Author: Dana Turnmire
Date: 09:51:10 10/25/03
Go up one level in this thread
Ruffian - a new chess program of the top class Chapter 1: A comet hits It was for instance in the middle of September 2002. I was not astonished badly as I the name of a chess program in a forum for the first time read "Ruffian". Now, that is actually nothing special, because in the meantime we have far more than 100 of such programs. My being astonished was directed toward the results, which the up to then unknown program had erspielt. The program was not accessible yet for all, but some testers brought almost sensational results in some computer chess forums. Thus it went some days long, until with me the patience burst (I knew that the program soon would freely be offered). "with it we want to see that ago", then I wrote in some forums and brought in such a way mean displeasure and my impatience, because the program stood for few testers to the time only for order. Before commercial programs are offered for the purchase, it is usual that them by so-called beta testers is before tested, is just as usual it that results of these programs is published (naturally the new beta program always wins). With Ruffian it did not act however around a commercial project (at least up to now not) and like that was simple my patience at the end. In this time there were many speculations, some meant in the apron that it with Ruffian a Crafty clone concerns (Crafty is a program of Professor Hyatt). Others meant even that the program are not available, because it is a computer virus. These discussions does not have I more near pursued because it in the English CCC forum to have taken place, and my English is very poor. To 23.09.2002 was it then so far: One could down-load the program Ruffian 1.0.0 in the InterNet, here the address: http://ruffian.hkust.se/ Naturally clearly that I already got myself on the first day Ruffian, and I accomplished some test immediately. In the English CCC forum immediately someone brought a contribution (addressed to Professor Hyatt) with the alluding question: "now is Crafty theclone"? I answered to it: "no, I do not believe that it a Crafty clone am, but perhaps a Fritz clone?". The definition "clone" from an encyclopedia (excerpt): "clones" Klonieren, English cloning, the production of genetically identical descendants (clones). They can by mitotic cell division (Mitose) or by transmission of the cell core of the individual who can be increased into a Eizelle (Kerntransplantation), entkernte before, to be produced. The following sentence is from great importance: the production of genetically identical descendants. It concerns thus with clones the production of identical descendants. That Ruffian however neither to Crafty nor Fritz (all the same which version one takes), is identical, which I completely easily could determine after the first tests. E.G. Ruffian cannot checkmate up to now (version 1.0.1) yet with Springer and Laeufer without the final game data bases, equally Ruffian evaluates some differently as Fritz, and also differently than Crafty! By correct "clones" therefore no speech can be absolute, even if I can prove the fact that Ruffian some techniques nearly - or uses even identically in such a way like Fritz, these techniques makes the new program therefore particularly strongly, and all other amateur programs clearly superior! As these techniques function in practice and where they with Ruffian those from Fritz are very similar, which we want to regard soon. Here however before some notes about clones like I for me understanding would like. I wrote the following statement in addition in a forum at the 24.Sept. (thus the knowledge level at that time of me, not the today's): When I tested Ruffian, he reminded me from the outset, of Fritz (against whom I had played thousands of portions) and I know really well! I know the strengths and the weaknesses of Fritz, I know its kind to play of analyzing its kind etc.. I have never stated that Ruffian a Fritz am that cannot it not be, because Fritz can e.g. with Springer and Laeufer already for a long time to checkmate (also without final game data bases) - I this matte guidance in former times often trained - Ruffian cannot do that! Ruffian can be thus no Fritz. Nearly all analyses nevertheless remind me "of the style" from Fritz, whether that is the kind of the search, the evaluation, search depth and, and..., everything reminds me of Fritz. I do not see anything, where I can say that in the program Ruffian is something "original"! Sorry I cannot do anything for it that I to this result comes, but is personal for me now once so. Whether therefore into Ruffian a code of Fritz is, or whether the programmer did not only take over many characteristics of Fritz (thus not stolen), that has for me as a user a meaning (at the most for programmers). I know many programs, and all have somewhat strange, thus something the one other program do not have! With Ruffian I do not see anything such. A Shredder has its own style, also the kind of the evaluation is singular. A ChessMaster (at least with the Standardsettings) plays singularly. The same meets just as too on ChessTiger and junior! Yes, and the Fritz was up to now singular for me likewise, until now Ruffian came. We have 2 programs the one similar style to prefer beautifully, now, similarly analyze etc.. Fact is that there is Fritz already for a long time and so far does not have tried, its program to only arrange in such a way that it similarly plays as the Fritzi. With Ruffian we have now a program, which can thus muster for instance the play strength of Fritz 6 (to quite also defeat knows a Fritz 7) and in addition still with similar style plays! Contrary to Fritz, Ruffian (when stock - and UCI engine) can be merged under many other GUIs, and with the style similarly of Fritz including the play strength of Fritz 6, it will conquer many hearts there it a Freewareprogramm is. In the usual rank lists Ruffian becomes 1.0.0 anyhow all past amateurs over wings. It gave last to admire such a thing in the year 1996, when the amateur Shredder in Jakarta got the WM title, and many hearts to conquer could. After now 6 years, now again a similar throw (quasi 0 to 100) succeeded to a programmer, the play strength after even more strongly to settle than at that time with Shredder 1. On this statement I got and others also the following answer: "actually amazingly that so far still nobody tried Fritz to copy. Ruffian comes just as fast on depth as Fritz owing to its excessive Prunings. Thus it at first attempt all amateurs ueberfluegelt. I do not know naturally, what gezaubert still in such a way by Ola everything into its program, but that aggressive Pruning and aggressive zero-move attitudes the play strength show now not only Fritz, but evenly also Ruffian positively affect." I answered to it in such a way: "after Ruffian is, must one now there this question seriously place itself. But I assume, how it already suggested Ulrich that each programmer wants to create its own creature. I on the idea would never come something would copy, I search always mean original way that Ruffian not Fritz am, would like I to again stress. It actually concerns its own program. But the programmer took over much of Fritz ideas, otherwise Ruffian would be like that as he now is not possible,!" Some short information still to Ruffian, before I turn to the practical side of the program: The programmer comes from Sweden, is 36 years old (one read it in such a way) and is technician of telecommunications. Its name: By Ola Valfridsson. Why is Ruffian called in such a way? One does not know it, but Ruffian is a black horse. When did the project Ruffian begin? 1998. Which minutes does Ruffian support? Stock I, stocks II and UCI. Under which operating systems does the program run? Linux, Bd, OSX and Windows. How much does it cost? At present the version is 1.0.0 and 1.0.1 Freeware. One can down-load the following versions under the address mentioned (at present): ruffian engine Windows version. Release 1.0.1. (September 29, 2002) latest ruffian engine Windows version. Release 1.0.0. (September 23, 2002)
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.