|
< Earlier Kibitzing · PAGE 699 OF 707 ·
Later Kibitzing> |
| Jan-09-15 | | ajile: <AylerKupp:> I think you missed the point here. The point is for us to have fun while we wait for another game. And by "we" I mean the whole Team not just 2 people in a scientifically perfect lab bubble. |
|
Jan-09-15
 | | AylerKupp: <ajile> I guess I did miss the point. But you did say that the idea was "to test to test how much (if any) human guidance counts in a match of this type." If it is also fun that you are interested for the majority of us (or at least getting others involved) we could have a team (whoever is interested in participating) instead of just one person guiding the engine in whatever way each player prefers and vote on the moves to be played while the other player is restricted to playing the moves that the engine recommends as the top move. We can then make it both participatory (i.e. fun) and more scientific by having the same engine used by both sides; that way the number of variables can be minimized. So I would therefore suggest that Stockfish be the engine of choice since that is apparently the strongest free engine available. Or we could wait until the next Team Challenge (we probably won't have to wait too long) and, with both <chessgames.com>'s and <RandomVisitor>'s concurrence, have only <RandomVisitor> sign up for one side and everyone else sign up for the other side. That I think would achieve most of your goals, particularly if the team agrees to only use Rybka 4.1 for the analysis. |
|
| Jan-09-15 | | ajile: I like the idea of using RV and Rybka 4.1 since he is already here and has the hardware to do it easily. Also he knows how the games work and timetables. Plus his analysis has guided the Team for many games now and is considered the gold standard. Of course all this is speculation before the fact since he hasn't been asked yet if he would be able or would want to do it. |
|
| Jan-10-15 | | kb2ct: It would have been a good game even if he had found 23....Pa5 :0) |
|
| Jan-10-15 | | cormier: http://flightsquare.net/ |
|
| Jan-10-15 | | RandomVisitor: Interesting ideas. |
|
| Jan-10-15 | | kwid: <RandomVisitor>: <Interesting ideas.> No offence but I have Deep Rybka 4 only but have observed your lines output over many years now which leads me to belief that R4.1 has no chance against the latest SF versions at one day thinking time. Just imagine at least 100 SF beasts running at depth's of 40 ply and deeper guided by the anticipated moves from your Rybka 4.1 running on less than 10 computers. If we can take R4.1 out of its guiding opening book you have absolutely no chance unless you apply our sliding method on at least one comp to update the others with human intervention. Or you may have to retool and forget about R4.1 to make this an enjoyable contest. |
|
| Jan-10-15 | | Kinghunt: Suppose <RandomVisitor> weren't using the outdated Rybka 4.1, but rather was running Komodo 8. That would be a much tougher challenge. For that matter, it may be worth updating to Komodo 8 even for general World Team analysis. I am sure the team would be willing to cover the software cost as long as you still handle the electricity bills! (I'd be willing to cover it myself, even.) |
|
| Jan-10-15 | | thegoodanarchist: <DPLeo:
No need for excuses because it is simply a fact.
There are separate ratings for classical, rapid, and blitz not because there is a need for excuses but because they are simply different formats.> So if you lose in a different format, you still consider yourself undefeated? Not agreeing. A loss is a loss. If you lose a game you are not undefeated, sorry. |
|
| Jan-10-15 | | John Mulder: I'd love to see the World take on <RV> using Deep Rybka 4.1. That would be a really interesting chess game. |
|
| Jan-10-15 | | DPLeo: <thegoodanarchist: ... A loss is a loss. If you lose a game you are not undefeated ...> In terms of absolutes, yes, but I was not referring to absolutes. I was speaking in terms of different formats. If a player is undefeated in classical play and loses a blitz game, the loss doesn't affect the classical rating or record. Likewise, if our team is undefeated in games playing against 1 GM and loses a game against another team, the loss doesn't affect our undefeated record against single GM's. |
|
| Jan-10-15 | | DanLanglois: <So if you lose in a different format, you still consider yourself undefeated?
Not agreeing. A loss is a loss. If you lose a game you are not undefeated, sorry.> But, you are undefeated in a different format. A loose analogy: if the world champion loses a blitz game, then who is the world champion? |
|
| Jan-10-15 | | thegoodanarchist: <DanLanglois: <So if you lose in a different format, you still consider yourself undefeated? Not agreeing. A loss is a loss. If you lose a game you are not undefeated, sorry.>
But, you are undefeated in a different format. A loose analogy: if the world champion loses a blitz game, then who is the world champion?> False analogy. A World Champion is determined in a match of <multiple> games, not one game. You are comparing the outcome of a match with multiple games to the outcome of one game |
|
| Jan-10-15 | | thegoodanarchist: <DPLeo: <thegoodanarchist: ... A loss is a loss. If you lose a game you are not undefeated ...>
In terms of absolutes, yes, but I was not referring to absolutes. I was speaking in terms of different formats. If a player is undefeated in classical play and loses a blitz game, the loss doesn't affect the classical rating or record. Likewise, if our team is undefeated in games playing against 1 GM and loses a game against another team, the loss doesn't affect our undefeated record against single GM's.> You are sounding more and more like a bully:
<Hey everyone! We steamrolled one guy alone, when we had hundreds! But a team of 5 guys can beat our team of hundreds, yet we didn't loose!Why? Because we didn't have the opportunity to beat up on just one lone guy by himself!!!! <thegoodanarchist> is just a jerk! We're number one!!!> |
|
| Jan-10-15 | | ChemMac: I'm reminded about what my late wife scornfully called Mensa, which I had just joined, "A society for self-congratulation". Actually, it's such a nice idea that so many chess players of a large range of abilities, at least those that voted, can all say, " *I* was part of the team that just defeated the strongest player of Germany!. Yes, *I* voted for the winning moves! " I for one have learned a lot.
I told a friend that the World Team of which I was a part had just won against our Grandmaster opponent., "Ooh, that's so impressive!!" !! I had to explain that my part was quite minimal, but I'm sure that many others wouldn't bother to do that!
! : -)
Let us not forget that, for Naiditsch as for any other professional opponent, this was an exhibition game, played as is stated for our enjoyment. HE therefore succeeded in this very well, so let us, again, thank him and wish him both chess success - and marital happiness. |
|
| Jan-11-15 | | DanLanglois: Is there such a thing as a false analogy? There'd have to be such a thing as a true analogy.. |
|
| Jan-11-15 | | thegoodanarchist: <DanLanglois: Is there such a thing as a false analogy? There'd have to be such a thing as a true analogy..> You know what I mean. Should I have said you're comparing apples to oranges? |
|
Jan-11-15
| | AylerKupp: <<thegoodanarchist> is just a jerk! We're number one!!!> Actually, we're probably number 23. I've estimated the World Team's ranking before (incorrectly, I had one of V.Akobian's ratings wrong). Besides, FIDE changed its Rating Regulations effective Jul-01-14 so my earlier calculations are no longer valid. So this time I used FIDE's Initial Rating calculator, http://ratings.fide.com/calculator_..., treated each game as though it was part of the same tournament, and used the ratings of each of our opponents at the time that each game started. I also used the correspondence ratings rather than the OTB ratings of A.Nickel, G.Timmerman, and M.Umansky which were considerably higher than their OTB ratings. I'm not sure if mixing OTB and correspondence ratings is correct but that's what gives us the highest rating so that's what I'm going with. :-) So, after 10 games (N) our opponents' average rating (Rc) is 2610 and our score (W) is 8.0 (+6, -0, =4). That gives us an Initial Rating = 2730, putting us between both the live rating (http://www.2700chess.com/) and FIDE rating (http://ratings.fide.com/top.phtml?l...) of Liren Ding (2732), ranked #22 and David Navara (2729), ranked #23. So we would replace Navara as the #23 ranked "player" in the world. Congratulations, World Team! |
|
| Jan-11-15 | | DanLanglois: <thegoodanarchist: You know what I mean. Should I have said you're comparing apples to oranges?> That's what I was doing. |
|
| Jan-11-15 | | DanLanglois: The statement that something is like comparing apples and oranges is a kind of analogy itself. |
|
| Jan-12-15 | | ajile: So have the naysayers ruined the match idea? RV doesn't appear overly excited. |
|
| Jan-12-15 | | kwid: <ajile: So have the naysayers ruined the match idea? RV doesn't appear overly excited> Is SF's rating not 200 ELO higher than Rbka? If so what chances will RV have against only some of our members with fast CBU's running SF and or KM on auto pilot? |
|
Jan-12-15
 | | kutztown46: <ajile> RV is a man of few words. I sensed interest from him. Why don't you post your idea at the cg support forum? If they go along with it, I think there are enough interested people to make it a fun experiment while we are waiting for the next game. |
|
Jan-12-15
 | | moronovich: I am mainly a lurker and a bypasser,but I think that <ajiles> idea is excellent. |
|
| Jan-12-15 | | team kids can win: IMO, the kind of GM who specializes as a second/researcher for a WC match would be an interesting opponent, for multiple reasons: 1) They are already proficient and accomplished in using computers to search for new innovations, especially in the opening where the World Team is weakest. 2) They might work a little harder than other OTB type GMs because the match style is closer to their specialty and they could enhance their reputation for something they do professionally. |
|
 |
|
|
< Earlier Kibitzing · PAGE 699 OF 707 ·
Later Kibitzing> |
Kibitzing
