Stockfish 2.2 etc tests

1+1 Blitz
No books No Bases No Large pages

[You must be registered and logged in to see this link.]


not sure another 50 games would see it do any better and quite honestly have better things to do with lap top and our time to bother with any more, hopefully the next release might be an improvement


looking at some of the games its clear that the time management is highly suspect even with increment , but worse is that i have seen it make incredible blunders within the first 20 moves in 2 games , it needs :gal4:

Intel Xeon CPU X5675 6X3.07GHz :
Time 2'+2" . Gui Fritz 13 . Hash 128 . Nalimov TB's . Ponder off.
Book: HS-7moves plus.ctg :

Houdini 2.0c Pro x64 - Stockfish 2.2 JA 64bit :
34/60 - 26/60 . 56.67%-43.33% . ( +22 - 14 = 24 )

Sandi says...
Im not excited about this new Stockfish.
The time management is not good.
Velmarin or Peterpan could compile this source better.


###

Well i agree the compile sucks but as far as that maggot velmarin's concerned



JA compile >> << velmarin



final thought

of all the overhyped release's , critter1.4 . komodo 4 etc , only this stockfish "IF"correctly compiled by a competent compiler such as peter pants stand any chance of slapping houdarts plagerized compiles into 2nd place

critter1.4 and komodo 4 FAIL to catch houdini 1.5a despite all the hyped up fantasy testings
less eggnog and more honest testings please

Critter 1.4 64-bit SSE4 -Stockfish 2.2 JA 64bit :
54.5/100 - 45.5/100 . 54.5%-45.5% . ( + 30 - 21 = 49 )

wow critter

Stockfish 2.2 vs 2.1 (2), Blitz:1'+1" 0


1 Stockfish 2.2 JA 64bit +37/=136/-34 50.72% 105.0/207
2 Stockfish 2.1.1 JA 64bit +34/=136/-37 49.28% 102.0/207

Also, 3+0:

1 Stockfish 2.1.1 JA 64bit 89.5
2 Stockfish 2.2 JA 64bit 60.5

Posted: Mon Jan 02, 2012 5:14 pm
Hi,

I stopped testing this version because of an obvious timing bug.
Even if the IPON is acceptin losses on time as an "engine feature" which belongs to the playing strenth in a tourney, it makes no sence to leave an engine in the list whose rating would be worse than the predecessor because of a very high loss on time rate ... !

Bye
Ingo

[You must be registered and logged in to see this link.]


~~~~~~~~~~~~~


Even this fool finally see the futility of continuing to test this buggy compile

Intel Xeon CPU X5675 6X3.07 GHz


Stockfish VE_06 64bit +13 =25 -2 25.5/40 - 14.5/40
Stockfish 2.2 JA 64bit +2 =25 -13 14.5/40 - 25.5/40


needed

Intel Xeon CPU X5675 6X3.07 GHz :
Time 3'+0" . GUI Fritz 13 . Ponder off . Hash 128 . Nalimov TB's . Book : HS-7moves plus.ctg :
Critter 1.4 64-bit SSE4 - Stockfish 120116 PP 64bit SSE :
53.5/100 - 46.5/100 . 53.5%-46.5% . ( +27 - 20 = 53 )

This Peter pan compile's not even the same strength as the official version of Stockfish. and that's nothing to write home about

Good job he only made it for " his friends " i can not offend him by saying " thanks a bunch , its pants "