If anything, I think it tells us is that a lot of ground has been covered since then. Alpha-beta-based search techniques grew to completely dominate computer chess--nearly all successful techniques discovered or invented since then have been incremental innovations in the context of alpha-beta-based programs. We're now up to the point where a new engine has to implement many of these specific enhancements in order to be a strong competitor. Perhaps, like specialization in any branch of science or mathematics, the deeper you go, the more esoteric it becomes? Perhaps computer chess programmers have mined most of the gold from alpha-beta techniques? There will probably continue to be improvements, like the LMR techniques that have become so popular lately, but nobody has shown anything wildly different from alpha-beta which could compete with it in overall effectiveness at playing chess. I hope SJE's efforts with Lisp will bear fruit, but even if they don't, I think its important and useful that he's trying something radically different (instead of "running with the pack", i.e. making yet another alpha-beta-based engine).sje wrote:Back in the Old Days when one could play with Chess 4.x on a 1960s CDC 6500 console, that program could scale its transposition table requirements down to a mere 256 entries. Each entry took a word and a half (96 bits) of ECS (Extended Core Storage). On that CDC CPU model, the program only scored a few hundred positions per second. Close to, but not quite as good as what I was able to do with a 1986 Macintosh Plus running at 8 MHz.Guetti wrote:6) 4 MB? This leaves not much space for the transposition table.
And yet, back then in the 1960s and early 1970s with punch cards, paper tape, and no Internet, the relative paucity of chess program authors seemed to produce new ideas and discoveries rather faster than is done today.
Does this tell us something?
What bothers me is that as computing resources continue to grow, computer-vs-human chess will basically cease to be interesting (if it hasn't already). Will the engine authors then continue to compete just to see who can make the strongest computer player? Perhaps some of them will switch to writing engines for Go instead, but many people have invested a lot of effort into computer chess, and into their existing alpha-beta-based engines.
Some people here have invested 15 or 20 years in it already... I am not a chess player, so I can't truly appreciate what drives most of you. But I think there must be something marvellous about chess that it still compels your interest after so many years! What motivates you the most? Is it interest in the game of chess itself? Or the fun of competing against each other? Or the satisfaction when you find something that improves your engine's playing strength?
I'm also curious, do you guys still expect to be doing computer chess a few years from now? It feels like there are more engine authors than ever before, but does computer chess still feel like "undiscovered country" to you guys? I mean, do you feel like explorers going where no man has gone before, or do you feel more like laborers digging up small bits of hard-earned gold that was surveyed ages ago?
Are you interested in continuing, and pushing it as far as it can possibly go? How close do you feel we are to the goal, of making a chess engine "as strong as humans can make it" ? Do you expect that changing technology will make current designs sub-optimal, so that there will still be lots of room for discovery and tweaking your engines to make them more optimal on the technology of the day?
Sorry for all the questions. If anyone wants to, just answer the question(s) which you would most enjoy answering.
