Author: Uri Blass
Date: 05:54:45 10/27/03
Go up one level in this thread
On October 27, 2003 at 08:43:35, José Carlos wrote: >On October 27, 2003 at 06:47:15, martin fierz wrote: > >>On October 26, 2003 at 17:34:28, José Carlos wrote: >> >>>On October 26, 2003 at 17:22:31, Uri Blass wrote: >>> >>>>On October 26, 2003 at 17:16:05, José Carlos wrote: >>>> >>>>>On October 26, 2003 at 16:51:26, Uri Blass wrote: >>>>> >>>>>>On October 26, 2003 at 16:02:43, K. Burcham wrote: >>>>>> >>>>>>> >>>>>>>Ruffian!!!!!! Perola Valfridsson!!!!!! Congratulations!!!!!!!! >>>>>>> >>>>>>>1. all time world record for least amount of time writing and understanding >>>>>>>chess code capable of beating top commercial programs. >>>>>> >>>>>>You do not know it because you do not know how much time other people devote to >>>>>>this task. >>>>>> >>>>>>>2. all time world record for keeping secret while working long hours working on >>>>>>>chess program. >>>>>> >>>>>>I see no reason to be impressed by that record and I do not know if it is a >>>>>>record. >>>>>> >>>>>>Maybe somebody works 20 years about a secret chess program and still keep it as >>>>>>a secret so you do not know about it. >>>>>> >>>>>>>3. all time world record for least amount of bugs, needed patches, best >>>>>>>compatibility with operating systems, no complaints from endusers, etc. (most >>>>>>>near perfect code in my time)(for me this, by far, is the most unbelievable) >>>>>> >>>>>>How do you know? >>>>>>If you do not release versions with bugs it does not prove that you had less >>>>>>bugs in versions that you did not release. >>>>>> >>>>>>>4. all time world record for not asking questions from other top programmers, >>>>>>>and achieving on your own. >>>>>> >>>>>>How do you know that he did not ask other programmers questions by email? >>>>>> >>>>>>>5. all time world record for winning a top chess program tournament in least >>>>>>>amount of time after release. (modern day record, and including commercials) >>>>>>>6. all time world record for highest strength level achieved in shortest number >>>>>>>of released versions. >>>>>> >>>>>>I am not impressed by these records because I do not see what is so good in not >>>>>>releasing versions. >>>>>> >>>>>>Uri >>>>> >>>>> Professionality. A good professional never releases a program with bugs. >>>>> Ruffian is a great achievement, but we shouldn't forget that we have been >>>>>witness of a probably bigger achievement. I'm talking about GCP's Deep Sjeng. He >>>>>started off with something similar to my Averno, and he has a super strong >>>>>tactical monster. Tremendous merit IMO. >>>>> >>>>> José C. >>>> >>>>I do not see why it is a bigger achievement. >>>> >>>>Deep Sjeng has more processors than Ruffian but it did not win the tournament >>>>and we have no evidence that it is better than Ruffian. >>>> >>>>Uri >>> >>> Note that I didn't say it's bigger but it's probably bigger. The reason for >>>using probability is that, of course, we don't know the insights of both >>>programmers work, but we have seen GCP upgrade his program version after >>>version, and get to a fantastic level. For some time it was an opponent for my >>>program. Then he jumped far away to a much higher level. >>> That is a great achievement. >>> >>> José C. >> >>i don't understand this argument at all. you have only seen ruffian 1.01 and >>later, but before that, very probably, version after version of ruffian got >>better and better. perhaps version 0.x of ruffian would also have been an >>opponent for your program. >>of course, both programs are great achievements. but if i had to choose, i would >>definitely pick the stronger program as the greater achievement.... >> >>cheers >> martin > > Hi Martin, > English is not an easy languaje for me so that's probably the reason of the >misunderstanding. I only wanted to express that compliments for Ruffian are >totally deserved, but we don't know much of the development of it. However we've >seen here how Sjeng has grown from a weak thing up to the top. > So basically I agree with you. > Which one is stronger is not clear so far. Until previous version of Ruffian, >DS seemed stronger according to tests results posted here (I think). But the new >version of Ruffian, after 11 games, looks really strong. > I hope it's clearer now. > > José C. Ruffian1.0.1 performed better than Deep Sjeng in Leo's tournament. I do not think that DS seemed stronger on one processor based on WBEC. Uri
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.