Author: Ralf Elvsén
Date: 04:40:06 08/17/99
Go up one level in this thread
On August 17, 1999 at 04:21:48, Thorsten Czub wrote: >On August 17, 1999 at 01:41:13, Ralf Elvsén wrote: > >>Hi Thorsten >> >>The problem I have with your posts about CSTal is your >>(over?)enthusiasm and the unrestrained ovations to its >>style and strength. > >IMO you read something in my posts that is not there. >I like CSTal. no doubt. but i do in the same way like >hiarcs or mchess or rebel or chess-tiger or zarkov or >wchess or isichess and other programs too. > Ok, it was never there... Maybe a more measured way of writing would prevent people (sorry, just me?) from getting that false impression, but I guess that is not your style. >>As I remember it (and correct me if >>I'm wrong) you posted three games vs Hiarcs 7.32, all won by >>CSTal. One win was due to a mistake by Hiarcs other people >>couldn't reproduce. I also can't recall you saying that these >>were the only three games the programs had played (but then >>again, I don't have the post in front of me). > >when i post 3 games i have played 3 games. >i do play many games. in the moment i do play cstal-anmon. >When i am playing x vs. y i cannot in the same time play >x vs. z. That is obvious and not at all something I asked for. >I created a style that overstresses the tal-attack because hiarcs >has still a king-safety problem. > > >>I am not saying that you are giving us flawed, selected >>games. It just would be nice if you added something like >>"...I have played these three games only..." and "...due >>to an inexplicable move by Hiarcs in one game..." . Then I >>would give more attention to the results (since I absolutely >>have no reason to believe otherwise than that you are a honest >>and credible person), and (who knows) I might buy a copy of >>that excellent program :) > >it is not my intention that you should buy something. >i want to know about the programs. buying them is a very >nice thing, but also a very passive consumer thing. > I was under the impression that you had a more or less organized cooperation with Chris W. Granted, such cooperation can take place without you being interested in the commercial success of CSTal. If I misunderstood this I sincerely apologize. >>As for Hiarcs having kingside attack-problems >>as you mentioned, that is true. > >so - what is your problem. when i am posting the real-things, >and you believe the same, you only wonder because i got my results >in a different way YOU got them? >Look - you play with the programs and get your results. >i do the same. >i don't need many games to find out something. >i have created a special style against hiarcs. >no wonder that this style gets a 2-0 against hiarcs, because in my >style i have stressed the king-attack sliders. therefore tal using it >plays crazy king-attacks. since you and i come to the same conclusion >that hiarcs has a problem with this, it is only logical that >i get 3-0 and others 2-0. >you would find it better if i would have played 20 or 50 games ? >why ? >because than you would better believe the impossible ? 2-0 is not impossible (obviously). 20-0 is not impossible either, but very improbable (not that you implied such a superiority). I guess you and I just have different opinions on what conclusions can be drawn from a few games. Let us continue to stay pleased in our two parallel universes. Btw, what ARE your conclusions from these games? CSTal is clearly better than Hiarcs? Hiarcs have kingside-problems? Ah, I see below: "all i want is to show that cstal is able to play good results against the top programs IF you give it fair conditions". Ok, if that is your claim, I will consider the hypothesis worth testing, although not yet verified or rejected. I personally would be glad if the "CSTal-approach" would prove to be succesful in this respect. It opens up doors for (in my opinion) more interesting programming questions than the "usual" stuff. Too bad the programming techniques seem to be so secret (?) >because presenting 2 games gives you the unconscious feeling that >my data is not right. ok . i do understand. Thanks :) ...eeh, substitute "too few" for "not right"... >you need to get more data to believe. you are a believer in the amount >of data. >i have nothing with you believing in quantity. but i work in a different >way. i don't eat much fast food, and believe it would do me a good thing, >i like to eat good food, of high quality. fresh salad and no fast food. >if you believe that your mass of fast food would be better for your >body, because it is a bigger amount of it, than that is YOUR believe. >please see this as a metapher. as an image for the amount of chess-games. > I think I understand your metaphor, although I consider it to be slightly insulting :) > >>If you can exploit >>it easily(?), that is interesting. > >>And again, no insinuations intended, just some friendly >>advice for your "marketing campaign" :) > >there is no marketing campaign. >i like mark and his program. in the same way i like chris and his program. >and others too. >i work with anybody i like and i have fun and we get good progress. >no matter how strong his program is. >Remember that predecessors of cstal were not much strong. >all i want is to show that cstal is able to play good results against >the top programs IF you give it fair conditions. >If you give the slowest NPS-program the slowest CPU (Cyrix-bullshit e.g.) >you will get bad results. Ok, I think I know better where you stand now. Thanks for your answer Ralf >> Regards, Ralf
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.