Grand Blitz Tournament - Predict the Top 6 Engines!

[swami]

The King has yet to play the 7 of the engines from the Top 15...

and it has so far shown to be the good destroyer of the weaker engines.

Frenzee is also extremely good against weaker engines, rarely gives away a point.

Glaurung even with 2 cores, is still at 11th placings so far. Maybe Glaurung needs better 'contempt factor'

[swami]

I think I'm kind of interested in organizing the similar new Grand Blitz tournament for engines rated <2200 or <2300

Such that the division 7 engines and below qualify, and 101 will be selected.

Is that a good idea or should I continue doing usual Edition 10 division tournaments?

I'm more inclined to opt for the former rather than latter.

83/84 rounds finished btw.

[Tony Thomas]

I think I'm kind of interested in organizing the similar new Grand Blitz tournament for engines rated <2200 or <2300

Such that the division 7 engines and below qualify, and 101 will be selected.

Is that a good idea or should I continue doing usual Edition 10 division tournaments?

I'm more inclined to opt for the former rather than latter.

83/84 rounds finished btw.

I think that the programmers of weaker engines would love to see their programs compete in a tournament. It is bound to increase their interest, which will in turn give is stronger (at least bug fixed) versions of those engines. Then again I am a whore for weak engines.

[swami]

Tony, For some reason, I believe I see more of programmers who have <2300 engine actively particpating in the forum talks
than those who have >2300 who usually don't participate ongoing discussion in here, they either remain silent and improve their engines or they have long since given up.

[pedrox]

I do not agree, Swami.

I believe that almost 50% of programmers of your great tournament are participating in the forums. I think that if you pick 100 engines below the 2300, then it will be harder to overcome the percentage of programmers, as many programs below this ELO are not being updated.

Furthermore I believe that all active programmers are willing to see their engines play, does not matter if the engine is 1500 or 2700.

I also enjoy all divisions, regardless of the strength of the engines.

Pedro

[swami]

Hi Pedro,

You are partly right, I have installed a lot of <2300 engines, 1/4th of them seem to have some kind of major bugs. As evidenced in Tony's no play list.

Some of them lose on time, some of them don't play the predifined opening positions/only play with the book, some lose on time when given check, a lot of them unupdated older ones etc These type of bugs are rare in engines >2300.

You are also right in your estimation that there are still 50% of authors active in comp chess. But I'm getting the impression that there are more authors who have <2300 engine than those who have >2300, That's just me.

Anyway, I'd be interested in making a poll about it. I just can't decide what to do when it comes to comp chess testing and tournament creating, I feel any kind of tournament set up is just too good

Regards,
Swami

[swami]

- Results after 86/87 Rounds.

- Results from 4324 games.



http://computerchessblogger.googlepages ... Rounds.jpg

[Tony Thomas]

Argh..I forgot to consider the fact that The King is using 2 CPU's...I am such a dolt..

[swami]

Ofcourse - The King Archangel had used 1 CPU in Division 2 Edition 9 last month and it won the division easily.

The King here uses 2 CPU, and it is beating a hell out of lotsa weaker engines...but the fact remains that it still has to play stronger engines...so I'd guess Thinker has a chance to climb up the ladder once again.

Also, Notice the Huge_Leap in the placings of Cyrano!! from mid 30 something to 21!

[Graham Banks]

Argh..I forgot to consider the fact that The King is using 2 CPU's...I am such a dolt..

Yeah I also forgot that some engines were using 2CPUs and others only 1CPU.
Oh well.