Monday, January 18, 2016

iCE wins division 2 in Grahams Amateur Tournament

Five years ago early iCE started in the lowest divison 7 of this touranment. Now it is finally entering the top league. It was a long journey.

Thursday, January 7, 2016

iCE 3 ELO gain

iCE ELO gain (absolute values relate to the private ELO list of Lars Hallerström)

The first testing results are now available. I rely envy those guys for their hardware. Looks like iCE 3 gained something in the area of about 140 ELO. This number matches the early results from the CEGT tests so is pretty reliable, at least for TC of up to 5 minutes per game.

Below is the attached cross table. Thank you, Lars, for running the test!

Tournament M5 Cross Table at Round 40
PosNAME Rtg Fed Pts Results Pct
1Ice 3.0 (64)2980 GER607.5
+451 =313 -236
2Rybka 4.1 SSE42 (64-4cores)3169 SWE30.0
+24 =12 -4
3Critter 1.6a (64-4cores)3199 SVK29.0
+20 =18 -2
4Hannibal 1.5 (64-4cores)3086 ITA24.5
+15 =19 -6
5Texel 1.05 (64-4cores)3066 SWE24.0
+17 =14 -9
6Chiron 2 (64-4cores)3109 ITA23.0
+18 =10 -12
7Spike 1.4 (4cores)21.5
+12 =19 -9
8Senpai 1.0 sse42 (64-4cores)3052 FRA20.5
+13 =15 -12
9Gaviota 1.0 (64-4cores)2951 ARG19.5
+9 =21 -10
10Ex 7.88b (64-4cores)2866 USA18.5
+9 =19 -12
11Arasan 18.1 (64-4cores)2914 USA17.5
+8 =19 -13
12SmarThink 1.80 (64)2988 RUS17.0
+10 =14 -16
13BobCat 7.1 (64-4cores)2800 NED16.5
+9 =15 -16
Spark 1.0 (64-4cores)2981 NED16.5
+10 =13 -17
14Cheng 4.37 rc0 (64-4cores)2887 CZE15.0
+10 =10 -20
Crafty 25.0 JA (64-4cores)2892 USA15.0
+8 =14 -18
15Scorpio 2.7.6 JA (64-4cores)2887 ETH14.0
+8 =12 -20
16Sjeng WC2008 (64-4cores)2912 BEL13.5
+8 =11 -21
17Tornado 7 (64-4cores)2944 GER10.5
+6 =9 -25
18Deuterium 14.01 (64)2946 PHI8.0
+5 =6 -29
19Pedone 1.3 (64-4cores)2699 ITA7.5
+5 =5 -30
20Rodent 1.7 (64)2883 POL7.0
+3 =8 -29
21Octo 5190 (64-4cores)2865 GER6.5
+3 =7 -30
Ktulu 96.5
+2 =9 -29
22Fruit 2.3.12937 ROU6.0
+2 =8 -30
23Bug2 1.9 (64-4cores)2829 BRA5.0
+2 =6 -32
Tournament start05 Jan 2016 00:00
Latest update10 Jan 2016 17:45
Site/ CountrySkrutten644

Sunday, January 3, 2016

iCE 3 gets released

I'm finally ready to release version 3 of my little chess engine project. It is still only single core as I haven't found the motivation to implement SMP. I'm still on the way to make the engine smarter not faster. So SMP is maybe something for the next release then.

What did change in v3

UCI Options Cleanup The UCI options of iCE are restructured and renamed. Details can be found here: iCE 3 UCI Options

Multi PV support iCE is now supporting Multi PV mode.

Bugfix The ponder protocol implementation is fixed. In Shredder GUI it did not work.

Evaluation: A bit of weight tuning was perfromed. However most changes failed or had only minor impact.

Search Changes: iCE 2 added History Heuristics, Late Move Pruning, Razoring and Counter Move Heuristics. For iCE 3 I optimized the conditions a bit further to make smarter pruning decisions.

Books: I finally touched my book code again. It was necessary as the books contained some errors coming from changes to the internal move format. I also improved the book building framework to use perfromance data of iCE in certain lines.

Time Management: I changed the algorithm for time allocation to not use all available time on the last move before TC.

What do you get

I make 3 versions of iCE available
  • ice3-x64-modern - This is the release you probably want to use. It makes use of some HW instructions found on recent CPU.
  • ice3-x64 - This is the fall back release if your HW doesn't support the instructions used in the modern compile
  • ice3-x32 - As the name suggests, if you still on a 32 bit platform like Windows XP use this one
The versions along with new books can be found at my download site

How to use

iCE is only a chess engine so it integrates into a graphical interface and communicates with it via a protocol. All interfaces that support the UCI (Universal Chess Interface) are usable. Among the available interfaces are
The above interface are tested in at least some version, but others should work as well. Pick your favorite choice.

iCE 3 playing in the Shredder GUI

Saturday, January 2, 2016

Short Algebraic Notation Madness

Chess games collections are usually stored in PGN and the moves are given in SAN (Short Algebraic Notation). This is meant to make it more readable for a human reader but for a human writer it is sometimes hard to follow the complicated rules especially if he writes the score sheet while playing a game.

For me as a programmer the existence of SAN is also really annoying. LAN (long algebraic notation) would make life much easier. The problem with SAN is the handling of move ambiguity. As usually the source square of a move is not given it must be calculated using the current layout of the board (which piece of the given type can travel to the target square).

So in order to translate SAN into a binary move object with (source and target square) a corresponding board class is needed which implements operations like get me all squares a bishop can travel to from source sq x. What makes it even more difficult is the rules for handling cases when more than one piece can travel to a certain square. Then file or rank of the source square are given, sometimes even both.

But not enough. If two pieces can reach the target square from a source square then the source file or rank indicator rule is not applied if one of the pieces is pinned and cannot move.

A lot of code is needed to handle all the special SAN cases and I estimate that a a fair number of club level written score sheets are not correct either.

So why am I complaining. My just published new books contain some errors as I was not handling the "two pieces can reach the target square but one of them is pinned to the king" case correctly. I fixed my book code and I'm in the process of rebuilding the books now.

So an update is coming soon.