Komodo 5 running for the IPON

[IWB]

Here you find more info:

http://www.inwoba.de

Have fun
Ingo

[Bram Visser]

Nice! I guess this test will get you some extra site-visitors !

[IWB]

It already happens - and would be nice if I would earn money with this

Bye
Ingo

[gleperlier]

It already happens - and would be nice if I would earn money with this

Bye
Ingo

[Rebel]

Code: Select all

Komodo 5

Komodo 5 - Houdini 2.0 STD (3028)  12.5 - 12.5  50.00%  Perf=3028 
Komodo 5 - Critter 1.6a (2973)  15.0 - 10.0  60.00%  Perf=3043 
Komodo 5 - Stockfish 2.2.2 JA (2966)  12.5 - 11.5  52.08%  Perf=2980 
Komodo 5 - Deep Rybka 4.1 (2956)  14.0 - 10.0  58.33%  Perf=3014 
Komodo 5 - Naum 4.2 (2834)  19.0 - 5.0  79.17%  Perf=3065 
Komodo 5 - Deep Shredder 12 (2800)  21.0 - 4.0  84.00%  Perf=3088 
Komodo 5 - Gull 1.2 (2796)  19.0 - 5.0  79.17%  Perf=3027 
Komodo 5 - Hannibal 1.2 (2792)  19.0 - 6.0  76.00%  Perf=2992 
Komodo 5 - Deep Sjeng c't 2010 32b (2791)  20.0 - 4.0  83.33%  Perf=3070 
Komodo 5 - Spike 1.4 32b (2782)  17.0 - 6.0  73.91%  Perf=2962 
Komodo 5 - spark-1.0 (2764)  19.0 - 5.0  79.17%  Perf=2995 
Komodo 5 - Protector 1.4.0 (2755)  21.0 - 3.0  87.50%  Perf=3093 
Komodo 5 - Deep Junior 13.3 (2744)  22.0 - 3.0  88.00%  Perf=3090 
Komodo 5 - HIARCS 13.2 MP 32b (2743)  19.0 - 5.0  79.17%  Perf=2974 
Komodo 5 - Quazar 0.4 (2732)  20.0 - 3.0  86.96%  Perf=3061 
Komodo 5 - Zappa Mexico II (2699)  19.5 - 4.5  81.25%  Perf=2953 
Komodo 5 - MinkoChess 1.3 (2691)  18.0 - 6.0  75.00%  Perf=2881 

307.5 - 103.5 74.82% Perf=3004

And climbing

[Laskos]

Here you find more info:

http://www.inwoba.de

Have fun
Ingo

Hi Ingo,

Do you plan to use Ordo for rating, or adjusted Bayeselo as shown by Adam Hair? The default Bayeselo heavily compresses your ratings, as shown again by Adam Hair (therefore all these complains that the average of ratings is not close to the Bayeselo rating were justified).
Adam Hair's posts are in this thread: http://talkchess.com/forum/viewtopic.ph ... 80&start=0

Kai

[IWB]

Hi Ingo,

Do you plan to use Ordo for rating, or adjusted Bayeselo as shown by Adam Hair? The default Bayeselo heavily compresses your ratings, as shown again by Adam Hair (therefore all these complains that the average of ratings is not close to the Bayeselo rating were justified).
Adam Hair's posts are in this thread: http://talkchess.com/forum/viewtopic.ph ... 80&start=0

Kai

Thx for the link, I have to study that with more time. I have to admit I do not understand half of it, but I will compare the ratings as I do it now (mm 0 1) and with mm 1 1 and scale set to 1 after mm.
Nonetheless I do see a comparision of my result file made with bayeselo default and computed but I do not run bayeselo default - I wun mm 0 1 (see conditions).

I will check in a few days, but as I publish the results, an elostat rating and bayes with mm 01 everyone can make his own list ...

Bye
Ingo

[Vinvin]

Code: Select all

Komodo 5

Komodo 5 - Houdini 2.0 STD (3028)  12.5 - 12.5  50.00%  Perf=3028 
Komodo 5 - Critter 1.6a (2973)  15.0 - 10.0  60.00%  Perf=3043 
Komodo 5 - Stockfish 2.2.2 JA (2966)  12.5 - 11.5  52.08%  Perf=2980 
Komodo 5 - Deep Rybka 4.1 (2956)  14.0 - 10.0  58.33%  Perf=3014 
Komodo 5 - Naum 4.2 (2834)  19.0 - 5.0  79.17%  Perf=3065 
Komodo 5 - Deep Shredder 12 (2800)  21.0 - 4.0  84.00%  Perf=3088 
Komodo 5 - Gull 1.2 (2796)  19.0 - 5.0  79.17%  Perf=3027 
Komodo 5 - Hannibal 1.2 (2792)  19.0 - 6.0  76.00%  Perf=2992 
Komodo 5 - Deep Sjeng c't 2010 32b (2791)  20.0 - 4.0  83.33%  Perf=3070 
Komodo 5 - Spike 1.4 32b (2782)  17.0 - 6.0  73.91%  Perf=2962 
Komodo 5 - spark-1.0 (2764)  19.0 - 5.0  79.17%  Perf=2995 
Komodo 5 - Protector 1.4.0 (2755)  21.0 - 3.0  87.50%  Perf=3093 
Komodo 5 - Deep Junior 13.3 (2744)  22.0 - 3.0  88.00%  Perf=3090 
Komodo 5 - HIARCS 13.2 MP 32b (2743)  19.0 - 5.0  79.17%  Perf=2974 
Komodo 5 - Quazar 0.4 (2732)  20.0 - 3.0  86.96%  Perf=3061 
Komodo 5 - Zappa Mexico II (2699)  19.5 - 4.5  81.25%  Perf=2953 
Komodo 5 - MinkoChess 1.3 (2691)  18.0 - 6.0  75.00%  Perf=2881 

307.5 - 103.5 74.82% Perf=3004

And climbing

now :

Code: Select all

Komodo 5 - Critter 1.6a (2973)	19.5 -11.5	 62.90% Perf=3064
Komodo 5 - Stockfish 2.2.2 JA (2966)	14.5 -15.5	 48.33% Perf=2955

Awesome against Critter and deceiving against SF, we will see if the pattern still ...

[Laskos]

Hi Ingo,

Do you plan to use Ordo for rating, or adjusted Bayeselo as shown by Adam Hair? The default Bayeselo heavily compresses your ratings, as shown again by Adam Hair (therefore all these complains that the average of ratings is not close to the Bayeselo rating were justified).
Adam Hair's posts are in this thread: http://talkchess.com/forum/viewtopic.ph ... 80&start=0

Kai

Thx for the link, I have to study that with more time. I have to admit I do not understand half of it, but I will compare the ratings as I do it now (mm 0 1) and with mm 1 1 and scale set to 1 after mm.
Nonetheless I do see a comparision of my result file made with bayeselo default and computed but I do not run bayeselo default - I wun mm 0 1 (see conditions).

I will check in a few days, but as I publish the results, an elostat rating and bayes with mm 01 everyone can make his own list ...

Bye
Ingo

It seems that for IPON as well as for CCRL setting mm 1 1 and then scale to 1 gives the most accurate results. Or you could use Ordo. What you see in Adam posts as 340 instead of 400 (all these fitted numbers must be compared with the default 400 of the logistic) is giving the magnitude of the compression, in this case some 15%. Not negligible at all. Hope you use Bayeselo mm 1 1 scale 1 prior 0.1 or Ordo.

Kai

[lkaufman]

We would appreciate any comments you might wish to make here regarding Komodo's time utilization in your test. Do we generally end up with more time, less time, or about the same as our top opponents? Are there any time forfeits? The reason I ask is that we never test with Ponder on, so your feedback is about all we have to go by regarding ponder-on testing.