Author: Komputer Korner
Date: 00:03:12 11/27/98
Go up one level in this thread
On November 26, 1998 at 20:09:54, Amir Ban wrote: >On November 26, 1998 at 17:44:11, Ralph E. Carter wrote: > >>Do not read the following if you don't like chess gossip. >>Others are off-topic continuously- here are my two cents. >> >>[My comments are in brackets.] >> >>CASE A: >> >>1) Mr. Vasquez said: >>>>>>I found that after overnight analysis (depth 19 and over) Junior 5.0 shows >>>>>>scrambled lines. >>[It is good and helpful to find bugs.] >> >>2) Mr. Ban said: >>>>>All the above is about how the line is displayed rather than about the line >>>>>itself. >>[The actual ANALYSIS is not scrambled, it is only a graphics problem.] >> >>3) Mr. KK said: >>>>...we have no reason to doubt you...the proof will be whether the problem shows up in Nimzo 99 as well. >>[Chess programmers sometimes make mistakes, let me investigate...] >> >>4) Mr. Ban said: >>>Korner, you assume I'm correct about what ? What is there to doubt here ? >>>You are not replacing Thorsten, are you ? >>[You seem to be calling me a liar. It is wrong to accuse me of lying without >>proof. You can be ejected from this club for that.] >> > >There's some paranoia going on here, so a clarification: > >You should read my question to KK literally: I was surprised by what he said, >because I was not aware of saying anything that requires trust. I'm not really >sure what he meant, but my three top guesses are: > >(1) I said I'm only involved with the engine, and not in the other stuff. >(2) I said the music problem has not been reported yet, so far as I know. >(3) I said the scrambled line is a display not an engine problem. > >I thought (1) is common knowledge. I can't prove (2) right now, but I don't >think it's important either. I thought (3) was evident from the description of >the problem, but maybe I was wrong there, so I'll explain why: > >The Junior engine transmits lines to the Junior5/Fritz5 shell in binary form in >from-square/to-square notation. If you understand that, you realize that it's >not possible for the engine to have a bug in who's to play, in which moves are >check, or in which piece moves, because that's not part of the information it >gives. I don't care whether this shows up on Nimzo99 or not, this is not an >engine bug. > >The object of all this, by the way, is not blame evasion, but sorting out who's >in a position to handle the complaint, a routine step in tech support. > > >>CASE B: >> >>1) Mr. Kasparov said: [The DB team cheated.] >>2) Mr. Ban said: [The burden of proof is on the DB team.] >> >>THE POINT: >>If you and others are going to perpetuate case B in this club, on and on, ad >>infinitum, I do not care to hear about your indignation about the insult you >>thought was made to YOU in case A. > >I was not expressing indignation, and was not asking for your sympathy. Yours is >the first post in this thread that I do find offensive. > >Amir I have run Junior 5 overnight and have found 2 exact lines when using scroll main lines. The 2 lines have different total nodes with different times of course but the PV and depth was exactly the same. My window had enough space for further plies but the lines both ended at the same ply. Perhaps there is something wrong with the way that the 32 bit GUI displays things. But there is something wrong. Amir, this is affecting your engine and since Junior 5 is involved one would think that you would want to get to the bottom of this. I haven't found mixed up lines like the other poster but I agree there is something wrong. -- Komputer Korner
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.