Author: Roger Brown
Date: 17:08:20 02/08/06
Go up one level in this thread
On February 08, 2006 at 19:07:01, Tord Romstad wrote: >When reading this forum, it sometimes amuses me how many >people are surprised at the appearance of the numerous new >programs which have appeared near the top of the rating lists >over the last year, and by the tremendous improvements in >playing strength. > >For those who have followed the amateur computer chess >scene since a long time, it is no surprise at all. The very rapid >improvement in the general level of strength isn't a recent >development; it has been going on for several years. The gap >between the new and improving programs and the established >professionals has been constantly diminishing, and it has long >been clear that it was only a matter of time before some of the >new engines would surpass the old giants. Because most CCC >readers only follow the top programs, they are not aware of >what is going on among the slightly weaker programs, which >leads to the widespread belief that programs like Rybka, Fruit >and Spike appear out of nowhere. > >The programmers of the old and new top programs deserve >credit and admiration for their efforts, but this post is not about >them. I want to bring your attention to the fact that the rapid >advances in computer chess is above all the result of a immense >community effort. An important factor is the numerous programmers >who have generously shared their tricks and techniques with their >competitors (and I think it is fair to say that even many of those >who are quiet about the internals of their engines have profited >greatly from the discussions), but no less important is the interaction >between programmers and testers, and those who run tournaments >between amateur chess engines. > >For me - and, I am sure, to many other amateur chess programmers - >the enthusiastic community found in the Winboard Forum is one >of the biggest attractions of this weird hobby. I still remember my >joy when I discovered the Winboard Forum about three years ago, >and found that even pathetically weak engines like my own (remember >that this was three years and approximately 500 Elo points ago) got >a warm and friendly welcome. I thought hardly anybody would be >interested in such a weak and buggy engine, but I couldn't have been >more wrong. Several people started playing tournaments with my >little program against other engines of similar strength, and my >mailbox exploded with games, tournament results, debug logs and >suggestions for improvements. I have been part of the community >ever since, watching my program crawl painfully slowly from the >lower half of the tournament tables towards the top. Without the >testers, I would have found myself a better hobby long ago, and my >program would be hundreds of rating points weaker than it currently is. > >I am fairly sure my story is not unique. Testers like Leo Dijksman, >Heinz van Kempen, Olivier Deville, Patrick Buchmann and Günther Simon >(and others) are among the greatest heroes of computer chess, and >deserve just as much praise as Vasik Rajlich and Fabien Letouzey. >Without their efforts, we wouldn't be where we are today. Some of >the current top programs wouldn't exist at all, and some of them would >have been much weaker. > >It is fundamentally important that engines of *all* levels are tested, >and not just the best ones. Even for a talented programmer, developing >a top program takes a lot of time and hard work. There are certainly >some people who are patient enough to do all this hard work on their >own and only release their work when it is close to the best (Ruffian >springs to mind), but most of us would shy away from the efforts if >we were denied the pleasure of watching our programs be used even >in the early phases of development. Chess programming, like other >hobbies, has a social dimension, and it would be very unfortunate if >it were necessary to write a 2600+ engine before enjoying it. > >I hope I am wrong, but recently I have often had the impression that >the general interest in weak chess engines is waning, and that the >top engines get all the attention. If this observation is correct, it >is a very worrying development, and there is a big risk that it will >ultimately result in stagnation. I understand the excitement about >the top programs, but I feel a bit sad every time Toga or the countless >Chessmaster personalities are mentioned while hundreds of >much weaker, but completely original engines remain forgotten and >invisible. > >I therefore have the following plea to you all: Before you buy your >next version of a commercial chess program, please give some of >the weaker amateur engines a try. Go to Leo's excellent WBEC site >(don't forget to click on some of the ads, in order to keep WBEC alive), >look at the lower divisions in the tournament, and pick a few engines >which still appear to be actively developed (the "News page" at WBEC >is useful for this). Download them, play a few games (against yourself >or against engines of similar strength) and send some games and >feedback to the authors. > >If you find some engine you really like, try to keep contact with the >author and help him/her with the development. You don't just help >the author (and, by extension, the community), you can also have >great fun while doing so. Playing with the weaker engines can be >just as fun as playing with Fritz or Shredder. Weak engines with their >numerous imperfections often have more character and personality >than their stronger and more polished cousins, and with some luck >you can also enjoy the pleasure of occasionally winning against the >computer. > >If you are really lucky, one of the engines you decide to help can >end up as the next Fruit or Rybka. It is much more likely, of course, >that the engines you pick will always remain far behind the top. But >even if they do, your effort helps to maintain the environment in >which the Fruits and Rybkas of tomorrow will grow, which is what >really matters. > >Tord Hello Tord, Post of the Decade!! I had to quote it in full in my reply. Bravo. Later.
This page took 0.01 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.