New engine releases 2020

Discussion of anything and everything relating to chess playing software and machines.

Moderators: bob, hgm, Harvey Williamson

Forum rules
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
Guenther
Posts: 3342
Joined: Wed Oct 01, 2008 4:33 am
Location: Regensburg, Germany
Full name: Guenther Simon
Contact:

Re: New engine releases 2020

Post by Guenther » Mon Sep 07, 2020 11:32 am

Gabor Szots wrote:
Sun Sep 06, 2020 9:32 am
Guenther wrote:
Sun Sep 06, 2020 9:27 am
The author created a fix for the nps overflow after my report, I will update my download soon.
Surely he likes to be listed at CCRL, or are you currently testing as private person?
(The nps overflow fix will have no influence on strength.)
I have already submitted my games to CCRL, Neocortex is likely to appear on our blitz list tomorrow, maybe even today.
I have updated my binaries of neocortex now with the nps output fix (no change for playing strength).
Justin told me that he will release binaries himself in the future (not so far away), beginning with version 1.1.
https://rwbc-chess.de
Greg Strong@ovyron wrote: What I think is that you have a very, very, VERY strong desire to post, and very little self-control.
All the time. In every conversation. You must post. Something. It's like you just can't help yourself.

Guenther
Posts: 3342
Joined: Wed Oct 01, 2008 4:33 am
Location: Regensburg, Germany
Full name: Guenther Simon
Contact:

Re: New engine releases 2020

Post by Guenther » Tue Sep 08, 2020 1:01 pm

New releases not mentioned yet:

MrBob 0.80
https://github.com/bobsquared/Mr_Bob_Chess

Scorpio 3.09
https://github.com/dshawul/Scorpio/releases

LC0 0.26.2 (identical to 0.26.2 rc1)
https://github.com/LeelaChessZero/lc0/releases

Rodent IV 0.29 [src only - I will compile it today, except someone is faster ;-) ]
https://github.com/nescitus/rodent-iv/commits/master

Bad news for older bitbucket repos, read here:
https://bitbucket.org/blog/sunsetting-m ... -bitbucket

This affects at least RuyDos, Ifrit, Brainless, Giraffe and Felpo
I am in the process of checking what wayback has saved there and add those links to the XB/UCI chronology.

For RuyDos it looks good (not checked yet others)
https://web.archive.org/web/*/https://b ... oh/ruydos/*
https://rwbc-chess.de
Greg Strong@ovyron wrote: What I think is that you have a very, very, VERY strong desire to post, and very little self-control.
All the time. In every conversation. You must post. Something. It's like you just can't help yourself.

Guenther
Posts: 3342
Joined: Wed Oct 01, 2008 4:33 am
Location: Regensburg, Germany
Full name: Guenther Simon
Contact:

Re: New engine releases 2020

Post by Guenther » Tue Sep 08, 2020 4:00 pm

Guenther wrote:
Tue Sep 08, 2020 1:01 pm
New releases not mentioned yet:

Rodent IV 0.29 [src only - I will compile it today, except someone is faster ;-) ]
https://github.com/nescitus/rodent-iv/commits/master
Actually it is already at 0.30. There was no makefile or build instructions, but I could deduct it from Rodent III
in the hope of no dramatic changes in the build process and source structure and that worked.

I don't know though, if it has to be build separately for personalities (never used them before)?
Edit: Ah, I see it reads automatically the given personality files from the source folder for Rodent IV
(haven't checked though, if there are new ones)

If you wanna use certain given books with it look here (I didn't want to make the upload too big):
https://github.com/nescitus/rodent-iv/tree/master/books

Anyhow, I have uploaded a bmi2 and an 'old' (my hardware) compilation now.
https://rwbc-chess.de/download.htm

Below is the output from the 'old' compilation (removed all currlines etc.)

Code: Select all

uci
id name Rodent IV 0.30
id author Pawel Koziol, Bernhard C. Maerz (based on Sungorus 1.4 by Pablo Vazquez)
option name Clear Hash type button
option name Hash type spin default 16 min 1 max 16384
option name Threads type spin default 1 min 1 max 48
option name MultiPV type spin default 1 min 1 max 12
option name TimeBuffer type spin default 10 min 0 max 1000
option name PrintPv type check default true
option name Taunting type check default false
option name UCI_LimitStrength type check default true
option name UCI_Elo type spin default 2800 min 800 max 2800
option name UCI_Chess960 type check default false
option name Personality type combo default --- var --- var Amanda var Ampere var Cloe var Deborah var Defender var Dynamic var Grumpy var Morphy var Nimzowitsch var Partisan var Pawnsacker var Pedrita var Preston var Sim
ple var Spitfire var Strangler var Steinitz var Tarrasch var Tal
option name Verbose type check default false
option name Ponder type check default false
option name UseBook type check default true
option name VerboseBook type check default false
option name MobilityRebalancing type check default false
uciok
ucinewgame
go infinite
info depth 1
info depth 1 time 0 nodes 2 nps 0 score cp -6 pv a2a4
info depth 1 time 15 nodes 6 nps 400 score cp 26 pv d2d4
info depth 1 time 15 nodes 21 nps 1400 score cp 33 pv b1c3
info depth 1 time 15 nodes 23 nps 1533 score cp 37 pv g1f3
info depth 2 time 31 nodes 51 nps 1645 score cp 15 pv g1f3 g8f6
info depth 3 time 47 nodes 157 nps 3340 score cp 34 pv g1f3 g8f6 b1c3
info depth 4 time 62 nodes 448 nps 7225 score cp 17 pv g1f3 g8f6 b1c3 d7d5
info depth 5 time 78 nodes 1066 nps 13666 score cp 24 pv g1f3 g8f6 b1c3 d7d5 d2d3
info depth 6 time 93 nodes 1781 nps 19150 score cp 16 pv g1f3 g8f6 b1c3 d7d5 d2d3 d5d4
info depth 7 time 109 nodes 3458 nps 31724 score cp 18 pv g1f3 g8f6 b1c3 d7d5 d2d4 b8c6 c1f4
info depth 8 time 125 nodes 6783 nps 54264 score cp 15 pv g1f3 g8f6 b1c3 d7d5 d2d4 b8c6 c1f4 c8f5
info depth 9 time 171 nodes 15498 nps 90631 score cp 13 pv g1f3 g8f6 b1c3 d7d5 d2d4 b8c6 c1f4 c8g4 h2h3
info depth 9 time 203 nodes 23123 nps 113906 score cp 18 pv e2e4 d7d5 e4d5 g8f6 c2c4 c7c6 d2d4 c6d5 c1f4
info depth 10 time 234 nodes 35912 nps 153470 score cp 26 pv g1f3
info depth 10 time 265 nodes 39475 nps 148962 score cp 30 pv g1f3 g8f6 b1c3 d7d5 d2d4 b8c6 c1f4 c8g4 c3b5 a8c8
info depth 11 time 452 nodes 123135 nps 272422 score cp 16 pv g1f3 d7d5 e2e3 g8f6 b1c3 c8g4 h2h3 g4f3 d1f3 e7e5 e3e4 c7c6 e4d5 c6d5
info depth 12 time 530 nodes 152014 nps 286818 score cp 16 pv g1f3 d7d5 e2e3 g8f6 b1c3 c8g4 h2h3 g4f3 d1f3 e7e5 e3e4 c7c6 e4d5 c6d5
info depth 13 time 764 nodes 252142 nps 330028 score cp 24 pv e2e4
info depth 13 time 827 nodes 281526 nps 340418 score cp 32 pv e2e4
info depth 13 time 1373 nodes 538502 nps 392208 score cp 34 pv e2e4 e7e5 b1c3 b8c6 g1f3 g8f6 f1b5 f8b4 e1g1 e8g8 b5c6 d7c6 f3e5 b4c3 d2c3 d8d1 f1d1 f6e4
info depth 14 time 1653 nodes 653886 nps 395575 score cp 28 pv e2e4 e7e5 g1f3 g8f6 f3e5 d7d6 e5f3 f6e4 d2d3 e4f6 b1c3 f8e7 f1e2 e8g8 e1g1
info depth 15 time 1934 nodes 775157 nps 400805 score cp 28 pv e2e4 e7e5 g1f3 g8f6 f3e5 d7d6 e5f3 f6e4 d2d3 e4f6 b1c3 f8e7 f1e2 e8g8 e1g1
info depth 16 time 3213 nodes 1359974 nps 423272 score cp 19 pv e2e4 e7e5 g1f3 g8f6 b1c3 b8c6 f1b5 f8b4 e1g1 e8g8 d2d3 d7d6 c1g5 b4c3 b2c3 c8g4 a1b1 g4f3 d1f3
info depth 17 time 4383 nodes 1877131 nps 428275 score cp 21 pv e2e4 e7e5 g1f3 g8f6 f3e5 d7d6 e5f3 f6e4 d2d3 e4f6 f1e2 f8e7 e1g1 c8d7 b1c3 b8c6 c1e3 e8g8
info depth 18 time 6224 nodes 2657165 nps 426922 score cp 29 pv e2e4
info depth 18 time 7113 nodes 3070091 nps 431616 score cp 37 pv e2e4
info depth 18 time 10374 nodes 4581572 nps 441639 score cp 47 pv e2e4 e7e5 g1f3 g8f6 f3e5 d7d6 e5f3 f6e4 b1c3 e4c3 d2c3 f8e7 c1e3 c8e6 f1d3 b8d7 e1g1 e8g8 f1e1
info depth 19 time 12683 nodes 5518161 nps 435083 score cp 39 pv e2e4 e7e5 g1f3 g8f6 f3e5 d7d6 e5f3 f6e4 b1c3 e4c3 d2c3 b8c6 f1d3 f8e7 c1f4 c8e6 e1g1 e8g8 f1e1 f8e8
 
https://rwbc-chess.de
Greg Strong@ovyron wrote: What I think is that you have a very, very, VERY strong desire to post, and very little self-control.
All the time. In every conversation. You must post. Something. It's like you just can't help yourself.

User avatar
Scally
Posts: 133
Joined: Thu Sep 28, 2017 7:34 pm
Location: Bermondsey, London
Full name: Alan Cooper
Contact:

Re: New engine releases 2020

Post by Scally » Wed Sep 09, 2020 2:41 pm

Hi all,

Rodent IV is now v0.31

Here’s what I use to compile it, (I use Picochess paths instead):

Code: Select all

#makefile to compile Rodent IV on Linux
PREFIX=/usr/local

#define the directory for the executable file
BINDIR = $(PREFIX)/bin

#define the directory for the data files
DATADIRBOOKS = $(PREFIX)/share/rodentIV-books
DATADIRPERSONALITIES = $(PREFIX)/share/rodentIV-persons

# Define the CPP compiler
CXX = g++

# Define features
# Other options are -DBOOKGEN -DNO_THREADS
CFG = -DNDEBUG -DUSEGEN -DUSE_THREADS -DBOOKSPATH=$(DATADIRBOOKS) -DPERSONALITIESPATH=$(DATADIRPERSONALITIES)
CFG1 = -DUSEGEN -DUSE_THREADS -DBOOKSPATH=$(DATADIRBOOKS) -DPERSONALITIESPATH=$(DATADIRPERSONALITIES)

# Define compiler flags
CXXFLAGS = -g -std=c++14 -Wall -Wno-unknown-pragmas -O3 -fno-rtti -pthread $(CFG)
CXX1FLAGS = -g -std=c++14 -Wall -Wno-unknown-pragmas -pthread $(CFG1)

# Define linker options
LDFLAGS = -s -Wl,--no-as-needed,--gc-sections
LD1FLAGS = -lm

# Command line switches
lto = yes
native = no

# Compile with Link Time Optimisation
ifeq ($(lto),yes)
	CXXFLAGS += -flto -latomic
	LDFLAGS += -flto
endif

# Optimise for local machines
ifeq ($(native),yes)
	CXXFLAGS += -march=native
	LDFLAGS += -march=native
endif

# Define outpout name and settings file
EXENAME= rodentIV
CONFIGFILE = basic.ini

.PHONY: clean install update remove help

default: build

build:
	@echo "Type make help for additional options"
	$(CXX) $(CXXFLAGS) $(LDFLAGS) -o $(EXENAME) src/*.cpp
	@echo "SHOW_OPTIONS" > $(CONFIGFILE)

build-popcnt:
	@echo "Type make help for additional options"
	$(CXX) $(CXXFLAGS) -msse3 -mpopcnt $(LDFLAGS) -o $(EXENAME) src/*.cpp
	@echo "SHOW_OPTIONS" > $(CONFIGFILE)

clang-profiled:
	@echo "Type make help for additional options"
#	$(CXX) $(LDFLAGS) $(CXXFLAGS) -fprofile-instr-generate -o $(EXENAME) src/*.cpp
	$(CXX) $(LDFLAGS) $(CXXFLAGS) -fprofile-generate -o $(EXENAME) src/*.cpp
	@echo bench | ./rodentIV > /dev/null
	@llvm-profdata merge -output=default.profdata default.profraw
	$(CXX) $(CXXFLAGS) -fprofile-instr-use=default.profdata $(LDFLAGS) -o $(EXENAME) src/*.cpp
	@echo "SHOW_OPTIONS" > $(CONFIGFILE)
	@rm -f *.profdata *.profraw

clang-popcnt-profiled:
	@echo "Type make help for additional options"
	$(CXX) $(CXXFLAGS) -msse3 -mpopcnt -fprofile-instr-generate $(LDFLAGS) -o $(EXENAME) src/*.cpp
	@echo bench | ./rodentIV > /dev/null
	@llvm-profdata merge -output=default.profdata default.profraw
	$(CXX) $(CXXFLAGS) -msse3 -mpopcnt -fprofile-instr-use=default.profdata $(LDFLAGS) -o $(EXENAME) src/*.cpp
	@echo "SHOW_OPTIONS" > $(CONFIGFILE)
	@rm -f *.profdata *.profraw

gcc-profiled:
	@echo "Type make help for additional options"
	$(CXX) $(CXXFLAGS) -fprofile-generate $(LDFLAGS) -o $(EXENAME) src/*.cpp
	@echo bench | ./rodentIV > /dev/null
	$(CXX) $(CXXFLAGS) -fprofile-use $(LDFLAGS) -o $(EXENAME) src/*.cpp
	@echo "SHOW_OPTIONS" > $(CONFIGFILE)
	@rm -f *.gcda

gcc-popcnt-profiled:
	@echo "Type make help for additional options"
	$(CXX) $(CXXFLAGS) -msse3 -mpopcnt -fprofile-generate $(LDFLAGS) -o $(EXENAME) src/*.cpp
	@echo bench | ./rodentIV > /dev/null
	$(CXX) $(CXXFLAGS) -msse3 -mpopcnt -fprofile-use $(LDFLAGS) -o $(EXENAME) src/*.cpp
	@echo "SHOW_OPTIONS" > $(CONFIGFILE)
	@rm -f *.gcda

build-static:
	@echo "Type make help for additional options"
	$(CXX) $(CXXFLAGS) $(LDFLAGS) -static -o $(EXENAME) src/*.cpp
	@echo "SHOW_OPTIONS" > $(CONFIGFILE)

build-debug:
	@echo "Type make help for additional options"
	$(CXX) $(CXX1FLAGS) $(LD1FLAGS) -o $(EXENAME) src/*.cpp
	@echo "SHOW_OPTIONS" > $(CONFIGFILE)

gcc-lcov:
	@echo "Type make help for additional options"
	$(CXX) $(CXXFLAGS) -fprofile-generate -ftest-coverage $(LDFLAGS) -o $(EXENAME) src/*.cpp
	@echo bench | ./rodentIV > /dev/null
	@lcov --capture --directory . --no-external --gcov-tool gcov --output-file coverage.info > /dev/null
	@genhtml coverage.info --output-directory coveragedir > /dev/null
	@rm -f rodentIV *.gcda *.gcno coverage.info

bookgen:
	@echo "Type make help for additional options"
	$(CXX) $(CXXFLAGS) -DBOOKGEN $(LDFLAGS) -o $(EXENAME)-bookgen src/*.cpp
	@echo "SHOW_OPTIONS" > $(CONFIGFILE)
	@./$(EXENAME)-bookgen
	@rm -f $(EXENAME)-bookgen

clean:
	rm -rf $(EXENAME) $(EXENAME).exe book_gen.h basic.ini *.profdata *.profraw *.gcda *.gcno coverage.info coveragedir

install:
	mkdir -p $(BINDIR)
	mkdir -p $(DATADIRBOOKS)
	mkdir -p $(DATADIRPERSONALITIES)
	cp $(EXENAME) $(BINDIR)
	cp $(CONFIGFILE) $(DATADIRPERSONALITIES)
	#find ../books/ -name '*.bin' -type f -exec cp '{}' $(DATADIRBOOKS) \;
	#find ../personalities/ -name '*.ini' -type f -exec cp '{}' $(DATADIRPERSONALITIES) \;
	cp -r ../books/* $(DATADIRBOOKS)
	cp -r ../personalities/* $(DATADIRPERSONALITIES)
	chmod 755 $(BINDIR)/$(EXENAME)
	chmod -R 644 $(DATADIRBOOKS)/*
	find $(DATADIRBOOKS) -type d -exec chmod 755 {} \;
	chmod -R 644 $(DATADIRPERSONALITIES)/*
	find $(DATADIRPERSONALITIES) -type d -exec chmod 755 {} \;

# This just performs a copy of rodentIV engine
update:
	cp $(EXENAME) $(BINDIR)

remove:
	rm -f $(BINDIR)/$(EXENAME)
	rm -rf $(DATADIRBOOKS)
	rm -rf $(DATADIRPERSONALITIES)

help:
	@echo ""
	@echo "To compile RodentIV type:"
	@echo ""
	@echo "make build			> Build Rodent IV"
	@echo "make build-popcnt		> Build with fast popcnt"
	@echo "make clang-profiled		> Build with pgo optimisations (clang++ required)"
	@echo "make clang-popcnt-profiled	> Build with pgo+popcnt optimisations (clang++ required)"
	@echo "make gcc-profiled		> Build with pgo optimisations (g++ required)"
	@echo "make gcc-popcnt-profiled	> Build with pgo+popcnt optimisations (g++ required)"
	@echo "make build-static		> Build a static binary"
	@echo "make build-debug		> Build a debug version"
	@echo "make gcc-lcov			> Code coverage with lcov (g++ required)"
	@echo "make bookgen			> Build internal book"
	@echo "make clean			> Clean up"
	@echo "make install			> Install RodentIV (root privileges required)"
	@echo "make update			> Update RodentIV engine (root privileges required)"
	@echo "make remove			> Remove RodentIV from your system (root privileges required)"

Cheers,

Al.
Alan Cooper
My Chess Computers

Gabor Szots
Posts: 582
Joined: Sat Jul 21, 2018 5:43 am
Location: Szentendre, Hungary
Full name: Gabor Szots

Re: New engine releases 2020

Post by Gabor Szots » Wed Sep 16, 2020 8:27 am

Scally wrote:
Wed Sep 09, 2020 2:41 pm
Hi all,

Rodent IV is now v0.31
Now at 0.32.
Gabor Szots
CCRL testing group

Guenther
Posts: 3342
Joined: Wed Oct 01, 2008 4:33 am
Location: Regensburg, Germany
Full name: Guenther Simon
Contact:

Re: New engine releases 2020

Post by Guenther » Wed Sep 16, 2020 10:14 am

Uploaded several compiles for Defenchess 2.3 dev which seems it was not compiled before?
(dev and dev2 are for older hardware w/o popcount, for dev3 and dev4 you need to check, if and how fast it runs on your hardware,
one is bmi2 + avx the other bmi2 + avx2)

It had a lot of changes + bugfixes and chess960 support added until January 2020.

https://github.com/cetincan0/Defenchess/commits/master
https://rwbc-chess.de/download.htm
https://rwbc-chess.de
Greg Strong@ovyron wrote: What I think is that you have a very, very, VERY strong desire to post, and very little self-control.
All the time. In every conversation. You must post. Something. It's like you just can't help yourself.

carldaman
Posts: 1985
Joined: Sat Jun 02, 2012 12:13 am

Re: New engine releases 2020

Post by carldaman » Fri Sep 18, 2020 10:56 am

Any download links for Barbarossa 0.6.0 (github only has v0.5) or Fruit Reloaded 15A?

I seem to recall that Daniel Mehrmann was considering taking his Fruit fork private, but I'm not sure.

Guenther
Posts: 3342
Joined: Wed Oct 01, 2008 4:33 am
Location: Regensburg, Germany
Full name: Guenther Simon
Contact:

Re: New engine releases 2020

Post by Guenther » Fri Sep 18, 2020 11:13 am

carldaman wrote:
Fri Sep 18, 2020 10:56 am
Any download links for Barbarossa 0.6.0 (github only has v0.5)
Obviously it is private for now and no commits were done to github anymore after 0.50.
You have to ask CCRL/Graham, who tested it since around 2019-08, or even better the author himself (but better in another thread).
https://rwbc-chess.de
Greg Strong@ovyron wrote: What I think is that you have a very, very, VERY strong desire to post, and very little self-control.
All the time. In every conversation. You must post. Something. It's like you just can't help yourself.

JohnWoe
Posts: 189
Joined: Sat Mar 02, 2013 10:31 pm

Re: New engine releases 2020

Post by JohnWoe » Sat Sep 19, 2020 7:41 pm

2 more engines. Both are 100% Sapeli clones.

Mayhem 0.1:
Sapeli in C++14. Got rid of C ugliness. No more strcpy/strncmp/etc stupidity.
This uses huge databases.
Those are automically generated by Stockfish 12.
Mayhem scans the database with extra thread while searching.
If Mayhem plays some position badly. Just add the correct move to the list.
Do it for all moves and Mayhem plays perfect chess :D
Release: https://github.com/SamuraiDangyo/mayhem ... s/tag/v0.1

MiniSapeli 0.1:
Just a cleaned up UCI only Sapeli. Not even worth of own repository/branch. 1681 sloc.
Release: https://github.com/SamuraiDangyo/Sapeli ... MiniSapeli

Terje
Posts: 225
Joined: Tue Nov 19, 2019 3:34 am
Location: https://github.com/TerjeKir/weiss
Full name: Terje Kirstihagen

Re: New engine releases 2020

Post by Terje » Sat Sep 19, 2020 8:54 pm

JohnWoe wrote:
Sat Sep 19, 2020 7:41 pm
2 more engines. Both are 100% Sapeli clones.

Mayhem 0.1:
Sapeli in C++14. Got rid of C ugliness. No more strcpy/strncmp/etc stupidity.
This uses huge databases.
Those are automically generated by Stockfish 12.
Mayhem scans the database with extra thread while searching.
If Mayhem plays some position badly. Just add the correct move to the list.
Do it for all moves and Mayhem plays perfect chess :D
Release: https://github.com/SamuraiDangyo/mayhem ... s/tag/v0.1

MiniSapeli 0.1:
Just a cleaned up UCI only Sapeli. Not even worth of own repository/branch. 1681 sloc.
Release: https://github.com/SamuraiDangyo/Sapeli ... MiniSapeli
Using more than the expected number of threads (1, unless Threads is set to something else) for heavy work can negatively impact testing by taking thinking time away from other engines (or itself) when multiple testing games are run simultaneously on the same machine.

Post Reply