Not logged inCSS-Forum
Forum CSS-Online Help Search Login
CSS-Shop Impressum Datenschutz
Up Topic Hauptforen / CSS-Forum / Rybka 4 against Houdini 1.01 (4-cpu)
- - By Orlando de la Vega Date 2010-06-08 23:23
Houdini 1.01 against Rybka 4 (4 cpu)

PlaceEngineAuthorCountryPointsSingle results
1Houdini 1.01Robert Houdart30.0/54½111½½½100-½0½1001011-1½½00½½½½0-½11½1½½1½½-½1½½11010½-0½½1
2Rybka 4V. Rajlich & L. Kaufman24.0/54½000½½½011-½1½0110100-0½½11½½½½1-½00½0½½0½½-½0½½00101½-1½½0




50 of 50 matches played.
Houdini 1.01 v/s Rybka 4 (54: + 18,= 24,- 12)

Start of tournament: ......... 2010.06.07, 18:24:15
Town/ Country: .............. Dax, France
Level: ............................ Blitz 40/1' + 1" repeated
Hardware: ...................... AMD Phenom(TM) 9750 Quad-Core
Engines: ......................... Houdini 1.01 / Rybka 4 both 32-bits using 4 threads
Hash: ............................. 256 MB / 256 MB
GUI: ............................... Shredder Classic 3
OS: ................................ Windows Seven Ultimate
Ponder OFF
Nalimov TB ..................... 3+4+5
Book: ............................. Shredder book, stopped using after 5 moves.
Openings: ....................... 27 different openings randomly chosen by GUI, large mode.

Sorry, I had to stop the tournament earlier and publish it quickly because I had to go away !

Best regards,
Orlando
Parent - - By Axel Caro Date 2010-06-09 07:07
Hi Orlando,

may you repeat this match with LTC (e.g. 30'+10")? I have some examples which show increasing scoring of R4 with longer time controls (I will post them soon).

Best regards
Axel
Parent - - By Orlando de la Vega Date 2010-06-09 08:41
Hi Axel,

OK, I'll try it, because I see it interessant too. But it will take longer time

Best,
Orlando
Parent - By Axel Caro Date 2010-06-09 22:08
Hi Orlando,
Zitat:
OK, I'll try it, because I see it interessant too.

fine!

Zitat:
But it will take longer time

As a german proverb says: "Good things need their time!"

Here my latest results with different time controls and variable ponder-settings:

Nunn2, ponder on, 10'+5"
Rybka 4 x64 - FireBird 1.2 x64 SP                    19.5  -20.5 

Nunn2, ponder off, 10'+5"
Rybka 4 x64 SP - FireBird 1.2 x64 SP               22.5  -17.5     

Nunn2, ponder off, 30'+30"
Deep Rybka 4 x64 - FireBird 1.2 x64 MP           27.0  -23.0 (50 games, I missed to limit match to 40 games) 

With ponder on FireBird has massive advantage by the well known "famous" ponderhits.

Regards,
Axel

PS: Not enough games to extract valid conclusions, but maybe a hint to a tendency (increasing strength at longer time controls).
Parent - - By Bram Mourik Date 2010-06-09 10:19
[quote="Axel Caro"]
Hi Orlando,

may you repeat this match with LTC (e.g. 30'+10")? I have some examples which show increasing scoring of R4 with longer time controls (I will post them soon).

Best regards
Axel
[/quote]

Vielleicht bezeichnet dieses auch was

Blitz games
T8100 (6,29 Fritzmark) Blitz 4m+2s  Nunn2
                           
1   Houdini 1.01 w32 2_CPU  +17/=27/-6 61.00%   30.5/50
2   Deep Rybka 4 w32        +6/=27/-17 39.00%   19.5/50

LTC games
T8100 (6,29 Fritzmark) LTC - 20m/40+10m/20+(10m+12s) Nunn2
                        
1   Houdini 1.01 w32 2_CPU  +11/=33/-6 55.00%   27.5/50
2   Deep Rybka 4 w32        +6/=33/-11 45.00%   22.5/50

grüsse Bram
Parent - By Kowolik Marcel Date 2010-06-09 16:46
Deep Rybka4(64bit)-Houdini1.01(64bit)   17-13   +8,=18,-4     
T7300,2Threads,Vista64bit,4GB Ram,Deep Fritz11 ctg bis Zug 8,Deep Fritz11GUI,ca 50GB Tablebase.

Bedenkzeit 30Min pro Partie+15S pro Zug.

MfG Marcel
Parent - - By Ernest Bonnem Date 2010-06-10 02:51
You are back to your quick-quick-test-post bad habits...

54 games (why not 52 or 56?) at 40/1' + 1" repeated... you should be ashamed!  
Parent - - By Damir Desevac Date 2010-06-10 22:42
When you are such an expert, why not make those tests yourself?
Bye
Damir
Parent - By Ernest Bonnem Date 2010-06-11 01:07
I have (400+ games), and don't find at all that kind of result!...  
Up Topic Hauptforen / CSS-Forum / Rybka 4 against Houdini 1.01 (4-cpu)

Powered by mwForum 2.29.3 © 1999-2014 Markus Wichitill