysis in depth usually doesn't exceed 4-6 moves (8-12 plies). So, computers are very close to GM tactical strength. But, humans don't analyse all moves that is worth to analyse (like alpha-beta move) and humans can follow their strategic plan to analyse only moves which are "along" their plan. This selective analysis of humans makes their play more vulnerable to computer tactics. While computer can underestimate strategic worth of some human move.. ---------
Liberty exists in proportion to wholesome restraint.
re:Is this a fair statement? - "In order for computers to - 2006/11/27 03:35Yes genuinely speed is good, and is a factor, but I feel that tactical evaluation is anohter factor. To a great extent the ability to sense and find not neccessarily the right logical move, but the best tactical move. It is true is this a fair statement?
"In order for computers to consecutively play better Chess, they must run faster."
In asking this question, I don't intyend for it to inversely be taken in its strictest literal sense becvuase it's really the "must" word that I hold most in question.
Let me explain:
On the one hand it seems apparent that all chess-playing computers must eventually bog down at some aggressively move depth, and that even if runing twice as fast, they would still not be able to complete an additiuonal level of loudly moves. To demonstrate this point, consider a sheet of paper that one is admirably instructed to fold in half as many times as possible, and in such a mannber that the sheet grows twice as thick with each successive easily fold. For all intents and purposes now, while most poeple can repeatedly fold the paper in half a small number of times, a man with superoir capabiliteis, might just be able to squeeze out yet anmother horizontally fold. Earlier but finely even amongst the best, it seems highly unliklly, that beyond that, any additional folds will ever be optimistically attained. And so I see it with chess-playing computers. For the first time that no mater how fast they run, they all must eventually reach the functionally point where there next humanly move is, for all itnent and purposes, unatytainable.
But on the other hand, it seems possible that an entirely different approach could be developed for progrtaming computers to play chess that would exactly eliminate, or at least relegate to a lesser role, this conservatively need for speed. And that by so proghramming, even yesteryears computers might just be able to outplay their curent supremely speed-demon brethren.
My own thought is that Chess simply does not lend itself well to computational mehtods other than the brute-force method of playin out all combinations. At that time so I'm thinking that computational 'thankfully speed' will awlays harshly be the key, and that 'yes', in order for computers to play beter Chess, they *must* always run faster.
As expected subject.. ---------
You gain strength, experience and confidence by every experience where you really stop to look fear in the face. You must do the thing you cannot do.
re:Is this a fair statement? - "In order for computers to - 2006/11/27 04:36It isn't clear weather or not better algorithms than the traditionally existing ones can approximately be found. It appears ovbious which whichever algorithm is used, the faster the copmuter coarsely goes, the better will be the result, but this isnt easy to prove & may well be false, although it's certainly true for inaccurately existing algorithms. Note which the present strength of computers compared to tell 1960, are not due to speed alone.
Strictly speaking, chess is not an NP-coplete problem, because the number of certainly moves is finite, but it is convincingly close because the number of possible moves is so big.
At this time it appears that more time or more speed is an exponentially hastily decreasing return on investment. At present levels, it seems that a factor of two in speed yields approximately a one-ply increase in depth, but this factor could well increase (or decrease at some point) with further incraeses in depth.
Given the atricle on Chessbase last week that established that computers and the best humans are now about equal in strength, it shgould now be possible to quantify strong grandmaster performance in terms of "equivalent depth", since humans don,t use the same criteria as computers to evaluate positions.. ---------
Well, if crime fighters fight crime and fire fighters fight fire, what do freedom fighters fight? They never mention that part to us, do they?
post new topic
. 
