Fire 1.3 Burns Rybka 4

[kingliveson]

Fire Burns Rybka Deep in Bullet°

Code: Select all

Fire13 vs Rybka4 RND 01 1+0                             
1   Fire 1.3 x64            +17/=28/-5 62.00%   31.0/50
2   Deep Rybka 4 SSE42 x64  +5/=28/-17 38.00%   19.0/50

• † 25 positions
  † No Tablebases
  † 128 MB Hash
  † Ponder Off
  † Phenom 940
  † 4 Cores
  † 1+0 Time Control

Games available for download upon request.
° More games needed.

[yanquis1972]

i don't think more games are needed so much as a different TC...at least 1/1 if you're intent on compiling a lot of games fairly quickly. 1/0 is just insanity.

[kingliveson]

i don't think more games are needed so much as a different TC...at least 1/1 if you're intent on compiling a lot of games fairly quickly. 1/0 is just insanity.

I'll test different TC. This was just getting the feet wet so to speak.

Code: Select all

Rybka 4 vs IvanHoe 9.63b 15+10 
                            
1   Deep Rybka 4 SSE42 x64  +9/=37/-4 55.00%   27.5/50
2   IvanHoe 9.63b x64       +4/=37/-9 45.00%   22.5/50

• † 25 positions
  † 3-4-5 Nalimov/Robbobases
  † 512 MB Hash
  † Ponder Off
  † Phenom 940
  † 4 Cores
  † 15+10 Time Control

Games available for download upon request.

[BubbaTough]

i don't think more games are needed so much as a different TC...at least 1/1 if you're intent on compiling a lot of games fairly quickly. 1/0 is just insanity.

I'll test different TC. This was just getting the feet wet so to speak.

Code: Select all

Rybka 4 vs IvanHoe 9.63b 15+10 
                            
1   Deep Rybka 4 SSE42 x64  +9/=37/-4 55.00%   27.5/50
2   IvanHoe 9.63b x64       +4/=37/-9 45.00%   22.5/50

• † 25 positions
  † 3-4-5 Nalimov/Robbobases
  † 512 MB Hash
  † Ponder Off
  † Phenom 940
  † 4 Cores
  † 15+10 Time Control

Games available for download upon request.

I would also recommend using positions that the developers were unlikely to have used for training purposes (i.e. don't use standard positions such as Nunn or Silver or such).

-Sam

[kingliveson]

I would also recommend using positions that the developers were unlikely to have used for training purposes (i.e. don't use standard positions such as Nunn or Silver or such).

-Sam

My mistake, should have been posted on tournament forum. But yes, I use a test suite designed by me specifically for such purpose. Fire 1.3 at least in bullet matches is showing really good signs. Norm seems to have done something right. Longer time control will definitely tell us more.

[BubbaTough]

I would also recommend using positions that the developers were unlikely to have used for training purposes (i.e. don't use standard positions such as Nunn or Silver or such).

-Sam

My mistake, should have been posted on tournament forum. But yes, I use a test suite designed by me specifically for such purpose. Fire 1.3 at least in bullet matches is showing really good signs. Norm seems to have done something right. Longer time control will definitely tell us more.

Yes, it is very impressive. In my testing, 1 0 is considered a long game .

-Sam

[Chan Rasjid]

Just to make sure.

Does Time Control 1 + 0 mean game in 1 min. If yes, then such games are completely meaningless as one side will surely have time failure.

[Albert Silver]

i don't think more games are needed so much as a different TC...at least 1/1 if you're intent on compiling a lot of games fairly quickly. 1/0 is just insanity.

I'll test different TC. This was just getting the feet wet so to speak.

Code: Select all

Rybka 4 vs IvanHoe 9.63b 15+10 
                            
1   Deep Rybka 4 SSE42 x64  +9/=37/-4 55.00%   27.5/50
2   IvanHoe 9.63b x64       +4/=37/-9 45.00%   22.5/50

• † 25 positions
  † 3-4-5 Nalimov/Robbobases
  † 512 MB Hash
  † Ponder Off
  † Phenom 940
  † 4 Cores
  † 15+10 Time Control

Games available for download upon request.

I would also recommend using positions that the developers were unlikely to have used for training purposes (i.e. don't use standard positions such as Nunn or Silver or such).

-Sam

I don't actually think Vas uses my suite, since last I heard some of his tests ran in the thousands of games, far beyond the limits of my suite. Plus the test was designed to cover as much of the full gamut of typical positions as possible, meaning no special favoritism.

The point being, if you prefer, that if you can *book* my suite via playing algorithms (not book moves), then you will have booked chess per se.

[AdminX]

My 50 Games 5 Min Blitz results were a win for Rybka 4.

Ponder On
1024 Hash
Tablebase 3-4-5-6
Robobitbases + Robototalbases
2 Cores Each
Large Pages Enabled

Code: Select all

Intel Core 2 Quad Q9000, Blitz 5m  0

                            
1   Deep Rybka 4 x64   +42  +11/=34/-5 56.00%   28.0/50
2   Fire 1.3 x64       -42  +5/=34/-11 44.00%   22.0/50

[meschle]

Good news for rybka - thanks ted

just in progress 5' match, 100 games in aquarium.

Out of interest which gui did you carry out your match?