Spike 1.4 and 6-man Nalimov

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

Moderators: hgm, Rebel, chrisw

MOBMAT
Posts: 385
Joined: Sat Feb 04, 2017 11:57 pm
Location: USA

Spike 1.4 and 6-man Nalimov

Post by MOBMAT »

I guess Spike 1.4 won't work the 6-man Nalimov

i tried this...

Code: Select all

setoption name Hash value 1024
setoption name NalimovPath value H:\nalimov
and get the error...

*** Unable to read H:\nalimov\kqrbkb.nbb.emd - out of memory

has anyone had success?

BTW, it uses the 5-man files just fine
i7-6700K @ 4.00Ghz 32Gb, Win 10 Home, EGTBs on PCI SSD
Benchmark: Stockfish15.1 NNUE x64 bmi2 (nps): 1277K
chysiddh14
Posts: 38
Joined: Tue Jan 01, 2019 9:34 am
Full name: Siddhartha Chaudhary

Re: Spike 1.4 and 6-man Nalimov

Post by chysiddh14 »

my fav engine but it never gets update
Colin-G
Posts: 191
Joined: Mon Oct 31, 2016 6:30 pm
Location: England

Re: Spike 1.4 and 6-man Nalimov

Post by Colin-G »

chysiddh14 wrote: Sat Jun 06, 2020 6:01 am my fav engine but it never gets update
Its also a pity that it never had a native linux version, unlike the earlier Spike 1.2
Ferdy
Posts: 4833
Joined: Sun Aug 10, 2008 3:15 pm
Location: Philippines

Re: Spike 1.4 and 6-man Nalimov

Post by Ferdy »

MOBMAT wrote: Sat Jun 06, 2020 4:44 am I guess Spike 1.4 won't work the 6-man Nalimov

i tried this...

Code: Select all

setoption name Hash value 1024
setoption name NalimovPath value H:\nalimov
and get the error...

*** Unable to read H:\nalimov\kqrbkb.nbb.emd - out of memory

has anyone had success?

BTW, it uses the 5-man files just fine
krbknn works here.

Code: Select all

Spike 1.4 (Build 84) by Volker Boehm & Ralf Schaefer, Book by Timo Klaustermeyer
uci
Failed to open configuration file Spike.cnfg
CPUs used      : 1
Log file       : disabled
Book file      : internal book
Max. book depth: 60
Book author    : Timo Klaustermeyer
Book name      : Spike Standard Book 2010
Book version   : 1.4
Book type (no.): 2
Book moves     : 234226
Hash size      : 32
Resign value   : 700 cp
Contempt       : 0
EGTB Horizont  : 4
Engame tables  : not found
Pos. learning  : false
Learn filename : SpikePosLearn.txt
Book learning  : true
Book learn file: SpikeBookLearn.scbl
id name Spike 1.4
id author Ralf SchΣfer und Volker B÷hm
option name UCI_ShowCurrLine type check default false
option name UCI_ShowRefutations type check default false
option name MultiPV type spin default 1 min 1 max 99
option name CPUs type spin default 1 min 1 max 12
option name OwnBook type check default true
option name Hash type spin default 32 min 2 max 1024
option name Logfile type check default false
option name Position Learn type check default false
option name Position Learn Filename type string default SpikePosLearn.txt
option name Book Learn type check default true
option name Book Learn Filename type string default SpikeBookLearn.scbl
option name NalimovPath type string default C:\TB
option name NalimovCache type spin min 1 max 2048 default 4
option name EGTB depth type combo default desktop var never var rootonly var laptop var desktop var server var usb var allways
option name UCI_Chess960 type check default false
option name UCI_LimitStrength type check default false
option name UCI_Elo type spin default 1300 min 1100 max 2500
option name UCI_EngineAbout type string default Spike by Ralf SchΣfer & Volker B÷hm, http:\spike.lazypics.de
uciok
setoption name NalimovPath value d:\Tablebases\TB6
Using 4.000 MB cache for TBs
isready
readyok
ucinewgame
position fen 4nk2/4n3/R7/4K3/4B3/8/8/8 w - - 0 1
go movetime 5000
info depth 0 score cp 30 pv nm  hashfull 0 time 0 nodes 0 tbhits 0
info depth 1 score cp 30 pv e4f3   hashfull 0 time 0 nodes 3 tbhits 0
info depth 1 score cp 30 pv e4f3   hashfull 0 time 0 nodes 3 tbhits 0
info depth 2 score mate 24 pv e4f3   hashfull 0 time 0 nodes 5 tbhits 1
info depth 2 score mate 24 pv e4f3   hashfull 0 time 0 nodes 5 tbhits 1
info depth 3 score mate 24 pv e4f3   hashfull 0 time 0 nodes 7 tbhits 2
info depth 3 score mate 24 pv e4f3   hashfull 0 time 0 nodes 7 tbhits 2
info depth 4 score mate 24 pv e4f3   hashfull 0 time 0 nodes 9 tbhits 3
info depth 4 score mate 24 pv e4f3   hashfull 0 time 0 nodes 9 tbhits 3
info depth 5 score mate 24 pv e4f3   hashfull 0 time 0 nodes 11 tbhits 4
info depth 5 score mate 24 pv e4f3   hashfull 0 time 0 nodes 11 tbhits 4
info depth 6 score mate 24 pv e4f3 e7c8   hashfull 0 time 0 nodes 43 tbhits 33
info depth 6 score mate 24 pv e4f3 e7c8   hashfull 0 time 0 nodes 43 tbhits 33
info depth 7 score mate 24 pv e4f3 e7c8 f3h5   hashfull 0 nps 7687 time 16 nodes 123 tbhits 107
info depth 7 score mate 24 pv e4f3 e7c8 f3h5   hashfull 0 nps 7687 time 16 nodes 123 tbhits 107
info depth 8 score mate 24 pv e4f3 e7c8 f3h5 c8d6   hashfull 0 nps 12187 time 16 nodes 195 tbhits 170
info depth 8 score mate 24 pv e4f3 e7c8 f3h5 c8d6   hashfull 0 nps 12187 time 16 nodes 195 tbhits 170
info depth 9 score mate 24 pv e4f3 e7c8 f3h5 c8d6 a6a7   hashfull 0 nps 19250 time 16 nodes 308 tbhits 257
info depth 9 score mate 24 pv e4f3 e7c8 f3h5 c8d6 a6a7   hashfull 0 nps 19250 time 16 nodes 308 tbhits 257
info depth 10 score mate 24 pv e4f3 e7c8 f3h5 c8d6 a6a7 d6c4   hashfull 0 nps 32312 time 16 nodes 517 tbhits 346
info depth 10 score mate 24 pv e4f3 e7c8 f3h5 c8d6 a6a7 d6c4   hashfull 0 nps 32312 time 16 nodes 517 tbhits 346
bestmove e4f3 ponder e7c8
MOBMAT
Posts: 385
Joined: Sat Feb 04, 2017 11:57 pm
Location: USA

Re: Spike 1.4 and 6-man Nalimov

Post by MOBMAT »

I am using Arena, which doesn't seem to matter since the engine process stops even when run from a command line.
Here is the communication to/from Arena

Code: Select all

2020-06-06 19:31:58.305*1*------------------------Starting engine 1 Spike 1.4-------------------------
2020-06-06 19:31:58.305*1*Configured Engine 1 Type:   UCI
2020-06-06 19:31:58.306*1*Engine 1 dir: E:\Chess\Arena\Engines\Spike
2020-06-06 19:31:58.306*1*Engine 1 commandline: E:\Chess\Arena\Engines\Spike\Spike1.4.exe 
2020-06-06 19:31:58.412*1*Child Process Prio Adj:  PID 4552 conhost.exe
2020-06-06 19:31:58.412*1*Engine 1 ProcessID: 7624
2020-06-06 19:31:58.412*1*Engine 1 Prio:32 ThreadPrio:0
2020-06-06 19:31:58.433<--1:Spike 1.4 (Build 84) by Volker Boehm & Ralf Schaefer, Book by Timo Klaustermeyer
2020-06-06 19:31:58.433-->1:uci
2020-06-06 19:31:59.087<--1:CPUs used      : 1
2020-06-06 19:31:59.088<--1:Log file       : disabled
2020-06-06 19:31:59.088<--1:Book file      : internal book
2020-06-06 19:31:59.088<--1:Max. book depth: 60 
2020-06-06 19:31:59.089<--1:Book author    : Timo Klaustermeyer
2020-06-06 19:31:59.089<--1:Book name      : Spike Standard Book 2010
2020-06-06 19:31:59.089<--1:Book version   : 1.4
2020-06-06 19:31:59.089<--1:Book type (no.): 2
2020-06-06 19:31:59.089<--1:Book moves     : 234226
2020-06-06 19:31:59.089<--1:Hash size      : 32
2020-06-06 19:31:59.089<--1:Resign value   : 900 cp
2020-06-06 19:31:59.089<--1:Contempt       : 0
2020-06-06 19:31:59.089<--1:EGTB Horizont  : 4
2020-06-06 19:31:59.123<--1:Engame tables  : not found
2020-06-06 19:31:59.123<--1:Pos. learning  : false
2020-06-06 19:31:59.124<--1:Learn filename : 
2020-06-06 19:31:59.124<--1:Book learning  : false
2020-06-06 19:31:59.124<--1:Book learn file: 
2020-06-06 19:31:59.124<--1:id name Spike 1.4
2020-06-06 19:31:59.124<--1:id author Ralf Schäfer und Volker Böhm
2020-06-06 19:31:59.124<--1:option name UCI_ShowCurrLine type check default false
2020-06-06 19:31:59.124<--1:option name UCI_ShowRefutations type check default false
2020-06-06 19:31:59.124<--1:option name MultiPV type spin default 1 min 1 max 99
2020-06-06 19:31:59.124<--1:option name CPUs type spin default 1 min 1 max 12
2020-06-06 19:31:59.125<--1:option name OwnBook type check default true
2020-06-06 19:31:59.125<--1:option name Hash type spin default 32 min 2 max 1024
2020-06-06 19:31:59.125<--1:option name Logfile type check default false
2020-06-06 19:31:59.125<--1:option name Position Learn type check default false
2020-06-06 19:31:59.125<--1:option name Position Learn Filename type string default 
2020-06-06 19:31:59.125<--1:option name Book Learn type check default false
2020-06-06 19:31:59.125<--1:option name Book Learn Filename type string default 
2020-06-06 19:31:59.125<--1:option name NalimovPath type string default 
2020-06-06 19:31:59.125<--1:option name NalimovCache type spin min 1 max 2048 default 4
2020-06-06 19:31:59.125<--1:option name EGTB depth type combo default desktop var never var rootonly var laptop var desktop var server var usb var allways
2020-06-06 19:31:59.125<--1:option name UCI_Chess960 type check default false
2020-06-06 19:31:59.125<--1:option name UCI_LimitStrength type check default false
2020-06-06 19:31:59.125<--1:option name UCI_Elo type spin default 1300 min 1100 max 2500
2020-06-06 19:31:59.125<--1:option name UCI_EngineAbout type string default Spike by Ralf Schäfer & Volker Böhm, http:\spike.lazypics.de
2020-06-06 19:31:59.125<--1:uciok
2020-06-06 19:31:59.125-->1:setoption name OwnBook value false
2020-06-06 19:31:59.125-->1:setoption name Hash value 1024
2020-06-06 19:31:59.125-->1:setoption name NalimovPath value h:\nalimov
2020-06-06 19:31:59.125-->1:setoption name EGTB depth value usb
2020-06-06 19:31:59.130*1*Child Process Prio Adj:  PID 4552 conhost.exe
2020-06-06 19:31:59.130-->1:isready
2020-06-06 19:32:10.226<--1:*** Unable to read h:\nalimov\kqrbkb.nbb.emd - out of memory
And the engine is dead, it never starts searching after clicking the "analyze" button.
When I run it from the command prompt, the process aborts some seconds after I issue the "setoption name NalimovPath value h:\nalimov" command.
I have 32Gb of memory, so I'm not really out of memory.
Running Windows 10 64-bit.
The EGTB files are fine, I ran the checksum test prior to testing.
My guess is there is a memory issue when starting a process, perhaps a limit on the number of files that can be opened.
More research required...
i7-6700K @ 4.00Ghz 32Gb, Win 10 Home, EGTBs on PCI SSD
Benchmark: Stockfish15.1 NNUE x64 bmi2 (nps): 1277K