v0.4.0 (#49)

ELO   | 150.10 +- 12.36 (95%)
CONF  | 60.0+0.60s Threads=1 Hash=128MB
GAMES | N: 2000 W: 1019 L: 205 D: 776
https://chess.swehosting.se/test/2280/

Bench: 13055552

Cargo.toml

@@ -1,9 +1,11 @@
 [package]
 name = "akimbo"
-version = "0.3.0"
+version = "0.4.0"
 edition = "2021"
 authors = ["Jamie Whiting"]
 
 [profile.release]
 panic = 'abort'
 strip = true
+codegen-units = 1
+lto = true

README.md

@@ -2,20 +2,31 @@
 
 # akimbo
 
+![License](https://img.shields.io/github/license/JacquesRW/akimbo?style=for-the-badge)
 [![GitHub release (latest by date)](https://img.shields.io/github/v/release/JacquesRW/akimbo?style=for-the-badge)](https://github.com/JacquesRW/akimbo/releases/latest)
 
 </div>
 
 A compact UCI Chess engine written in Rust.
 
-The main aim is to stay under 1000 SLOC (excluding comments and blank lines) and 1500 TLOC (including the aforementioned).
+Huge thanks to all of the members of [this OpenBench Instance](https://chess.swehosting.se/users/) who have provided support and guidance in the development
+of akimbo, as well as facilitating far faster testing than on my own.
+
+## Aims
+
+#### Small
+akimbo has a hard upper limit of 1000 lines of code, excluding blank lines and comments. Stats for each version are included below.
+
+#### PST Only
+akimbo's evaluation consists only of a set of tapered piece-square tables, tuned using Texel's tuning method.
 
 ## Stats
-|                           Version                                |     Release Date     | SLOC | TLOC | [CCRL 40/15](https://www.computerchess.org.uk/ccrl/4040/) |
-| :---------------------------------------------------------------:|:--------------------:|:----:|:----:|:------:|
-| [0.1.1](https://github.com/JacquesRW/akimbo/releases/tag/v0.1.1) |    8th January 2023  | 1167 | 1381 |  2469  |
-| [0.2.0](https://github.com/JacquesRW/akimbo/releases/tag/v0.2.0) |   10th   April 2023  |  866 | 1000 |  2525  |
-| [0.3.0](https://github.com/JacquesRW/akimbo/releases/tag/v0.3.0) |   16th     May 2023  |  891 | 1050 |  2587  |
+|                           Version                                |     Release Date     | SLOC | [CCRL 40/15](https://www.computerchess.org.uk/ccrl/4040/) |
+| :---------------------------------------------------------------:|:--------------------:|:----:|:------:|
+| [0.1.1](https://github.com/JacquesRW/akimbo/releases/tag/v0.1.1) |    8th January 2023  | 1167 |  2469  |
+| [0.2.0](https://github.com/JacquesRW/akimbo/releases/tag/v0.2.0) |   10th   April 2023  |  866 |  2525  |
+| [0.3.0](https://github.com/JacquesRW/akimbo/releases/tag/v0.3.0) |   16th     May 2023  |  891 |  2587  |
+| [0.4.0](https://github.com/JacquesRW/akimbo/releases/tag/v0.4.0) |    4th    July 2023  |  852 |   TBD  |
 
 ## Features
 
@@ -28,12 +39,13 @@ The main aim is to stay under 1000 SLOC (excluding comments and blank lines) and
 - Late Move Reductions
 - Reverse Futility Pruning (with Improving)
 - Null Move Pruning
-- Razoring
 - Internal Iterative Reductions
+- Late Move Pruning
+- Razoring
 
 #### Move Ordering
 1. Hash Move
 2. Captures (MVV-LVA)
 3. Promotions
 4. Killer Moves
-5. History Heuristic
+5. History Heuristic (with Gravity)

src/main.rs

@@ -33,7 +33,7 @@ fn main() {
         for fen in bench_fens {
             pos = Position::from_fen(fen);
             let timer = Instant::now();
-            go(&pos, &mut eng, false, 8);
+            go(&pos, &mut eng, false, 11);
             total_time += timer.elapsed().as_millis();
             total_nodes += eng.nodes + QNODES.load(std::sync::atomic::Ordering::Relaxed);
         }