Page 3 of 5

Re: Lc0 - macOS binary requested

Posted: Sat May 25, 2019 10:46 pm
by smcracraft
Thanks,

I was able to get it working on the Mac with no GPU but it plays horribly compared to the AWS Ubuntu and Windows 10 instances with GPU.

Check this out. Thoughts?

Stuarts-MacBook-Pro:release stuartcracraft$ ./lc0
_
| _ | |
|_ |_ |_| v0.22.0-dev built May 25 2019
ucinewgame
Found pb network file: ./weights_run1_42372.pb.gz
Creating backend [opencl]...
OpenCL, maximum batch size set to 16.
Initializing OpenCL.
Detected 1 OpenCL platforms.
Platform version: OpenCL 1.2 (Feb 22 2019 20:16:07)
Platform profile: FULL_PROFILE
Platform name: Apple
Platform vendor: Apple
Device ID: 0
Device name: Intel(R) Core(TM) i7-7920HQ CPU @ 3.10GHz
Device type: CPU
Device vendor: Intel
Device driver: 1.1
Device speed: 3100 MHZ
Device cores: 8 CU
Device score: 512
Device ID: 1
Device name: Intel(R) HD Graphics 630
Device type: GPU
Device vendor: Intel Inc.
Device driver: 1.2(Mar 11 2019 21:25:40)
Device speed: 1100 MHZ
Device cores: 24 CU
Device score: 612
Device ID: 2
Device name: AMD Radeon Pro 560 Compute Engine
Device type: GPU
Device vendor: AMD
Device driver: 1.2 (Mar 11 2019 21:14:43)
Device speed: 300 MHZ
Device cores: 16 CU
Device score: 1112
Selected platform: Apple
Selected device: AMD Radeon Pro 560 Compute Engine
with OpenCL 1.2 capability.
Started OpenCL SGEMM tuner with batch size 16.
Will try 578 valid configurations.
(1/578) KWG=32 KWI=2 MDIMA=8 MDIMC=8 MWG=16 NDIMB=8 NDIMC=8 NWG=16 SA=0 SB=0 STRM=0 STRN=0 VWM=1 VWN=1 10955.8 us (49.0 GFLOPS)
(2/578) KWG=32 KWI=2 MDIMA=8 MDIMC=8 MWG=32 NDIMB=8 NDIMC=8 NWG=16 SA=0 SB=0 STRM=0 STRN=0 VWM=1 VWN=1 6267.1 us (85.7 GFLOPS)
(4/578) KWG=32 KWI=2 MDIMA=8 MDIMC=8 MWG=16 NDIMB=8 NDIMC=8 NWG=32 SA=0 SB=0 STRM=0 STRN=0 VWM=1 VWN=1 6194.0 us (86.7 GFLOPS)
(5/578) KWG=32 KWI=2 MDIMA=8 MDIMC=8 MWG=32 NDIMB=8 NDIMC=8 NWG=32 SA=0 SB=0 STRM=0 STRN=0 VWM=1 VWN=1 2990.0 us (179.6 GFLOPS)
(8/578) KWG=32 KWI=2 MDIMA=8 MDIMC=8 MWG=32 NDIMB=8 NDIMC=8 NWG=64 SA=0 SB=0 STRM=0 STRN=0 VWM=1 VWN=1 2788.1 us (192.6 GFLOPS)
(115/578) KWG=32 KWI=2 MDIMA=8 MDIMC=8 MWG=32 NDIMB=8 NDIMC=8 NWG=32 SA=0 SB=0 STRM=0 STRN=0 VWM=4 VWN=1 2681.6 us (200.2 GFLOPS)
(117/578) KWG=32 KWI=2 MDIMA=8 MDIMC=8 MWG=32 NDIMB=8 NDIMC=8 NWG=64 SA=0 SB=0 STRM=0 STRN=0 VWM=4 VWN=1 2172.8 us (247.1 GFLOPS)
(121/578) KWG=32 KWI=2 MDIMA=16 MDIMC=16 MWG=64 NDIMB=8 NDIMC=8 NWG=64 SA=0 SB=0 STRM=0 STRN=0 VWM=4 VWN=1 2165.8 us (247.9 GFLOPS)
go nodes 1
(458/578) KWG=32 KWI=2 MDIMA=16 MDIMC=16 MWG=64 NDIMB=16 NDIMC=16 NWG=32 SA=1 SB=1 STRM=0 STRN=0 VWM=1 VWN=2 2041.8 us (262.9 GFLOPS)
Wavefront/Warp size: 64

Max workgroup size: 256
Max workgroup dimensions: 256 256 256
info depth 1 seldepth 1 time 102 nodes 1 score cp 12 hashfull 0 nps 9 tbhits 0 pv d2d4
bestmove d2d4
go nodes 10
info depth 2 seldepth 2 time 236 nodes 4 score cp 17 hashfull 0 nps 12 tbhits 0 pv e2e4 e7e5
info depth 2 seldepth 3 time 395 nodes 8 score cp 10 hashfull 0 nps 17 tbhits 0 pv g1f3 d7d5 d2d4
info depth 3 seldepth 4 time 731 nodes 17 score cp 11 hashfull 0 nps 21 tbhits 0 pv g1f3 d7d5 d2d4 g8f6
bestmove g1f3 ponder d7d5
go nodes 100
info depth 2 seldepth 2 time 102 nodes 21 score cp 11 hashfull 0 nps 19 tbhits 0 pv g1f3 e7e6 e2e4 d7d5
info depth 3 seldepth 5 time 433 nodes 30 score cp 10 hashfull 0 nps 25 tbhits 0 pv g1f3 d7d5 g2g3 c7c5 f1g2
info depth 4 seldepth 6 time 768 nodes 48 score cp 10 hashfull 0 nps 37 tbhits 0 pv g1f3 d7d5 g2g3 c7c5 f1g2
info depth 4 seldepth 7 time 1119 nodes 83 score cp 9 hashfull 1 nps 57 tbhits 0 pv g1f3 d7d5 g2g3 e7e6 f1g2 g7g6 g3g4
info depth 4 seldepth 7 time 1137 nodes 100 score cp -17 hashfull 1 nps 71 tbhits 0 pv g1f3 d7d5 g2g3 e7e6 f1g2 g7g6 d2d3
bestmove g1f3 ponder d7d5
position startpos moves g1f3 d7d5
go nodes 100
info depth 5 seldepth 6 time 555 nodes 35 score cp 180 hashfull 1 nps 32 tbhits 0 pv g2g3 g8f6 f1g2 c8d7 g3g4
info depth 6 seldepth 7 time 2337 nodes 218 score cp 103 hashfull 3 nps 86 tbhits 0 pv g2g3 c7c5 f1g2 d8c7 g2f1 g7g6 h2h3
bestmove g2g3 ponder c7c5
position startpos moves g1f3 d7d5 g2g3 c7c5
go nodes 1000
info depth 2 seldepth 2 time 431 nodes 299 score cp 140 hashfull 3 nps 18 tbhits 0 pv f1g2 d8c7 h1f1 e7e6 h2h3
info depth 2 seldepth 6 time 431 nodes 310 score cp 135 hashfull 3 nps 44 tbhits 0 pv f1g2 d8c7 h1f1 e7e6 h2h3 d5d4
info depth 5 seldepth 7 time 1603 nodes 418 score cp 60 hashfull 4 nps 79 tbhits 0 pv f1g2 d8c7 e2e3 c7c6 b2b4 h7h6
info depth 5 seldepth 7 time 4130 nodes 790 score cp 30 hashfull 5 nps 120 tbhits 0 pv f1g2 d8c7 h1f1 e7e5 h2h3 c7a5 h3h4
bestmove f1g2 ponder d8c7
position startpos moves g1f3 d7d5 g2g3 c7c5 f1g2 b8c6
go nodes 3000
info depth 1 seldepth 2 time 171 nodes 13 score cp -12800 hashfull 7 nps 76 tbhits 0 pv f3e5 d5d4
info depth 1 seldepth 2 time 171 nodes 27 score cp -12800 hashfull 7 nps 157 tbhits 0 pv g3g4 d5d4
info depth 2 seldepth 3 time 205 nodes 28 score cp 0 hashfull 7 nps 136 tbhits 0 pv g3g4 d8c7 g4g5
info depth 3 seldepth 4 time 1773 nodes 267 score cp 30 hashfull 8 nps 150 tbhits 0 pv g3g4 c6e5 h2h3
info depth 3 seldepth 4 time 2055 nodes 332 score cp -115 hashfull 8 nps 161 tbhits 0 pv d2d4 c8f5 h1f1 f5c2
info depth 3 seldepth 4 time 2191 nodes 388 score cp 31 hashfull 8 nps 177 tbhits 0 pv g2h3 d8d6 g3g4 d5d4
info depth 3 seldepth 4 time 3010 nodes 506 score cp -112 hashfull 9 nps 168 tbhits 0 pv e2e3 d8d6 h1f1 h7h6
info depth 3 seldepth 5 time 3084 nodes 512 score cp -112 hashfull 9 nps 166 tbhits 0 pv e2e3 d8d6 h1f1 h7h6
info depth 3 seldepth 5 time 3719 nodes 572 score cp -5 hashfull 10 nps 153 tbhits 0 pv b1c3 d8d7 d2d4 h7h6
info depth 3 seldepth 6 time 4365 nodes 623 score cp -5 hashfull 10 nps 142 tbhits 0 pv b1c3 d8d7 d2d4 h7h6
info depth 3 seldepth 6 time 4716 nodes 718 score cp -14 hashfull 10 nps 152 tbhits 0 pv e1f1 b7b5 b1c3 a7a6 f3g5 a8b8
info depth 4 seldepth 6 time 5073 nodes 761 score cp -57 hashfull 11 nps 150 tbhits 0 pv e1f1 b7b5 b1c3 a7a6 f3g5 a8b8
info depth 4 seldepth 6 time 6442 nodes 901 score cp 42 hashfull 11 nps 139 tbhits 0 pv g2h3 d8d6 a2a3 g8f6 a1a2
info depth 4 seldepth 7 time 10966 nodes 1652 score cp 0 hashfull 15 nps 150 tbhits 0 pv g2h3 c8h3 d2d3 a8c8 c1h6 a7a6
info depth 5 seldepth 7 time 14616 nodes 2269 score cp 0 hashfull 18 nps 155 tbhits 0 pv g2h3 c8h3 d2d3 a8c8 c1h6 a7a6
info depth 5 seldepth 7 time 17307 nodes 2896 score cp -22 hashfull 22 nps 167 tbhits 0 pv g2h3 c8h3 d2d3 a8c8 c1h6 a7a6
bestmove g2h3 ponder c8h3
position startpos moves g1f3 d7d5 g2g3 c7c5 f1g2 b8c6 g2h3 e7e5
position startpos moves g1f3 d7d5 g2g3 c7c5 f1g2 b8c6 g2h3 e7e5
go movetime 20000
info depth 4 seldepth 4 time 433 nodes 28 score cp 269 hashfull 24 nps 27 tbhits 0 pv h1g1 a7a6 f3h4 h7h6
info depth 3 seldepth 4 time 790 nodes 64 score cp -11 hashfull 24 nps 60 tbhits 0 pv h1g1 a7a6 f3h4 h7h6 h4g6
info depth 4 seldepth 5 time 1474 nodes 124 score cp 58 hashfull 25 nps 73 tbhits 0 pv h1g1 a7a6 f3d4 e8e7 g3g4
info depth 4 seldepth 6 time 3630 nodes 475 score cp 9 hashfull 26 nps 126 tbhits 0 pv h1g1 a7a6 e2e3 b7b5 g1h1 h7h6
info depth 4 seldepth 6 time 5867 nodes 801 score cp 21 hashfull 28 nps 133 tbhits 0 pv h1g1 a7a6 e2e3 b7b5 g1h1 h7h6
bestmove h1g1 ponder a7a6

Re: Lc0 - macOS binary requested

Posted: Sat May 25, 2019 11:38 pm
by smcracraft
Also it crashes on the mac.

Stuarts-MacBook-Pro:release stuartcracraft$ ./lc0
_
| _ | |
|_ |_ |_| v0.22.0-dev built May 25 2019
ucinewgame
Found pb network file: ./weights_run1_42372.pb.gz
Creating backend [opencl]...
OpenCL, maximum batch size set to 16.
Initializing OpenCL.
Detected 1 OpenCL platforms.
Platform version: OpenCL 1.2 (Feb 22 2019 20:16:07)
Platform profile: FULL_PROFILE
Platform name: Apple
Platform vendor: Apple
Device ID: 0
Device name: Intel(R) Core(TM) i7-7920HQ CPU @ 3.10GHz
Device type: CPU
Device vendor: Intel
Device driver: 1.1
Device speed: 3100 MHZ
Device cores: 8 CU
Device score: 512
Device ID: 1
Device name: Intel(R) HD Graphics 630
Device type: GPU
Device vendor: Intel Inc.
Device driver: 1.2(Mar 11 2019 21:25:40)
Device speed: 1100 MHZ
Device cores: 24 CU
Device score: 612
Device ID: 2
Device name: AMD Radeon Pro 560 Compute Engine
Device type: GPU
Device vendor: AMD
Device driver: 1.2 (Mar 11 2019 21:14:43)
Device speed: 300 MHZ
Device cores: 16 CU
Device score: 1112
Selected platform: Apple
Selected device: AMD Radeon Pro 560 Compute Engine
with OpenCL 1.2 capability.
Loaded existing SGEMM tuning for batch size 16.
Wavefront/Warp size: 64

Max workgroup size: 256
Max workgroup dimensions: 256 256 256
go nodes 3000
info depth 1 seldepth 2 time 209 nodes 4 score cp 17 hashfull 0 nps 19 tbhits 0 pv e2e4 e7e5
info depth 2 seldepth 3 time 361 nodes 8 score cp 10 hashfull 0 nps 22 tbhits 0 pv g1f3 d7d5 d2d4
info depth 2 seldepth 4 time 705 nodes 10 score cp -71 hashfull 0 nps 14 tbhits 0 pv g1f3 e7e6 c2c4 f8e7
info depth 3 seldepth 4 time 706 nodes 19 score cp -201 hashfull 0 nps 26 tbhits 0 pv g1f3 e7e6 d2d4 h7h6
info depth 3 seldepth 5 time 1039 nodes 34 score cp 130 hashfull 0 nps 32 tbhits 0 pv e2e4 e7e5 b1c3 a7a6 c3d5
info depth 3 seldepth 5 time 1524 nodes 86 score cp -4 hashfull 1 nps 56 tbhits 0 pv e2e3 g8f6 g1f3 h8g8
info depth 3 seldepth 6 time 1879 nodes 127 score cp -71 hashfull 1 nps 67 tbhits 0 pv e2e3 g8f6 g1f3 h8g8
info depth 4 seldepth 7 time 2673 nodes 236 score cp 26 hashfull 2 nps 88 tbhits 0 pv e2e4 c7c6 c2c4 d8c7 g2g3 c6c5
info depth 4 seldepth 7 time 4609 nodes 566 score cp 13 hashfull 4 nps 122 tbhits 0 pv b1c3 d7d5 a1b1 f7f5 a2a3 h7h6
info depth 4 seldepth 7 time 6259 nodes 753 score cp 124 hashfull 4 nps 120 tbhits 0 pv f2f4 a7a5 h2h4 c7c5 a2a3
info depth 5 seldepth 8 time 10000 nodes 1470 score cp 59 hashfull 8 nps 147 tbhits 0 pv f2f4 e7e6 c2c3 h7h6 a2a3 a7a6 h2h3
Segmentation fault: 11
Stuarts-MacBook-Pro:release stuartcracraft$

Re: Lc0 - macOS binary requested

Posted: Sun May 26, 2019 2:48 am
by MikeB
smcracraft wrote: Sat May 25, 2019 11:38 pm Also it crashes on the mac.

Stuarts-MacBook-Pro:release stuartcracraft$ ./lc0
_
| _ | |
|_ |_ |_| v0.22.0-dev built May 25 2019
ucinewgame
Found pb network file: ./weights_run1_42372.pb.gz
Creating backend [opencl]...
OpenCL, maximum batch size set to 16.
Initializing OpenCL.
Detected 1 OpenCL platforms.
Platform version: OpenCL 1.2 (Feb 22 2019 20:16:07)
Platform profile: FULL_PROFILE
Platform name: Apple
Platform vendor: Apple
Device ID: 0
Device name: Intel(R) Core(TM) i7-7920HQ CPU @ 3.10GHz
Device type: CPU
Device vendor: Intel
Device driver: 1.1
Device speed: 3100 MHZ
Device cores: 8 CU
Device score: 512
Device ID: 1
Device name: Intel(R) HD Graphics 630
Device type: GPU
Device vendor: Intel Inc.
Device driver: 1.2(Mar 11 2019 21:25:40)
Device speed: 1100 MHZ
Device cores: 24 CU
Device score: 612
Device ID: 2
Device name: AMD Radeon Pro 560 Compute Engine
Device type: GPU
Device vendor: AMD
Device driver: 1.2 (Mar 11 2019 21:14:43)
Device speed: 300 MHZ
Device cores: 16 CU
Device score: 1112
Selected platform: Apple
Selected device: AMD Radeon Pro 560 Compute Engine
with OpenCL 1.2 capability.
Loaded existing SGEMM tuning for batch size 16.
Wavefront/Warp size: 64

Max workgroup size: 256
Max workgroup dimensions: 256 256 256
go nodes 3000
info depth 1 seldepth 2 time 209 nodes 4 score cp 17 hashfull 0 nps 19 tbhits 0 pv e2e4 e7e5
info depth 2 seldepth 3 time 361 nodes 8 score cp 10 hashfull 0 nps 22 tbhits 0 pv g1f3 d7d5 d2d4
info depth 2 seldepth 4 time 705 nodes 10 score cp -71 hashfull 0 nps 14 tbhits 0 pv g1f3 e7e6 c2c4 f8e7
info depth 3 seldepth 4 time 706 nodes 19 score cp -201 hashfull 0 nps 26 tbhits 0 pv g1f3 e7e6 d2d4 h7h6
info depth 3 seldepth 5 time 1039 nodes 34 score cp 130 hashfull 0 nps 32 tbhits 0 pv e2e4 e7e5 b1c3 a7a6 c3d5
info depth 3 seldepth 5 time 1524 nodes 86 score cp -4 hashfull 1 nps 56 tbhits 0 pv e2e3 g8f6 g1f3 h8g8
info depth 3 seldepth 6 time 1879 nodes 127 score cp -71 hashfull 1 nps 67 tbhits 0 pv e2e3 g8f6 g1f3 h8g8
info depth 4 seldepth 7 time 2673 nodes 236 score cp 26 hashfull 2 nps 88 tbhits 0 pv e2e4 c7c6 c2c4 d8c7 g2g3 c6c5
info depth 4 seldepth 7 time 4609 nodes 566 score cp 13 hashfull 4 nps 122 tbhits 0 pv b1c3 d7d5 a1b1 f7f5 a2a3 h7h6
info depth 4 seldepth 7 time 6259 nodes 753 score cp 124 hashfull 4 nps 120 tbhits 0 pv f2f4 a7a5 h2h4 c7c5 a2a3
info depth 5 seldepth 8 time 10000 nodes 1470 score cp 59 hashfull 8 nps 147 tbhits 0 pv f2f4 e7e6 c2c3 h7h6 a2a3 a7a6 h2h3
Segmentation fault: 11
Stuarts-MacBook-Pro:release stuartcracraft$
I was able to duplicate your error:

Code: Select all

Mac-Pro:release michaelbyrne$ lc0
       _
|   _ | |
|_ |_ |_| v0.22.0-dev built May 25 2019

uci
id name Lc0 v0.22.0-dev
id author The LCZero Authors.
option name WeightsFile type string default <autodiscover>
option name Backend type combo default opencl var opencl var blas var check var random var roundrobin var multiplexing var demux
option name BackendOptions type string default 
option name Threads type spin default 2 min 1 max 128
option name NNCacheSize type spin default 200000 min 0 max 999999999
option name MinibatchSize type spin default 256 min 1 max 1024
option name MaxPrefetch type spin default 32 min 0 max 1024
option name CPuct type string default 3.000000
option name CPuctBase type string default 19652.000000
option name CPuctFactor type string default 2.000000
option name Temperature type string default 0.000000
option name TempDecayMoves type spin default 0 min 0 max 100
option name TempCutoffMove type spin default 0 min 0 max 1000
option name TempEndgame type string default 0.000000
option name TempValueCutoff type string default 100.000000
option name TempVisitOffset type string default 0.000000
option name DirichletNoise type check default false
option name VerboseMoveStats type check default false
option name SmartPruningFactor type string default 1.330000
option name FpuStrategy type combo default reduction var reduction var absolute
option name FpuValue type string default 1.200000
option name FpuStrategyAtRoot type combo default same var reduction var absolute var same
option name FpuValueAtRoot type string default 1.000000
option name CacheHistoryLength type spin default 0 min 0 max 7
option name PolicyTemperature type string default 2.200000
option name MaxCollisionEvents type spin default 32 min 1 max 1024
option name MaxCollisionVisits type spin default 9999 min 1 max 1000000
option name OutOfOrderEval type check default true
option name StickyEndgames type check default true
option name SyzygyFastPlay type check default true
option name MultiPV type spin default 1 min 1 max 500
option name ScoreType type combo default centipawn var centipawn var centipawn_2018 var win_percentage var Q
option name HistoryFill type combo default fen_only var no var fen_only var always
option name KLDGainAverageInterval type spin default 100 min 1 max 10000000
option name MinimumKLDGainPerNode type string default 0.000000
option name Slowmover type string default 1.000000
option name MoveOverheadMs type spin default 200 min 0 max 100000000
option name SyzygyPath type string default 
option name Ponder type check default true
option name ImmediateTimeUse type string default 1.000000
option name RamLimitMb type spin default 0 min 0 max 100000000
option name ConfigFile type string default lc0.config
option name LogFile type string default 
uciok
ucinewgame
Found pb network file: ./weights_run1_42372.pb.gz
Creating backend [opencl]...
OpenCL, maximum batch size set to 16.
Initializing OpenCL.
Detected 1 OpenCL platforms.
Platform version: OpenCL 1.2 (Apr  7 2019 18:38:19)
Platform profile: FULL_PROFILE
Platform name:    Apple
Platform vendor:  Apple
Device ID:      0
Device name:    Intel(R) Xeon(R) CPU           X5690  @ 3.47GHz
Device type:    CPU
Device vendor:  Intel
Device driver:  1.1
Device speed:   3460 MHZ
Device cores:   24 CU
Device score:   512
Device ID:      1
Device name:    AMD Radeon HD 7950 Compute Engine
Device type:    GPU
Device vendor:  AMD
Device driver:  1.2 (Apr 29 2019 22:28:38)
Device speed:   150 MHZ
Device cores:   28 CU
Device score:   1112
Selected platform: Apple
Selected device: AMD Radeon HD 7950 Compute Engine
with OpenCL 1.2 capability.
Started OpenCL SGEMM tuner with batch size 16.
Will try 578 valid configurations.
(1/578) KWG=32 KWI=2 MDIMA=8 MDIMC=8 MWG=16 NDIMB=8 NDIMC=8 NWG=16 SA=0 SB=0 STRM=0 STRN=0 VWM=1 VWN=1 2234.1 us (240.3 GFLOPS)
(2/578) KWG=32 KWI=2 MDIMA=8 MDIMC=8 MWG=32 NDIMB=8 NDIMC=8 NWG=16 SA=0 SB=0 STRM=0 STRN=0 VWM=1 VWN=1 874.5 us (613.9 GFLOPS)
(4/578) KWG=32 KWI=2 MDIMA=8 MDIMC=8 MWG=16 NDIMB=8 NDIMC=8 NWG=32 SA=0 SB=0 STRM=0 STRN=0 VWM=1 VWN=1 867.5 us (618.9 GFLOPS)
(5/578) KWG=32 KWI=2 MDIMA=8 MDIMC=8 MWG=32 NDIMB=8 NDIMC=8 NWG=32 SA=0 SB=0 STRM=0 STRN=0 VWM=1 VWN=1 436.1 us (1231.0 GFLOPS)
(15/578) KWG=32 KWI=2 MDIMA=16 MDIMC=16 MWG=64 NDIMB=8 NDIMC=8 NWG=32 SA=0 SB=0 STRM=0 STRN=0 VWM=1 VWN=1 431.5 us (1244.2 GFLOPS)
(115/578) KWG=32 KWI=2 MDIMA=8 MDIMC=8 MWG=32 NDIMB=8 NDIMC=8 NWG=32 SA=0 SB=0 STRM=0 STRN=0 VWM=4 VWN=1 430.8 us (1246.1 GFLOPS)
(117/578) KWG=32 KWI=2 MDIMA=8 MDIMC=8 MWG=32 NDIMB=8 NDIMC=8 NWG=64 SA=0 SB=0 STRM=0 STRN=0 VWM=4 VWN=1 362.3 us (1482.0 GFLOPS)
(283/578) KWG=32 KWI=2 MDIMA=8 MDIMC=8 MWG=32 NDIMB=8 NDIMC=8 NWG=64 SA=0 SB=0 STRM=0 STRN=0 VWM=4 VWN=4 360.3 us (1489.9 GFLOPS)
(478/578) KWG=32 KWI=2 MDIMA=8 MDIMC=8 MWG=32 NDIMB=8 NDIMC=8 NWG=32 SA=1 SB=1 STRM=0 STRN=0 VWM=2 VWN=2 354.0 us (1516.5 GFLOPS)
Wavefront/Warp size: 64

Max workgroup size: 256
Max workgroup dimensions: 256 256 256
go nodes 3000
info depth 1 seldepth 2 time 514 nodes 4 score cp 17 hashfull 0 nps 7 tbhits 0 pv e2e4 e7e5
info depth 2 seldepth 3 time 591 nodes 8 score cp 10 hashfull 0 nps 13 tbhits 0 pv g1f3 d7d5 d2d4
info depth 3 seldepth 4 time 744 nodes 17 score cp 11 hashfull 0 nps 22 tbhits 0 pv g1f3 d7d5 d2d4 g8f6
info depth 3 seldepth 5 time 834 nodes 34 score cp 20 hashfull 0 nps 40 tbhits 0 pv g1f3 e7e6 d2d4 d7d5 f3g5
info depth 3 seldepth 6 time 1000 nodes 38 score cp 20 hashfull 0 nps 38 tbhits 0 pv g1f3 e7e6 d2d4 d7d5 f3g5
info depth 4 seldepth 6 time 1067 nodes 94 score cp -131 hashfull 1 nps 88 tbhits 0 pv e2e3 b7b6 e3e4 h7h6
info depth 4 seldepth 6 time 1388 nodes 162 score cp -5 hashfull 1 nps 116 tbhits 0 pv g1f3 e7e6 d2d4 d7d5 e2e3 h7h6
info depth 4 seldepth 7 time 1564 nodes 227 score cp -68 hashfull 2 nps 145 tbhits 0 pv g1f3 e7e6 d2d4 d7d5 e2e3 h7h6
info depth 4 seldepth 7 time 1847 nodes 328 score cp -75 hashfull 2 nps 177 tbhits 0 pv c2c3 e7e5 g2g3 a7a6 h2h3
info depth 4 seldepth 8 time 2069 nodes 362 score cp -75 hashfull 2 nps 174 tbhits 0 pv c2c3 e7e5 g2g3 a7a6 h2h3
info depth 4 seldepth 9 time 2515 nodes 462 score cp -75 hashfull 3 nps 183 tbhits 0 pv c2c3 e7e5 g2g3 a7a6 h2h3
info depth 4 seldepth 9 time 3168 nodes 644 score cp 50 hashfull 4 nps 203 tbhits 0 pv e2e4 e7e5 g1f3 b8c6 d1e2 h7h6 b1a3 a7a6 c2c3
info depth 4 seldepth 9 time 6636 nodes 1720 score cp 59 hashfull 11 nps 259 tbhits 0 pv a2a4 e7e6 b1c3 g8f6 a4a5
info depth 5 seldepth 9 time 7534 nodes 1976 score cp 35 hashfull 12 nps 262 tbhits 0 pv a2a4 g7g6 e2e4 a7a5 f1c4 b7b6 h2h3
info depth 5 seldepth 9 time 10873 nodes 2904 score cp -2147483648 hashfull 18 nps 267 tbhits 0 pv e2e4 a7a6 b2b3 e7e6 d1e2 h7h6 f2f3 h6h5
Segmentation fault: 11
My initial thought that there was something wrong with the NN you downloaded , so I downloaded the NN you used and tried it. There were no issues.

Code: Select all

Mac-Pro:release michaelbyrne$ lc0
       _
|   _ | |
|_ |_ |_| v0.22.0-dev built May 25 2019
ucinewgame
Found pb network file: ./weights_run1_42372.pb.gz
Creating backend [opencl]...
OpenCL, maximum batch size set to 16.
Initializing OpenCL.
Detected 1 OpenCL platforms.
Platform version: OpenCL 1.2 (Apr  7 2019 18:38:19)
Platform profile: FULL_PROFILE
Platform name:    Apple
Platform vendor:  Apple
Device ID:      0
Device name:    Intel(R) Xeon(R) CPU           X5690  @ 3.47GHz
Device type:    CPU
Device vendor:  Intel
Device driver:  1.1
Device speed:   3460 MHZ
Device cores:   24 CU
Device score:   512
Device ID:      1
Device name:    AMD Radeon HD 7950 Compute Engine
Device type:    GPU
Device vendor:  AMD
Device driver:  1.2 (Apr 29 2019 22:28:38)
Device speed:   150 MHZ
Device cores:   28 CU
Device score:   1112
Selected platform: Apple
Selected device: AMD Radeon HD 7950 Compute Engine
with OpenCL 1.2 capability.
Loaded existing SGEMM tuning for batch size 16.
Wavefront/Warp size: 64

Max workgroup size: 256
Max workgroup dimensions: 256 256 256
go nodes 3000
info depth 1 seldepth 2 time 611 nodes 4 score cp 17 hashfull 0 nps 6 tbhits 0 pv e2e4 e7e5
info depth 2 seldepth 3 time 722 nodes 8 score cp 10 hashfull 0 nps 11 tbhits 0 pv g1f3 d7d5 d2d4
info depth 3 seldepth 4 time 886 nodes 17 score cp 11 hashfull 0 nps 19 tbhits 0 pv g1f3 d7d5 d2d4 g8f6
info depth 3 seldepth 5 time 1050 nodes 25 score cp 9 hashfull 0 nps 23 tbhits 0 pv g1f3 d7d5 g2g3 c7c5 f1g2
info depth 4 seldepth 6 time 1203 nodes 63 score cp 10 hashfull 1 nps 52 tbhits 0 pv g1f3 d7d5 g2g3 c7c5 f1g2
info depth 4 seldepth 7 time 1374 nodes 82 score cp 10 hashfull 1 nps 59 tbhits 0 pv g1f3 d7d5 g2g3 e7e6 f1g2 c7c5
info depth 4 seldepth 8 time 1539 nodes 111 score cp 15 hashfull 1 nps 72 tbhits 0 pv e2e4 e7e5 g1f3 b8c6 b1c3 g8f6 f1b5 f8d6
info depth 4 seldepth 8 time 1641 nodes 132 score cp 11 hashfull 1 nps 80 tbhits 0 pv d2d4 g8f6 c2c4 e7e6 b1c3 f8b4 e2e3 e8g8 d4d5
info depth 5 seldepth 9 time 1706 nodes 146 score cp 16 hashfull 1 nps 85 tbhits 0 pv e2e4 e7e5 g1f3 b8c6 b1c3 g8f6 f1b5 f8b4 e1g1
info depth 5 seldepth 10 time 1872 nodes 205 score cp 16 hashfull 1 nps 109 tbhits 0 pv e2e4 e7e5 g1f3 b8c6 f1b5 g8f6 b1c3 f8d6 e1g1 e8g8
info depth 5 seldepth 11 time 2036 nodes 252 score cp 16 hashfull 2 nps 123 tbhits 0 pv e2e4 e7e5 g1f3 b8c6 b1c3 g8f6 f1b5 f8d6 e1g1 e8g8 b5a4
info depth 6 seldepth 11 time 2037 nodes 267 score cp 16 hashfull 2 nps 131 tbhits 0 pv e2e4 e7e5 g1f3 b8c6 b1c3 g8f6 f1b5 f8d6 e1g1 e8g8 b5a4
info depth 6 seldepth 12 time 2201 nodes 311 score cp 16 hashfull 2 nps 141 tbhits 0 pv e2e4 e7e5 b1c3 g8f6 g1f3 b8c6 f1b5 f8d6 e1g1 e8g8 b5a4 h7h6
info depth 6 seldepth 13 time 2343 nodes 398 score cp 16 hashfull 2 nps 169 tbhits 0 pv e2e4 e7e5 b1c3 g8f6 g1f3 b8c6 f1b5 f8d6 d2d3 e8g8 b5a4 f8e8 e1g1
info depth 6 seldepth 14 time 2506 nodes 463 score cp 17 hashfull 2 nps 184 tbhits 0 pv e2e4 e7e5 b1c3 g8f6 g1f3 b8c6 f1b5 f8d6 d2d3 e8g8 b5a4 f8e8 e1g1 h7h6
info depth 7 seldepth 15 time 2782 nodes 620 score cp 18 hashfull 3 nps 222 tbhits 0 pv e2e4 e7e5 b1c3 g8f6 g1f3 b8c6 f1b5 f8d6 d2d3 e8g8 b5a4 f8e8 e1g1 h7h6 a2a3
info depth 7 seldepth 16 time 3000 nodes 720 score cp 18 hashfull 3 nps 240 tbhits 0 pv e2e4 e7e5 b1c3 g8f6 g1f3 b8c6 f1b5 f8d6 d2d3 e8g8 b5a4 f8e8 e1g1 h7h6 a2a3 h6h5
info depth 7 seldepth 17 time 3436 nodes 912 score cp 18 hashfull 4 nps 265 tbhits 0 pv e2e4 e7e5 b1c3 g8f6 g1f3 b8c6 f1b5 f8d6 d2d3 e8g8 b5a4 f8e8 e1g1 h7h6 a2a3 a7a6 a4c6
info depth 7 seldepth 18 time 3605 nodes 1041 score cp 19 hashfull 4 nps 288 tbhits 0 pv e2e4 e7e5 b1c3 g8f6 g1f3 b8c6 f1b5 f8d6 d2d3 e8g8 b5a4 f8e8 e1g1 h7h6 a2a3 a7a6 h2h3
info depth 7 seldepth 19 time 3901 nodes 1089 score cp 19 hashfull 5 nps 279 tbhits 0 pv e2e4 e7e5 b1c3 g8f6 g1f3 b8c6 f1b5 f8d6 d2d3 e8g8 b5a4 f8e8 e1g1 h7h6 a2a3 a7a6 h2h3
info depth 7 seldepth 20 time 4232 nodes 1268 score cp 19 hashfull 5 nps 299 tbhits 0 pv e2e4 e7e5 b1c3 g8f6 g1f3 b8c6 f1b5 f8d6 d2d3 e8g8 b5a4 f8e8 e1g1 h7h6 a2a3 a7a6 h2h3
info depth 8 seldepth 20 time 4619 nodes 1619 score cp 19 hashfull 6 nps 350 tbhits 0 pv e2e4 e7e5 b1c3 g8f6 g1f3 b8c6 f1b5 f8d6 d2d3 e8g8 b5a4 f8e8 e1g1 h7h6 a2a3 a7a6 h2h3 d6f8
info depth 8 seldepth 21 time 5399 nodes 1921 score cp 19 hashfull 8 nps 355 tbhits 0 pv e2e4 e7e5 b1c3 g8f6 g1f3 b8c6 f1b5 f8d6 d2d3 e8g8 b5a4 f8e8 e1g1 h7h6 a2a3 a7a6 h2h3 d6f8 a4b3
info depth 8 seldepth 22 time 6792 nodes 2800 score cp 17 hashfull 10 nps 412 tbhits 0 pv e2e4 e7e5 b1c3 g8f6 g1f3 b8c6 f1b5 f8d6 d2d3 e8g8 b5a4 f8e8 e1g1 h7h6 a2a3 a7a6 h2h3 d6f8 a4b3
info depth 8 seldepth 22 time 7265 nodes 3052 score cp 16 hashfull 11 nps 420 tbhits 0 pv e2e4 e7e5 b1c3 g8f6 g1f3 b8c6 f1b5 f8d6 d2d3 e8g8 b5a4 f8e8 e1g1 h7h6 a2a3 a7a6 h2h3 d6f8 a4b3
bestmove e2e4 ponder e7e5
quit
SoI took this one step further, deleted the leelaz_opencl_tuning, and used the net you were using to rebuild the leelaz_opencl_tuning file. Bingo, I got the seg 11 error. I deleted the weight file you were using, and the leelaz_opencl_tuning file. I put my weights file back.

Code: Select all

Mac-Pro:release michaelbyrne$ lc0
       _
|   _ | |
|_ |_ |_| v0.22.0-dev built May 25 2019
ucinewgame
Found pb network file: ./weights
Creating backend [opencl]...
OpenCL, maximum batch size set to 16.
Initializing OpenCL.
Detected 1 OpenCL platforms.
Platform version: OpenCL 1.2 (Apr  7 2019 18:38:19)
Platform profile: FULL_PROFILE
Platform name:    Apple
Platform vendor:  Apple
Device ID:      0
Device name:    Intel(R) Xeon(R) CPU           X5690  @ 3.47GHz
Device type:    CPU
Device vendor:  Intel
Device driver:  1.1
Device speed:   3460 MHZ
Device cores:   24 CU
Device score:   512
Device ID:      1
Device name:    AMD Radeon HD 7950 Compute Engine
Device type:    GPU
Device vendor:  AMD
Device driver:  1.2 (Apr 29 2019 22:28:38)
Device speed:   150 MHZ
Device cores:   28 CU
Device score:   1112
Selected platform: Apple
Selected device: AMD Radeon HD 7950 Compute Engine
with OpenCL 1.2 capability.
Started OpenCL SGEMM tuner with batch size 16.
Will try 578 valid configurations.
(2/578) KWG=32 KWI=2 MDIMA=8 MDIMC=8 MWG=32 NDIMB=8 NDIMC=8 NWG=16 SA=0 SB=0 STRM=0 STRN=0 VWM=1 VWN=1 249.9 us (411.1 GFLOPS)
(5/578) KWG=32 KWI=2 MDIMA=8 MDIMC=8 MWG=32 NDIMB=8 NDIMC=8 NWG=32 SA=0 SB=0 STRM=0 STRN=0 VWM=1 VWN=1 143.7 us (715.2 GFLOPS)
(15/578) KWG=32 KWI=2 MDIMA=16 MDIMC=16 MWG=64 NDIMB=8 NDIMC=8 NWG=32 SA=0 SB=0 STRM=0 STRN=0 VWM=1 VWN=1 141.1 us (728.2 GFLOPS)
(42/578) KWG=32 KWI=2 MDIMA=16 MDIMC=16 MWG=64 NDIMB=16 NDIMC=16 NWG=64 SA=0 SB=0 STRM=0 STRN=0 VWM=1 VWN=1 138.5 us (741.8 GFLOPS)
(115/578) KWG=32 KWI=2 MDIMA=8 MDIMC=8 MWG=32 NDIMB=8 NDIMC=8 NWG=32 SA=0 SB=0 STRM=0 STRN=0 VWM=4 VWN=1 126.5 us (812.1 GFLOPS)
(117/578) KWG=32 KWI=2 MDIMA=8 MDIMC=8 MWG=32 NDIMB=8 NDIMC=8 NWG=64 SA=0 SB=0 STRM=0 STRN=0 VWM=4 VWN=1 122.1 us (841.6 GFLOPS)
(121/578) KWG=32 KWI=2 MDIMA=16 MDIMC=16 MWG=64 NDIMB=8 NDIMC=8 NWG=64 SA=0 SB=0 STRM=0 STRN=0 VWM=4 VWN=1 121.2 us (847.8 GFLOPS)
(376/578) KWG=32 KWI=2 MDIMA=8 MDIMC=8 MWG=32 NDIMB=16 NDIMC=16 NWG=32 SA=1 SB=1 STRM=0 STRN=0 VWM=2 VWN=1 116.3 us (883.6 GFLOPS)
(485/578) KWG=32 KWI=2 MDIMA=16 MDIMC=16 MWG=32 NDIMB=8 NDIMC=8 NWG=32 SA=1 SB=1 STRM=0 STRN=0 VWM=2 VWN=2 111.1 us (925.2 GFLOPS)
Wavefront/Warp size: 64

Max workgroup size: 256
Max workgroup dimensions: 256 256 256
go nodes 3000     
info depth 1 seldepth 2 time 190 nodes 7 score cp 8 hashfull 0 nps 36 tbhits 0 pv d2d4 g8f6
info depth 2 seldepth 3 time 226 nodes 16 score cp 5 hashfull 0 nps 70 tbhits 0 pv g2g3 d7d5 g1f3
info depth 3 seldepth 4 time 262 nodes 35 score cp 6 hashfull 0 nps 133 tbhits 0 pv g2g3 e7e6 c2c4 d7d5
info depth 3 seldepth 5 time 302 nodes 57 score cp 5 hashfull 0 nps 188 tbhits 0 pv g2g3 e7e6 c2c4 d7d5 f1g2
info depth 3 seldepth 5 time 305 nodes 59 score cp 5 hashfull 0 nps 193 tbhits 0 pv c2c4 e7e5 d2d3 b8c6 g1f3
info depth 4 seldepth 5 time 324 nodes 67 score cp 5 hashfull 0 nps 206 tbhits 0 pv g2g3 d7d5 g1f3 c7c5 f1g2 d5d4
info depth 4 seldepth 6 time 350 nodes 92 score cp 6 hashfull 0 nps 262 tbhits 0 pv g2g3 e7e6 c2c4 d7d5 f1g2 d5d4
info depth 4 seldepth 7 time 395 nodes 128 score cp 6 hashfull 1 nps 324 tbhits 0 pv c2c4 e7e5 d2d3 f8e7 g1f3 b8c6
info depth 4 seldepth 8 time 445 nodes 179 score cp 6 hashfull 1 nps 402 tbhits 0 pv c2c4 e7e5 d2d3 f8e7 g1f3 b8c6
info depth 4 seldepth 8 time 494 nodes 235 score cp 7 hashfull 1 nps 475 tbhits 0 pv g1f3 d7d5 d2d4 g8f6 c1f4 c8f5 e2e3 e7e6
info depth 4 seldepth 9 time 520 nodes 260 score cp 7 hashfull 1 nps 500 tbhits 0 pv g1f3 d7d5 d2d4 g8f6 c1f4 c8f5 e2e3 e7e6
info depth 5 seldepth 9 time 568 nodes 306 score cp 6 hashfull 2 nps 538 tbhits 0 pv g1f3 d7d5 d2d4 g8f6 c1f4 c8f5 c2c4 e7e6 b1c3
info depth 5 seldepth 10 time 640 nodes 415 score cp 6 hashfull 2 nps 648 tbhits 0 pv g1f3 d7d5 d2d4 g8f6 c1f4 c7c5 e2e3 c5d4 e3d4
info depth 5 seldepth 11 time 763 nodes 537 score cp 7 hashfull 3 nps 703 tbhits 0 pv g1f3 d7d5 d2d4 g8f6 c1f4 c7c5 e2e3 c5d4 e3d4
info depth 5 seldepth 12 time 873 nodes 747 score cp 7 hashfull 3 nps 855 tbhits 0 pv g1f3 d7d5 d2d4 g8f6 c1f4 c7c5 e2e3 c5d4 e3d4
info depth 5 seldepth 13 time 1021 nodes 954 score cp 8 hashfull 4 nps 934 tbhits 0 pv g1f3 d7d5 d2d4 g8f6 c1f4 c7c5 e2e3 c5d4 e3d4
info depth 6 seldepth 13 time 1154 nodes 1143 score cp 8 hashfull 5 nps 990 tbhits 0 pv g1f3 d7d5 d2d4 g8f6 c1f4 c7c5 e2e3 c5d4 e3d4 b8c6
info depth 6 seldepth 14 time 1390 nodes 1483 score cp 9 hashfull 6 nps 1066 tbhits 0 pv g1f3 d7d5 d2d4 g8f6 c1f4 c7c5 e2e3 c5d4 e3d4 b8c6
info depth 6 seldepth 14 time 1820 nodes 2205 score cp 13 hashfull 9 nps 1211 tbhits 0 pv d2d4 g8f6 c2c4 e7e6 g2g3 f8b4 b1d2 d7d5 f1g2 e8g8 g1f3
info depth 6 seldepth 14 time 1919 nodes 2291 score cp 9 hashfull 9 nps 1193 tbhits 0 pv g1f3 d7d5 d2d4 g8f6 c1f4 c7c5 e2e3 c5d4 e3d4 b8c6 c2c3
info depth 6 seldepth 14 time 1943 nodes 2402 score cp 13 hashfull 10 nps 1236 tbhits 0 pv d2d4 g8f6 c2c4 e7e6 g2g3 f8b4 b1d2 d7d5 f1g2 e8g8 g1f3
info depth 6 seldepth 15 time 2265 nodes 2951 score cp 12 hashfull 12 nps 1302 tbhits 0 pv d2d4 g8f6 c2c4 e7e6 g2g3 f8b4 b1d2 d7d5 f1g2 e8g8 g1f3 d5c4
info depth 6 seldepth 15 time 2347 nodes 3003 score cp 12 hashfull 12 nps 1279 tbhits 0 pv d2d4 g8f6 c2c4 e7e6 g2g3 f8b4 b1d2 d7d5 f1g2 e8g8 g1f3 d5c4 d4d5
bestmove d2d4 ponder g8f6
Everything works - you still use the weights file if you want, but you can't tune Lc0 with it, it's creating a bad tuning file. So move your weights file to a different folder and place this one there instead.

https://www.dropbox.com/s/6iy7cqeu89ijl34/weights?dl=1
I'm not even sure which it was at this point but it work great ,and you will get much better NPS.

Delete your current leelaz_opencl_tuning file and follow the steps you did before when you first ran LcO.

Also, dkappe has the best nets for those with less than high-end hardware - your nps will increase by 10X easily or more and you will be much more satisfied with lc0.

https://github.com/dkappe/leela-chess-w ... d-Networks

Re: Lc0 - macOS binary requested

Posted: Sun May 26, 2019 3:25 am
by MikeB
Steppenwolf wrote: Sat May 25, 2019 9:01 am Hi MikeB,

great thanks!!!
It seems to work, see my screen shot.
Where can I see the kn/seconds?
They were not displayed in the Output like under SF.
They are missing in HIARCS Explorer - not sure why. I tried to change a few settings, but nothing worked.

You might want to let them know.

http://www.hiarcs.net/forums/viewforum.php?f=2

Re: Lc0 - macOS binary requested

Posted: Sun May 26, 2019 5:19 am
by smcracraft
MikeB wrote: Sun May 26, 2019 2:48 am
smcracraft wrote: Sat May 25, 2019 11:38 pm Also it crashes on the mac.

Stuarts-MacBook-Pro:release stuartcracraft$ ./lc0
_
| _ | |
|_ |_ |_| v0.22.0-dev built May 25 2019
ucinewgame
Found pb network file: ./weights_run1_42372.pb.gz
Creating backend [opencl]...
OpenCL, maximum batch size set to 16.
Initializing OpenCL.
Detected 1 OpenCL platforms.
Platform version: OpenCL 1.2 (Feb 22 2019 20:16:07)
Platform profile: FULL_PROFILE
Platform name: Apple
Platform vendor: Apple
Device ID: 0
Device name: Intel(R) Core(TM) i7-7920HQ CPU @ 3.10GHz
Device type: CPU
Device vendor: Intel
Device driver: 1.1
Device speed: 3100 MHZ
Device cores: 8 CU
Device score: 512
Device ID: 1
Device name: Intel(R) HD Graphics 630
Device type: GPU
Device vendor: Intel Inc.
Device driver: 1.2(Mar 11 2019 21:25:40)
Device speed: 1100 MHZ
Device cores: 24 CU
Device score: 612
Device ID: 2
Device name: AMD Radeon Pro 560 Compute Engine
Device type: GPU
Device vendor: AMD
Device driver: 1.2 (Mar 11 2019 21:14:43)
Device speed: 300 MHZ
Device cores: 16 CU
Device score: 1112
Selected platform: Apple
Selected device: AMD Radeon Pro 560 Compute Engine
with OpenCL 1.2 capability.
Loaded existing SGEMM tuning for batch size 16.
Wavefront/Warp size: 64

Max workgroup size: 256
Max workgroup dimensions: 256 256 256
go nodes 3000
info depth 1 seldepth 2 time 209 nodes 4 score cp 17 hashfull 0 nps 19 tbhits 0 pv e2e4 e7e5
info depth 2 seldepth 3 time 361 nodes 8 score cp 10 hashfull 0 nps 22 tbhits 0 pv g1f3 d7d5 d2d4
info depth 2 seldepth 4 time 705 nodes 10 score cp -71 hashfull 0 nps 14 tbhits 0 pv g1f3 e7e6 c2c4 f8e7
info depth 3 seldepth 4 time 706 nodes 19 score cp -201 hashfull 0 nps 26 tbhits 0 pv g1f3 e7e6 d2d4 h7h6
info depth 3 seldepth 5 time 1039 nodes 34 score cp 130 hashfull 0 nps 32 tbhits 0 pv e2e4 e7e5 b1c3 a7a6 c3d5
info depth 3 seldepth 5 time 1524 nodes 86 score cp -4 hashfull 1 nps 56 tbhits 0 pv e2e3 g8f6 g1f3 h8g8
info depth 3 seldepth 6 time 1879 nodes 127 score cp -71 hashfull 1 nps 67 tbhits 0 pv e2e3 g8f6 g1f3 h8g8
info depth 4 seldepth 7 time 2673 nodes 236 score cp 26 hashfull 2 nps 88 tbhits 0 pv e2e4 c7c6 c2c4 d8c7 g2g3 c6c5
info depth 4 seldepth 7 time 4609 nodes 566 score cp 13 hashfull 4 nps 122 tbhits 0 pv b1c3 d7d5 a1b1 f7f5 a2a3 h7h6
info depth 4 seldepth 7 time 6259 nodes 753 score cp 124 hashfull 4 nps 120 tbhits 0 pv f2f4 a7a5 h2h4 c7c5 a2a3
info depth 5 seldepth 8 time 10000 nodes 1470 score cp 59 hashfull 8 nps 147 tbhits 0 pv f2f4 e7e6 c2c3 h7h6 a2a3 a7a6 h2h3
Segmentation fault: 11
Stuarts-MacBook-Pro:release stuartcracraft$
I was able to duplicate your error:

Code: Select all

Mac-Pro:release michaelbyrne$ lc0
       _
|   _ | |
|_ |_ |_| v0.22.0-dev built May 25 2019

uci
id name Lc0 v0.22.0-dev
id author The LCZero Authors.
option name WeightsFile type string default <autodiscover>
option name Backend type combo default opencl var opencl var blas var check var random var roundrobin var multiplexing var demux
option name BackendOptions type string default 
option name Threads type spin default 2 min 1 max 128
option name NNCacheSize type spin default 200000 min 0 max 999999999
option name MinibatchSize type spin default 256 min 1 max 1024
option name MaxPrefetch type spin default 32 min 0 max 1024
option name CPuct type string default 3.000000
option name CPuctBase type string default 19652.000000
option name CPuctFactor type string default 2.000000
option name Temperature type string default 0.000000
option name TempDecayMoves type spin default 0 min 0 max 100
option name TempCutoffMove type spin default 0 min 0 max 1000
option name TempEndgame type string default 0.000000
option name TempValueCutoff type string default 100.000000
option name TempVisitOffset type string default 0.000000
option name DirichletNoise type check default false
option name VerboseMoveStats type check default false
option name SmartPruningFactor type string default 1.330000
option name FpuStrategy type combo default reduction var reduction var absolute
option name FpuValue type string default 1.200000
option name FpuStrategyAtRoot type combo default same var reduction var absolute var same
option name FpuValueAtRoot type string default 1.000000
option name CacheHistoryLength type spin default 0 min 0 max 7
option name PolicyTemperature type string default 2.200000
option name MaxCollisionEvents type spin default 32 min 1 max 1024
option name MaxCollisionVisits type spin default 9999 min 1 max 1000000
option name OutOfOrderEval type check default true
option name StickyEndgames type check default true
option name SyzygyFastPlay type check default true
option name MultiPV type spin default 1 min 1 max 500
option name ScoreType type combo default centipawn var centipawn var centipawn_2018 var win_percentage var Q
option name HistoryFill type combo default fen_only var no var fen_only var always
option name KLDGainAverageInterval type spin default 100 min 1 max 10000000
option name MinimumKLDGainPerNode type string default 0.000000
option name Slowmover type string default 1.000000
option name MoveOverheadMs type spin default 200 min 0 max 100000000
option name SyzygyPath type string default 
option name Ponder type check default true
option name ImmediateTimeUse type string default 1.000000
option name RamLimitMb type spin default 0 min 0 max 100000000
option name ConfigFile type string default lc0.config
option name LogFile type string default 
uciok
ucinewgame
Found pb network file: ./weights_run1_42372.pb.gz
Creating backend [opencl]...
OpenCL, maximum batch size set to 16.
Initializing OpenCL.
Detected 1 OpenCL platforms.
Platform version: OpenCL 1.2 (Apr  7 2019 18:38:19)
Platform profile: FULL_PROFILE
Platform name:    Apple
Platform vendor:  Apple
Device ID:      0
Device name:    Intel(R) Xeon(R) CPU           X5690  @ 3.47GHz
Device type:    CPU
Device vendor:  Intel
Device driver:  1.1
Device speed:   3460 MHZ
Device cores:   24 CU
Device score:   512
Device ID:      1
Device name:    AMD Radeon HD 7950 Compute Engine
Device type:    GPU
Device vendor:  AMD
Device driver:  1.2 (Apr 29 2019 22:28:38)
Device speed:   150 MHZ
Device cores:   28 CU
Device score:   1112
Selected platform: Apple
Selected device: AMD Radeon HD 7950 Compute Engine
with OpenCL 1.2 capability.
Started OpenCL SGEMM tuner with batch size 16.
Will try 578 valid configurations.
(1/578) KWG=32 KWI=2 MDIMA=8 MDIMC=8 MWG=16 NDIMB=8 NDIMC=8 NWG=16 SA=0 SB=0 STRM=0 STRN=0 VWM=1 VWN=1 2234.1 us (240.3 GFLOPS)
(2/578) KWG=32 KWI=2 MDIMA=8 MDIMC=8 MWG=32 NDIMB=8 NDIMC=8 NWG=16 SA=0 SB=0 STRM=0 STRN=0 VWM=1 VWN=1 874.5 us (613.9 GFLOPS)
(4/578) KWG=32 KWI=2 MDIMA=8 MDIMC=8 MWG=16 NDIMB=8 NDIMC=8 NWG=32 SA=0 SB=0 STRM=0 STRN=0 VWM=1 VWN=1 867.5 us (618.9 GFLOPS)
(5/578) KWG=32 KWI=2 MDIMA=8 MDIMC=8 MWG=32 NDIMB=8 NDIMC=8 NWG=32 SA=0 SB=0 STRM=0 STRN=0 VWM=1 VWN=1 436.1 us (1231.0 GFLOPS)
(15/578) KWG=32 KWI=2 MDIMA=16 MDIMC=16 MWG=64 NDIMB=8 NDIMC=8 NWG=32 SA=0 SB=0 STRM=0 STRN=0 VWM=1 VWN=1 431.5 us (1244.2 GFLOPS)
(115/578) KWG=32 KWI=2 MDIMA=8 MDIMC=8 MWG=32 NDIMB=8 NDIMC=8 NWG=32 SA=0 SB=0 STRM=0 STRN=0 VWM=4 VWN=1 430.8 us (1246.1 GFLOPS)
(117/578) KWG=32 KWI=2 MDIMA=8 MDIMC=8 MWG=32 NDIMB=8 NDIMC=8 NWG=64 SA=0 SB=0 STRM=0 STRN=0 VWM=4 VWN=1 362.3 us (1482.0 GFLOPS)
(283/578) KWG=32 KWI=2 MDIMA=8 MDIMC=8 MWG=32 NDIMB=8 NDIMC=8 NWG=64 SA=0 SB=0 STRM=0 STRN=0 VWM=4 VWN=4 360.3 us (1489.9 GFLOPS)
(478/578) KWG=32 KWI=2 MDIMA=8 MDIMC=8 MWG=32 NDIMB=8 NDIMC=8 NWG=32 SA=1 SB=1 STRM=0 STRN=0 VWM=2 VWN=2 354.0 us (1516.5 GFLOPS)
Wavefront/Warp size: 64

Max workgroup size: 256
Max workgroup dimensions: 256 256 256
go nodes 3000
info depth 1 seldepth 2 time 514 nodes 4 score cp 17 hashfull 0 nps 7 tbhits 0 pv e2e4 e7e5
info depth 2 seldepth 3 time 591 nodes 8 score cp 10 hashfull 0 nps 13 tbhits 0 pv g1f3 d7d5 d2d4
info depth 3 seldepth 4 time 744 nodes 17 score cp 11 hashfull 0 nps 22 tbhits 0 pv g1f3 d7d5 d2d4 g8f6
info depth 3 seldepth 5 time 834 nodes 34 score cp 20 hashfull 0 nps 40 tbhits 0 pv g1f3 e7e6 d2d4 d7d5 f3g5
info depth 3 seldepth 6 time 1000 nodes 38 score cp 20 hashfull 0 nps 38 tbhits 0 pv g1f3 e7e6 d2d4 d7d5 f3g5
info depth 4 seldepth 6 time 1067 nodes 94 score cp -131 hashfull 1 nps 88 tbhits 0 pv e2e3 b7b6 e3e4 h7h6
info depth 4 seldepth 6 time 1388 nodes 162 score cp -5 hashfull 1 nps 116 tbhits 0 pv g1f3 e7e6 d2d4 d7d5 e2e3 h7h6
info depth 4 seldepth 7 time 1564 nodes 227 score cp -68 hashfull 2 nps 145 tbhits 0 pv g1f3 e7e6 d2d4 d7d5 e2e3 h7h6
info depth 4 seldepth 7 time 1847 nodes 328 score cp -75 hashfull 2 nps 177 tbhits 0 pv c2c3 e7e5 g2g3 a7a6 h2h3
info depth 4 seldepth 8 time 2069 nodes 362 score cp -75 hashfull 2 nps 174 tbhits 0 pv c2c3 e7e5 g2g3 a7a6 h2h3
info depth 4 seldepth 9 time 2515 nodes 462 score cp -75 hashfull 3 nps 183 tbhits 0 pv c2c3 e7e5 g2g3 a7a6 h2h3
info depth 4 seldepth 9 time 3168 nodes 644 score cp 50 hashfull 4 nps 203 tbhits 0 pv e2e4 e7e5 g1f3 b8c6 d1e2 h7h6 b1a3 a7a6 c2c3
info depth 4 seldepth 9 time 6636 nodes 1720 score cp 59 hashfull 11 nps 259 tbhits 0 pv a2a4 e7e6 b1c3 g8f6 a4a5
info depth 5 seldepth 9 time 7534 nodes 1976 score cp 35 hashfull 12 nps 262 tbhits 0 pv a2a4 g7g6 e2e4 a7a5 f1c4 b7b6 h2h3
info depth 5 seldepth 9 time 10873 nodes 2904 score cp -2147483648 hashfull 18 nps 267 tbhits 0 pv e2e4 a7a6 b2b3 e7e6 d1e2 h7h6 f2f3 h6h5
Segmentation fault: 11
My initial thought that there was something wrong with the NN you downloaded , so I downloaded the NN you used and tried it. There were no issues.

Code: Select all

Mac-Pro:release michaelbyrne$ lc0
       _
|   _ | |
|_ |_ |_| v0.22.0-dev built May 25 2019
ucinewgame
Found pb network file: ./weights_run1_42372.pb.gz
Creating backend [opencl]...
OpenCL, maximum batch size set to 16.
Initializing OpenCL.
Detected 1 OpenCL platforms.
Platform version: OpenCL 1.2 (Apr  7 2019 18:38:19)
Platform profile: FULL_PROFILE
Platform name:    Apple
Platform vendor:  Apple
Device ID:      0
Device name:    Intel(R) Xeon(R) CPU           X5690  @ 3.47GHz
Device type:    CPU
Device vendor:  Intel
Device driver:  1.1
Device speed:   3460 MHZ
Device cores:   24 CU
Device score:   512
Device ID:      1
Device name:    AMD Radeon HD 7950 Compute Engine
Device type:    GPU
Device vendor:  AMD
Device driver:  1.2 (Apr 29 2019 22:28:38)
Device speed:   150 MHZ
Device cores:   28 CU
Device score:   1112
Selected platform: Apple
Selected device: AMD Radeon HD 7950 Compute Engine
with OpenCL 1.2 capability.
Loaded existing SGEMM tuning for batch size 16.
Wavefront/Warp size: 64

Max workgroup size: 256
Max workgroup dimensions: 256 256 256
go nodes 3000
info depth 1 seldepth 2 time 611 nodes 4 score cp 17 hashfull 0 nps 6 tbhits 0 pv e2e4 e7e5
info depth 2 seldepth 3 time 722 nodes 8 score cp 10 hashfull 0 nps 11 tbhits 0 pv g1f3 d7d5 d2d4
info depth 3 seldepth 4 time 886 nodes 17 score cp 11 hashfull 0 nps 19 tbhits 0 pv g1f3 d7d5 d2d4 g8f6
info depth 3 seldepth 5 time 1050 nodes 25 score cp 9 hashfull 0 nps 23 tbhits 0 pv g1f3 d7d5 g2g3 c7c5 f1g2
info depth 4 seldepth 6 time 1203 nodes 63 score cp 10 hashfull 1 nps 52 tbhits 0 pv g1f3 d7d5 g2g3 c7c5 f1g2
info depth 4 seldepth 7 time 1374 nodes 82 score cp 10 hashfull 1 nps 59 tbhits 0 pv g1f3 d7d5 g2g3 e7e6 f1g2 c7c5
info depth 4 seldepth 8 time 1539 nodes 111 score cp 15 hashfull 1 nps 72 tbhits 0 pv e2e4 e7e5 g1f3 b8c6 b1c3 g8f6 f1b5 f8d6
info depth 4 seldepth 8 time 1641 nodes 132 score cp 11 hashfull 1 nps 80 tbhits 0 pv d2d4 g8f6 c2c4 e7e6 b1c3 f8b4 e2e3 e8g8 d4d5
info depth 5 seldepth 9 time 1706 nodes 146 score cp 16 hashfull 1 nps 85 tbhits 0 pv e2e4 e7e5 g1f3 b8c6 b1c3 g8f6 f1b5 f8b4 e1g1
info depth 5 seldepth 10 time 1872 nodes 205 score cp 16 hashfull 1 nps 109 tbhits 0 pv e2e4 e7e5 g1f3 b8c6 f1b5 g8f6 b1c3 f8d6 e1g1 e8g8
info depth 5 seldepth 11 time 2036 nodes 252 score cp 16 hashfull 2 nps 123 tbhits 0 pv e2e4 e7e5 g1f3 b8c6 b1c3 g8f6 f1b5 f8d6 e1g1 e8g8 b5a4
info depth 6 seldepth 11 time 2037 nodes 267 score cp 16 hashfull 2 nps 131 tbhits 0 pv e2e4 e7e5 g1f3 b8c6 b1c3 g8f6 f1b5 f8d6 e1g1 e8g8 b5a4
info depth 6 seldepth 12 time 2201 nodes 311 score cp 16 hashfull 2 nps 141 tbhits 0 pv e2e4 e7e5 b1c3 g8f6 g1f3 b8c6 f1b5 f8d6 e1g1 e8g8 b5a4 h7h6
info depth 6 seldepth 13 time 2343 nodes 398 score cp 16 hashfull 2 nps 169 tbhits 0 pv e2e4 e7e5 b1c3 g8f6 g1f3 b8c6 f1b5 f8d6 d2d3 e8g8 b5a4 f8e8 e1g1
info depth 6 seldepth 14 time 2506 nodes 463 score cp 17 hashfull 2 nps 184 tbhits 0 pv e2e4 e7e5 b1c3 g8f6 g1f3 b8c6 f1b5 f8d6 d2d3 e8g8 b5a4 f8e8 e1g1 h7h6
info depth 7 seldepth 15 time 2782 nodes 620 score cp 18 hashfull 3 nps 222 tbhits 0 pv e2e4 e7e5 b1c3 g8f6 g1f3 b8c6 f1b5 f8d6 d2d3 e8g8 b5a4 f8e8 e1g1 h7h6 a2a3
info depth 7 seldepth 16 time 3000 nodes 720 score cp 18 hashfull 3 nps 240 tbhits 0 pv e2e4 e7e5 b1c3 g8f6 g1f3 b8c6 f1b5 f8d6 d2d3 e8g8 b5a4 f8e8 e1g1 h7h6 a2a3 h6h5
info depth 7 seldepth 17 time 3436 nodes 912 score cp 18 hashfull 4 nps 265 tbhits 0 pv e2e4 e7e5 b1c3 g8f6 g1f3 b8c6 f1b5 f8d6 d2d3 e8g8 b5a4 f8e8 e1g1 h7h6 a2a3 a7a6 a4c6
info depth 7 seldepth 18 time 3605 nodes 1041 score cp 19 hashfull 4 nps 288 tbhits 0 pv e2e4 e7e5 b1c3 g8f6 g1f3 b8c6 f1b5 f8d6 d2d3 e8g8 b5a4 f8e8 e1g1 h7h6 a2a3 a7a6 h2h3
info depth 7 seldepth 19 time 3901 nodes 1089 score cp 19 hashfull 5 nps 279 tbhits 0 pv e2e4 e7e5 b1c3 g8f6 g1f3 b8c6 f1b5 f8d6 d2d3 e8g8 b5a4 f8e8 e1g1 h7h6 a2a3 a7a6 h2h3
info depth 7 seldepth 20 time 4232 nodes 1268 score cp 19 hashfull 5 nps 299 tbhits 0 pv e2e4 e7e5 b1c3 g8f6 g1f3 b8c6 f1b5 f8d6 d2d3 e8g8 b5a4 f8e8 e1g1 h7h6 a2a3 a7a6 h2h3
info depth 8 seldepth 20 time 4619 nodes 1619 score cp 19 hashfull 6 nps 350 tbhits 0 pv e2e4 e7e5 b1c3 g8f6 g1f3 b8c6 f1b5 f8d6 d2d3 e8g8 b5a4 f8e8 e1g1 h7h6 a2a3 a7a6 h2h3 d6f8
info depth 8 seldepth 21 time 5399 nodes 1921 score cp 19 hashfull 8 nps 355 tbhits 0 pv e2e4 e7e5 b1c3 g8f6 g1f3 b8c6 f1b5 f8d6 d2d3 e8g8 b5a4 f8e8 e1g1 h7h6 a2a3 a7a6 h2h3 d6f8 a4b3
info depth 8 seldepth 22 time 6792 nodes 2800 score cp 17 hashfull 10 nps 412 tbhits 0 pv e2e4 e7e5 b1c3 g8f6 g1f3 b8c6 f1b5 f8d6 d2d3 e8g8 b5a4 f8e8 e1g1 h7h6 a2a3 a7a6 h2h3 d6f8 a4b3
info depth 8 seldepth 22 time 7265 nodes 3052 score cp 16 hashfull 11 nps 420 tbhits 0 pv e2e4 e7e5 b1c3 g8f6 g1f3 b8c6 f1b5 f8d6 d2d3 e8g8 b5a4 f8e8 e1g1 h7h6 a2a3 a7a6 h2h3 d6f8 a4b3
bestmove e2e4 ponder e7e5
quit
SoI took this one step further, deleted the leelaz_opencl_tuning, and used the net you were using to rebuild the leelaz_opencl_tuning file. Bingo, I got the seg 11 error. I deleted the weight file you were using, and the leelaz_opencl_tuning file. I put my weights file back.

Code: Select all

Mac-Pro:release michaelbyrne$ lc0
       _
|   _ | |
|_ |_ |_| v0.22.0-dev built May 25 2019
ucinewgame
Found pb network file: ./weights
Creating backend [opencl]...
OpenCL, maximum batch size set to 16.
Initializing OpenCL.
Detected 1 OpenCL platforms.
Platform version: OpenCL 1.2 (Apr  7 2019 18:38:19)
Platform profile: FULL_PROFILE
Platform name:    Apple
Platform vendor:  Apple
Device ID:      0
Device name:    Intel(R) Xeon(R) CPU           X5690  @ 3.47GHz
Device type:    CPU
Device vendor:  Intel
Device driver:  1.1
Device speed:   3460 MHZ
Device cores:   24 CU
Device score:   512
Device ID:      1
Device name:    AMD Radeon HD 7950 Compute Engine
Device type:    GPU
Device vendor:  AMD
Device driver:  1.2 (Apr 29 2019 22:28:38)
Device speed:   150 MHZ
Device cores:   28 CU
Device score:   1112
Selected platform: Apple
Selected device: AMD Radeon HD 7950 Compute Engine
with OpenCL 1.2 capability.
Started OpenCL SGEMM tuner with batch size 16.
Will try 578 valid configurations.
(2/578) KWG=32 KWI=2 MDIMA=8 MDIMC=8 MWG=32 NDIMB=8 NDIMC=8 NWG=16 SA=0 SB=0 STRM=0 STRN=0 VWM=1 VWN=1 249.9 us (411.1 GFLOPS)
(5/578) KWG=32 KWI=2 MDIMA=8 MDIMC=8 MWG=32 NDIMB=8 NDIMC=8 NWG=32 SA=0 SB=0 STRM=0 STRN=0 VWM=1 VWN=1 143.7 us (715.2 GFLOPS)
(15/578) KWG=32 KWI=2 MDIMA=16 MDIMC=16 MWG=64 NDIMB=8 NDIMC=8 NWG=32 SA=0 SB=0 STRM=0 STRN=0 VWM=1 VWN=1 141.1 us (728.2 GFLOPS)
(42/578) KWG=32 KWI=2 MDIMA=16 MDIMC=16 MWG=64 NDIMB=16 NDIMC=16 NWG=64 SA=0 SB=0 STRM=0 STRN=0 VWM=1 VWN=1 138.5 us (741.8 GFLOPS)
(115/578) KWG=32 KWI=2 MDIMA=8 MDIMC=8 MWG=32 NDIMB=8 NDIMC=8 NWG=32 SA=0 SB=0 STRM=0 STRN=0 VWM=4 VWN=1 126.5 us (812.1 GFLOPS)
(117/578) KWG=32 KWI=2 MDIMA=8 MDIMC=8 MWG=32 NDIMB=8 NDIMC=8 NWG=64 SA=0 SB=0 STRM=0 STRN=0 VWM=4 VWN=1 122.1 us (841.6 GFLOPS)
(121/578) KWG=32 KWI=2 MDIMA=16 MDIMC=16 MWG=64 NDIMB=8 NDIMC=8 NWG=64 SA=0 SB=0 STRM=0 STRN=0 VWM=4 VWN=1 121.2 us (847.8 GFLOPS)
(376/578) KWG=32 KWI=2 MDIMA=8 MDIMC=8 MWG=32 NDIMB=16 NDIMC=16 NWG=32 SA=1 SB=1 STRM=0 STRN=0 VWM=2 VWN=1 116.3 us (883.6 GFLOPS)
(485/578) KWG=32 KWI=2 MDIMA=16 MDIMC=16 MWG=32 NDIMB=8 NDIMC=8 NWG=32 SA=1 SB=1 STRM=0 STRN=0 VWM=2 VWN=2 111.1 us (925.2 GFLOPS)
Wavefront/Warp size: 64

Max workgroup size: 256
Max workgroup dimensions: 256 256 256
go nodes 3000     
info depth 1 seldepth 2 time 190 nodes 7 score cp 8 hashfull 0 nps 36 tbhits 0 pv d2d4 g8f6
info depth 2 seldepth 3 time 226 nodes 16 score cp 5 hashfull 0 nps 70 tbhits 0 pv g2g3 d7d5 g1f3
info depth 3 seldepth 4 time 262 nodes 35 score cp 6 hashfull 0 nps 133 tbhits 0 pv g2g3 e7e6 c2c4 d7d5
info depth 3 seldepth 5 time 302 nodes 57 score cp 5 hashfull 0 nps 188 tbhits 0 pv g2g3 e7e6 c2c4 d7d5 f1g2
info depth 3 seldepth 5 time 305 nodes 59 score cp 5 hashfull 0 nps 193 tbhits 0 pv c2c4 e7e5 d2d3 b8c6 g1f3
info depth 4 seldepth 5 time 324 nodes 67 score cp 5 hashfull 0 nps 206 tbhits 0 pv g2g3 d7d5 g1f3 c7c5 f1g2 d5d4
info depth 4 seldepth 6 time 350 nodes 92 score cp 6 hashfull 0 nps 262 tbhits 0 pv g2g3 e7e6 c2c4 d7d5 f1g2 d5d4
info depth 4 seldepth 7 time 395 nodes 128 score cp 6 hashfull 1 nps 324 tbhits 0 pv c2c4 e7e5 d2d3 f8e7 g1f3 b8c6
info depth 4 seldepth 8 time 445 nodes 179 score cp 6 hashfull 1 nps 402 tbhits 0 pv c2c4 e7e5 d2d3 f8e7 g1f3 b8c6
info depth 4 seldepth 8 time 494 nodes 235 score cp 7 hashfull 1 nps 475 tbhits 0 pv g1f3 d7d5 d2d4 g8f6 c1f4 c8f5 e2e3 e7e6
info depth 4 seldepth 9 time 520 nodes 260 score cp 7 hashfull 1 nps 500 tbhits 0 pv g1f3 d7d5 d2d4 g8f6 c1f4 c8f5 e2e3 e7e6
info depth 5 seldepth 9 time 568 nodes 306 score cp 6 hashfull 2 nps 538 tbhits 0 pv g1f3 d7d5 d2d4 g8f6 c1f4 c8f5 c2c4 e7e6 b1c3
info depth 5 seldepth 10 time 640 nodes 415 score cp 6 hashfull 2 nps 648 tbhits 0 pv g1f3 d7d5 d2d4 g8f6 c1f4 c7c5 e2e3 c5d4 e3d4
info depth 5 seldepth 11 time 763 nodes 537 score cp 7 hashfull 3 nps 703 tbhits 0 pv g1f3 d7d5 d2d4 g8f6 c1f4 c7c5 e2e3 c5d4 e3d4
info depth 5 seldepth 12 time 873 nodes 747 score cp 7 hashfull 3 nps 855 tbhits 0 pv g1f3 d7d5 d2d4 g8f6 c1f4 c7c5 e2e3 c5d4 e3d4
info depth 5 seldepth 13 time 1021 nodes 954 score cp 8 hashfull 4 nps 934 tbhits 0 pv g1f3 d7d5 d2d4 g8f6 c1f4 c7c5 e2e3 c5d4 e3d4
info depth 6 seldepth 13 time 1154 nodes 1143 score cp 8 hashfull 5 nps 990 tbhits 0 pv g1f3 d7d5 d2d4 g8f6 c1f4 c7c5 e2e3 c5d4 e3d4 b8c6
info depth 6 seldepth 14 time 1390 nodes 1483 score cp 9 hashfull 6 nps 1066 tbhits 0 pv g1f3 d7d5 d2d4 g8f6 c1f4 c7c5 e2e3 c5d4 e3d4 b8c6
info depth 6 seldepth 14 time 1820 nodes 2205 score cp 13 hashfull 9 nps 1211 tbhits 0 pv d2d4 g8f6 c2c4 e7e6 g2g3 f8b4 b1d2 d7d5 f1g2 e8g8 g1f3
info depth 6 seldepth 14 time 1919 nodes 2291 score cp 9 hashfull 9 nps 1193 tbhits 0 pv g1f3 d7d5 d2d4 g8f6 c1f4 c7c5 e2e3 c5d4 e3d4 b8c6 c2c3
info depth 6 seldepth 14 time 1943 nodes 2402 score cp 13 hashfull 10 nps 1236 tbhits 0 pv d2d4 g8f6 c2c4 e7e6 g2g3 f8b4 b1d2 d7d5 f1g2 e8g8 g1f3
info depth 6 seldepth 15 time 2265 nodes 2951 score cp 12 hashfull 12 nps 1302 tbhits 0 pv d2d4 g8f6 c2c4 e7e6 g2g3 f8b4 b1d2 d7d5 f1g2 e8g8 g1f3 d5c4
info depth 6 seldepth 15 time 2347 nodes 3003 score cp 12 hashfull 12 nps 1279 tbhits 0 pv d2d4 g8f6 c2c4 e7e6 g2g3 f8b4 b1d2 d7d5 f1g2 e8g8 g1f3 d5c4 d4d5
bestmove d2d4 ponder g8f6
Everything works - you still use the weights file if you want, but you can't tune Lc0 with it, it's creating a bad tuning file. So move your weights file to a different folder and place this one there instead.

https://www.dropbox.com/s/6iy7cqeu89ijl34/weights?dl=1
I'm not even sure which it was at this point but it work great ,and you will get much better NPS.

Delete your current leelaz_opencl_tuning file and follow the steps you did before when you first ran LcO.

Also, dkappe has the best nets for those with less than high-end hardware - your nps will increase by 10X easily or more and you will be much more satisfied with lc0.

https://github.com/dkappe/leela-chess-w ... d-Networks
Thanks - that worked perfectly. I played a game against it. It sac'd the exchange and then crushed me with double-attack. I like it!!! Thanks so much again!

Re: Lc0 - macOS binary requested

Posted: Sun May 26, 2019 5:34 am
by smcracraft
MikeB wrote: Sun May 26, 2019 2:48 am
smcracraft wrote: Sat May 25, 2019 11:38 pm Also it crashes on the mac.

Stuarts-MacBook-Pro:release stuartcracraft$ ./lc0
_
| _ | |
|_ |_ |_| v0.22.0-dev built May 25 2019
ucinewgame
Found pb network file: ./weights_run1_42372.pb.gz
Creating backend [opencl]...
OpenCL, maximum batch size set to 16.
Initializing OpenCL.
Detected 1 OpenCL platforms.
Platform version: OpenCL 1.2 (Feb 22 2019 20:16:07)
Platform profile: FULL_PROFILE
Platform name: Apple
Platform vendor: Apple
Device ID: 0
Device name: Intel(R) Core(TM) i7-7920HQ CPU @ 3.10GHz
Device type: CPU
Device vendor: Intel
Device driver: 1.1
Device speed: 3100 MHZ
Device cores: 8 CU
Device score: 512
Device ID: 1
Device name: Intel(R) HD Graphics 630
Device type: GPU
Device vendor: Intel Inc.
Device driver: 1.2(Mar 11 2019 21:25:40)
Device speed: 1100 MHZ
Device cores: 24 CU
Device score: 612
Device ID: 2
Device name: AMD Radeon Pro 560 Compute Engine
Device type: GPU
Device vendor: AMD
Device driver: 1.2 (Mar 11 2019 21:14:43)
Device speed: 300 MHZ
Device cores: 16 CU
Device score: 1112
Selected platform: Apple
Selected device: AMD Radeon Pro 560 Compute Engine
with OpenCL 1.2 capability.
Loaded existing SGEMM tuning for batch size 16.
Wavefront/Warp size: 64

Max workgroup size: 256
Max workgroup dimensions: 256 256 256
go nodes 3000
info depth 1 seldepth 2 time 209 nodes 4 score cp 17 hashfull 0 nps 19 tbhits 0 pv e2e4 e7e5
info depth 2 seldepth 3 time 361 nodes 8 score cp 10 hashfull 0 nps 22 tbhits 0 pv g1f3 d7d5 d2d4
info depth 2 seldepth 4 time 705 nodes 10 score cp -71 hashfull 0 nps 14 tbhits 0 pv g1f3 e7e6 c2c4 f8e7
info depth 3 seldepth 4 time 706 nodes 19 score cp -201 hashfull 0 nps 26 tbhits 0 pv g1f3 e7e6 d2d4 h7h6
info depth 3 seldepth 5 time 1039 nodes 34 score cp 130 hashfull 0 nps 32 tbhits 0 pv e2e4 e7e5 b1c3 a7a6 c3d5
info depth 3 seldepth 5 time 1524 nodes 86 score cp -4 hashfull 1 nps 56 tbhits 0 pv e2e3 g8f6 g1f3 h8g8
info depth 3 seldepth 6 time 1879 nodes 127 score cp -71 hashfull 1 nps 67 tbhits 0 pv e2e3 g8f6 g1f3 h8g8
info depth 4 seldepth 7 time 2673 nodes 236 score cp 26 hashfull 2 nps 88 tbhits 0 pv e2e4 c7c6 c2c4 d8c7 g2g3 c6c5
info depth 4 seldepth 7 time 4609 nodes 566 score cp 13 hashfull 4 nps 122 tbhits 0 pv b1c3 d7d5 a1b1 f7f5 a2a3 h7h6
info depth 4 seldepth 7 time 6259 nodes 753 score cp 124 hashfull 4 nps 120 tbhits 0 pv f2f4 a7a5 h2h4 c7c5 a2a3
info depth 5 seldepth 8 time 10000 nodes 1470 score cp 59 hashfull 8 nps 147 tbhits 0 pv f2f4 e7e6 c2c3 h7h6 a2a3 a7a6 h2h3
Segmentation fault: 11
Stuarts-MacBook-Pro:release stuartcracraft$
I was able to duplicate your error:

Code: Select all

Mac-Pro:release michaelbyrne$ lc0
       _
|   _ | |
|_ |_ |_| v0.22.0-dev built May 25 2019

uci
id name Lc0 v0.22.0-dev
id author The LCZero Authors.
option name WeightsFile type string default <autodiscover>
option name Backend type combo default opencl var opencl var blas var check var random var roundrobin var multiplexing var demux
option name BackendOptions type string default 
option name Threads type spin default 2 min 1 max 128
option name NNCacheSize type spin default 200000 min 0 max 999999999
option name MinibatchSize type spin default 256 min 1 max 1024
option name MaxPrefetch type spin default 32 min 0 max 1024
option name CPuct type string default 3.000000
option name CPuctBase type string default 19652.000000
option name CPuctFactor type string default 2.000000
option name Temperature type string default 0.000000
option name TempDecayMoves type spin default 0 min 0 max 100
option name TempCutoffMove type spin default 0 min 0 max 1000
option name TempEndgame type string default 0.000000
option name TempValueCutoff type string default 100.000000
option name TempVisitOffset type string default 0.000000
option name DirichletNoise type check default false
option name VerboseMoveStats type check default false
option name SmartPruningFactor type string default 1.330000
option name FpuStrategy type combo default reduction var reduction var absolute
option name FpuValue type string default 1.200000
option name FpuStrategyAtRoot type combo default same var reduction var absolute var same
option name FpuValueAtRoot type string default 1.000000
option name CacheHistoryLength type spin default 0 min 0 max 7
option name PolicyTemperature type string default 2.200000
option name MaxCollisionEvents type spin default 32 min 1 max 1024
option name MaxCollisionVisits type spin default 9999 min 1 max 1000000
option name OutOfOrderEval type check default true
option name StickyEndgames type check default true
option name SyzygyFastPlay type check default true
option name MultiPV type spin default 1 min 1 max 500
option name ScoreType type combo default centipawn var centipawn var centipawn_2018 var win_percentage var Q
option name HistoryFill type combo default fen_only var no var fen_only var always
option name KLDGainAverageInterval type spin default 100 min 1 max 10000000
option name MinimumKLDGainPerNode type string default 0.000000
option name Slowmover type string default 1.000000
option name MoveOverheadMs type spin default 200 min 0 max 100000000
option name SyzygyPath type string default 
option name Ponder type check default true
option name ImmediateTimeUse type string default 1.000000
option name RamLimitMb type spin default 0 min 0 max 100000000
option name ConfigFile type string default lc0.config
option name LogFile type string default 
uciok
ucinewgame
Found pb network file: ./weights_run1_42372.pb.gz
Creating backend [opencl]...
OpenCL, maximum batch size set to 16.
Initializing OpenCL.
Detected 1 OpenCL platforms.
Platform version: OpenCL 1.2 (Apr  7 2019 18:38:19)
Platform profile: FULL_PROFILE
Platform name:    Apple
Platform vendor:  Apple
Device ID:      0
Device name:    Intel(R) Xeon(R) CPU           X5690  @ 3.47GHz
Device type:    CPU
Device vendor:  Intel
Device driver:  1.1
Device speed:   3460 MHZ
Device cores:   24 CU
Device score:   512
Device ID:      1
Device name:    AMD Radeon HD 7950 Compute Engine
Device type:    GPU
Device vendor:  AMD
Device driver:  1.2 (Apr 29 2019 22:28:38)
Device speed:   150 MHZ
Device cores:   28 CU
Device score:   1112
Selected platform: Apple
Selected device: AMD Radeon HD 7950 Compute Engine
with OpenCL 1.2 capability.
Started OpenCL SGEMM tuner with batch size 16.
Will try 578 valid configurations.
(1/578) KWG=32 KWI=2 MDIMA=8 MDIMC=8 MWG=16 NDIMB=8 NDIMC=8 NWG=16 SA=0 SB=0 STRM=0 STRN=0 VWM=1 VWN=1 2234.1 us (240.3 GFLOPS)
(2/578) KWG=32 KWI=2 MDIMA=8 MDIMC=8 MWG=32 NDIMB=8 NDIMC=8 NWG=16 SA=0 SB=0 STRM=0 STRN=0 VWM=1 VWN=1 874.5 us (613.9 GFLOPS)
(4/578) KWG=32 KWI=2 MDIMA=8 MDIMC=8 MWG=16 NDIMB=8 NDIMC=8 NWG=32 SA=0 SB=0 STRM=0 STRN=0 VWM=1 VWN=1 867.5 us (618.9 GFLOPS)
(5/578) KWG=32 KWI=2 MDIMA=8 MDIMC=8 MWG=32 NDIMB=8 NDIMC=8 NWG=32 SA=0 SB=0 STRM=0 STRN=0 VWM=1 VWN=1 436.1 us (1231.0 GFLOPS)
(15/578) KWG=32 KWI=2 MDIMA=16 MDIMC=16 MWG=64 NDIMB=8 NDIMC=8 NWG=32 SA=0 SB=0 STRM=0 STRN=0 VWM=1 VWN=1 431.5 us (1244.2 GFLOPS)
(115/578) KWG=32 KWI=2 MDIMA=8 MDIMC=8 MWG=32 NDIMB=8 NDIMC=8 NWG=32 SA=0 SB=0 STRM=0 STRN=0 VWM=4 VWN=1 430.8 us (1246.1 GFLOPS)
(117/578) KWG=32 KWI=2 MDIMA=8 MDIMC=8 MWG=32 NDIMB=8 NDIMC=8 NWG=64 SA=0 SB=0 STRM=0 STRN=0 VWM=4 VWN=1 362.3 us (1482.0 GFLOPS)
(283/578) KWG=32 KWI=2 MDIMA=8 MDIMC=8 MWG=32 NDIMB=8 NDIMC=8 NWG=64 SA=0 SB=0 STRM=0 STRN=0 VWM=4 VWN=4 360.3 us (1489.9 GFLOPS)
(478/578) KWG=32 KWI=2 MDIMA=8 MDIMC=8 MWG=32 NDIMB=8 NDIMC=8 NWG=32 SA=1 SB=1 STRM=0 STRN=0 VWM=2 VWN=2 354.0 us (1516.5 GFLOPS)
Wavefront/Warp size: 64

Max workgroup size: 256
Max workgroup dimensions: 256 256 256
go nodes 3000
info depth 1 seldepth 2 time 514 nodes 4 score cp 17 hashfull 0 nps 7 tbhits 0 pv e2e4 e7e5
info depth 2 seldepth 3 time 591 nodes 8 score cp 10 hashfull 0 nps 13 tbhits 0 pv g1f3 d7d5 d2d4
info depth 3 seldepth 4 time 744 nodes 17 score cp 11 hashfull 0 nps 22 tbhits 0 pv g1f3 d7d5 d2d4 g8f6
info depth 3 seldepth 5 time 834 nodes 34 score cp 20 hashfull 0 nps 40 tbhits 0 pv g1f3 e7e6 d2d4 d7d5 f3g5
info depth 3 seldepth 6 time 1000 nodes 38 score cp 20 hashfull 0 nps 38 tbhits 0 pv g1f3 e7e6 d2d4 d7d5 f3g5
info depth 4 seldepth 6 time 1067 nodes 94 score cp -131 hashfull 1 nps 88 tbhits 0 pv e2e3 b7b6 e3e4 h7h6
info depth 4 seldepth 6 time 1388 nodes 162 score cp -5 hashfull 1 nps 116 tbhits 0 pv g1f3 e7e6 d2d4 d7d5 e2e3 h7h6
info depth 4 seldepth 7 time 1564 nodes 227 score cp -68 hashfull 2 nps 145 tbhits 0 pv g1f3 e7e6 d2d4 d7d5 e2e3 h7h6
info depth 4 seldepth 7 time 1847 nodes 328 score cp -75 hashfull 2 nps 177 tbhits 0 pv c2c3 e7e5 g2g3 a7a6 h2h3
info depth 4 seldepth 8 time 2069 nodes 362 score cp -75 hashfull 2 nps 174 tbhits 0 pv c2c3 e7e5 g2g3 a7a6 h2h3
info depth 4 seldepth 9 time 2515 nodes 462 score cp -75 hashfull 3 nps 183 tbhits 0 pv c2c3 e7e5 g2g3 a7a6 h2h3
info depth 4 seldepth 9 time 3168 nodes 644 score cp 50 hashfull 4 nps 203 tbhits 0 pv e2e4 e7e5 g1f3 b8c6 d1e2 h7h6 b1a3 a7a6 c2c3
info depth 4 seldepth 9 time 6636 nodes 1720 score cp 59 hashfull 11 nps 259 tbhits 0 pv a2a4 e7e6 b1c3 g8f6 a4a5
info depth 5 seldepth 9 time 7534 nodes 1976 score cp 35 hashfull 12 nps 262 tbhits 0 pv a2a4 g7g6 e2e4 a7a5 f1c4 b7b6 h2h3
info depth 5 seldepth 9 time 10873 nodes 2904 score cp -2147483648 hashfull 18 nps 267 tbhits 0 pv e2e4 a7a6 b2b3 e7e6 d1e2 h7h6 f2f3 h6h5
Segmentation fault: 11
My initial thought that there was something wrong with the NN you downloaded , so I downloaded the NN you used and tried it. There were no issues.

Code: Select all

Mac-Pro:release michaelbyrne$ lc0
       _
|   _ | |
|_ |_ |_| v0.22.0-dev built May 25 2019
ucinewgame
Found pb network file: ./weights_run1_42372.pb.gz
Creating backend [opencl]...
OpenCL, maximum batch size set to 16.
Initializing OpenCL.
Detected 1 OpenCL platforms.
Platform version: OpenCL 1.2 (Apr  7 2019 18:38:19)
Platform profile: FULL_PROFILE
Platform name:    Apple
Platform vendor:  Apple
Device ID:      0
Device name:    Intel(R) Xeon(R) CPU           X5690  @ 3.47GHz
Device type:    CPU
Device vendor:  Intel
Device driver:  1.1
Device speed:   3460 MHZ
Device cores:   24 CU
Device score:   512
Device ID:      1
Device name:    AMD Radeon HD 7950 Compute Engine
Device type:    GPU
Device vendor:  AMD
Device driver:  1.2 (Apr 29 2019 22:28:38)
Device speed:   150 MHZ
Device cores:   28 CU
Device score:   1112
Selected platform: Apple
Selected device: AMD Radeon HD 7950 Compute Engine
with OpenCL 1.2 capability.
Loaded existing SGEMM tuning for batch size 16.
Wavefront/Warp size: 64

Max workgroup size: 256
Max workgroup dimensions: 256 256 256
go nodes 3000
info depth 1 seldepth 2 time 611 nodes 4 score cp 17 hashfull 0 nps 6 tbhits 0 pv e2e4 e7e5
info depth 2 seldepth 3 time 722 nodes 8 score cp 10 hashfull 0 nps 11 tbhits 0 pv g1f3 d7d5 d2d4
info depth 3 seldepth 4 time 886 nodes 17 score cp 11 hashfull 0 nps 19 tbhits 0 pv g1f3 d7d5 d2d4 g8f6
info depth 3 seldepth 5 time 1050 nodes 25 score cp 9 hashfull 0 nps 23 tbhits 0 pv g1f3 d7d5 g2g3 c7c5 f1g2
info depth 4 seldepth 6 time 1203 nodes 63 score cp 10 hashfull 1 nps 52 tbhits 0 pv g1f3 d7d5 g2g3 c7c5 f1g2
info depth 4 seldepth 7 time 1374 nodes 82 score cp 10 hashfull 1 nps 59 tbhits 0 pv g1f3 d7d5 g2g3 e7e6 f1g2 c7c5
info depth 4 seldepth 8 time 1539 nodes 111 score cp 15 hashfull 1 nps 72 tbhits 0 pv e2e4 e7e5 g1f3 b8c6 b1c3 g8f6 f1b5 f8d6
info depth 4 seldepth 8 time 1641 nodes 132 score cp 11 hashfull 1 nps 80 tbhits 0 pv d2d4 g8f6 c2c4 e7e6 b1c3 f8b4 e2e3 e8g8 d4d5
info depth 5 seldepth 9 time 1706 nodes 146 score cp 16 hashfull 1 nps 85 tbhits 0 pv e2e4 e7e5 g1f3 b8c6 b1c3 g8f6 f1b5 f8b4 e1g1
info depth 5 seldepth 10 time 1872 nodes 205 score cp 16 hashfull 1 nps 109 tbhits 0 pv e2e4 e7e5 g1f3 b8c6 f1b5 g8f6 b1c3 f8d6 e1g1 e8g8
info depth 5 seldepth 11 time 2036 nodes 252 score cp 16 hashfull 2 nps 123 tbhits 0 pv e2e4 e7e5 g1f3 b8c6 b1c3 g8f6 f1b5 f8d6 e1g1 e8g8 b5a4
info depth 6 seldepth 11 time 2037 nodes 267 score cp 16 hashfull 2 nps 131 tbhits 0 pv e2e4 e7e5 g1f3 b8c6 b1c3 g8f6 f1b5 f8d6 e1g1 e8g8 b5a4
info depth 6 seldepth 12 time 2201 nodes 311 score cp 16 hashfull 2 nps 141 tbhits 0 pv e2e4 e7e5 b1c3 g8f6 g1f3 b8c6 f1b5 f8d6 e1g1 e8g8 b5a4 h7h6
info depth 6 seldepth 13 time 2343 nodes 398 score cp 16 hashfull 2 nps 169 tbhits 0 pv e2e4 e7e5 b1c3 g8f6 g1f3 b8c6 f1b5 f8d6 d2d3 e8g8 b5a4 f8e8 e1g1
info depth 6 seldepth 14 time 2506 nodes 463 score cp 17 hashfull 2 nps 184 tbhits 0 pv e2e4 e7e5 b1c3 g8f6 g1f3 b8c6 f1b5 f8d6 d2d3 e8g8 b5a4 f8e8 e1g1 h7h6
info depth 7 seldepth 15 time 2782 nodes 620 score cp 18 hashfull 3 nps 222 tbhits 0 pv e2e4 e7e5 b1c3 g8f6 g1f3 b8c6 f1b5 f8d6 d2d3 e8g8 b5a4 f8e8 e1g1 h7h6 a2a3
info depth 7 seldepth 16 time 3000 nodes 720 score cp 18 hashfull 3 nps 240 tbhits 0 pv e2e4 e7e5 b1c3 g8f6 g1f3 b8c6 f1b5 f8d6 d2d3 e8g8 b5a4 f8e8 e1g1 h7h6 a2a3 h6h5
info depth 7 seldepth 17 time 3436 nodes 912 score cp 18 hashfull 4 nps 265 tbhits 0 pv e2e4 e7e5 b1c3 g8f6 g1f3 b8c6 f1b5 f8d6 d2d3 e8g8 b5a4 f8e8 e1g1 h7h6 a2a3 a7a6 a4c6
info depth 7 seldepth 18 time 3605 nodes 1041 score cp 19 hashfull 4 nps 288 tbhits 0 pv e2e4 e7e5 b1c3 g8f6 g1f3 b8c6 f1b5 f8d6 d2d3 e8g8 b5a4 f8e8 e1g1 h7h6 a2a3 a7a6 h2h3
info depth 7 seldepth 19 time 3901 nodes 1089 score cp 19 hashfull 5 nps 279 tbhits 0 pv e2e4 e7e5 b1c3 g8f6 g1f3 b8c6 f1b5 f8d6 d2d3 e8g8 b5a4 f8e8 e1g1 h7h6 a2a3 a7a6 h2h3
info depth 7 seldepth 20 time 4232 nodes 1268 score cp 19 hashfull 5 nps 299 tbhits 0 pv e2e4 e7e5 b1c3 g8f6 g1f3 b8c6 f1b5 f8d6 d2d3 e8g8 b5a4 f8e8 e1g1 h7h6 a2a3 a7a6 h2h3
info depth 8 seldepth 20 time 4619 nodes 1619 score cp 19 hashfull 6 nps 350 tbhits 0 pv e2e4 e7e5 b1c3 g8f6 g1f3 b8c6 f1b5 f8d6 d2d3 e8g8 b5a4 f8e8 e1g1 h7h6 a2a3 a7a6 h2h3 d6f8
info depth 8 seldepth 21 time 5399 nodes 1921 score cp 19 hashfull 8 nps 355 tbhits 0 pv e2e4 e7e5 b1c3 g8f6 g1f3 b8c6 f1b5 f8d6 d2d3 e8g8 b5a4 f8e8 e1g1 h7h6 a2a3 a7a6 h2h3 d6f8 a4b3
info depth 8 seldepth 22 time 6792 nodes 2800 score cp 17 hashfull 10 nps 412 tbhits 0 pv e2e4 e7e5 b1c3 g8f6 g1f3 b8c6 f1b5 f8d6 d2d3 e8g8 b5a4 f8e8 e1g1 h7h6 a2a3 a7a6 h2h3 d6f8 a4b3
info depth 8 seldepth 22 time 7265 nodes 3052 score cp 16 hashfull 11 nps 420 tbhits 0 pv e2e4 e7e5 b1c3 g8f6 g1f3 b8c6 f1b5 f8d6 d2d3 e8g8 b5a4 f8e8 e1g1 h7h6 a2a3 a7a6 h2h3 d6f8 a4b3
bestmove e2e4 ponder e7e5
quit
SoI took this one step further, deleted the leelaz_opencl_tuning, and used the net you were using to rebuild the leelaz_opencl_tuning file. Bingo, I got the seg 11 error. I deleted the weight file you were using, and the leelaz_opencl_tuning file. I put my weights file back.

Code: Select all

Mac-Pro:release michaelbyrne$ lc0
       _
|   _ | |
|_ |_ |_| v0.22.0-dev built May 25 2019
ucinewgame
Found pb network file: ./weights
Creating backend [opencl]...
OpenCL, maximum batch size set to 16.
Initializing OpenCL.
Detected 1 OpenCL platforms.
Platform version: OpenCL 1.2 (Apr  7 2019 18:38:19)
Platform profile: FULL_PROFILE
Platform name:    Apple
Platform vendor:  Apple
Device ID:      0
Device name:    Intel(R) Xeon(R) CPU           X5690  @ 3.47GHz
Device type:    CPU
Device vendor:  Intel
Device driver:  1.1
Device speed:   3460 MHZ
Device cores:   24 CU
Device score:   512
Device ID:      1
Device name:    AMD Radeon HD 7950 Compute Engine
Device type:    GPU
Device vendor:  AMD
Device driver:  1.2 (Apr 29 2019 22:28:38)
Device speed:   150 MHZ
Device cores:   28 CU
Device score:   1112
Selected platform: Apple
Selected device: AMD Radeon HD 7950 Compute Engine
with OpenCL 1.2 capability.
Started OpenCL SGEMM tuner with batch size 16.
Will try 578 valid configurations.
(2/578) KWG=32 KWI=2 MDIMA=8 MDIMC=8 MWG=32 NDIMB=8 NDIMC=8 NWG=16 SA=0 SB=0 STRM=0 STRN=0 VWM=1 VWN=1 249.9 us (411.1 GFLOPS)
(5/578) KWG=32 KWI=2 MDIMA=8 MDIMC=8 MWG=32 NDIMB=8 NDIMC=8 NWG=32 SA=0 SB=0 STRM=0 STRN=0 VWM=1 VWN=1 143.7 us (715.2 GFLOPS)
(15/578) KWG=32 KWI=2 MDIMA=16 MDIMC=16 MWG=64 NDIMB=8 NDIMC=8 NWG=32 SA=0 SB=0 STRM=0 STRN=0 VWM=1 VWN=1 141.1 us (728.2 GFLOPS)
(42/578) KWG=32 KWI=2 MDIMA=16 MDIMC=16 MWG=64 NDIMB=16 NDIMC=16 NWG=64 SA=0 SB=0 STRM=0 STRN=0 VWM=1 VWN=1 138.5 us (741.8 GFLOPS)
(115/578) KWG=32 KWI=2 MDIMA=8 MDIMC=8 MWG=32 NDIMB=8 NDIMC=8 NWG=32 SA=0 SB=0 STRM=0 STRN=0 VWM=4 VWN=1 126.5 us (812.1 GFLOPS)
(117/578) KWG=32 KWI=2 MDIMA=8 MDIMC=8 MWG=32 NDIMB=8 NDIMC=8 NWG=64 SA=0 SB=0 STRM=0 STRN=0 VWM=4 VWN=1 122.1 us (841.6 GFLOPS)
(121/578) KWG=32 KWI=2 MDIMA=16 MDIMC=16 MWG=64 NDIMB=8 NDIMC=8 NWG=64 SA=0 SB=0 STRM=0 STRN=0 VWM=4 VWN=1 121.2 us (847.8 GFLOPS)
(376/578) KWG=32 KWI=2 MDIMA=8 MDIMC=8 MWG=32 NDIMB=16 NDIMC=16 NWG=32 SA=1 SB=1 STRM=0 STRN=0 VWM=2 VWN=1 116.3 us (883.6 GFLOPS)
(485/578) KWG=32 KWI=2 MDIMA=16 MDIMC=16 MWG=32 NDIMB=8 NDIMC=8 NWG=32 SA=1 SB=1 STRM=0 STRN=0 VWM=2 VWN=2 111.1 us (925.2 GFLOPS)
Wavefront/Warp size: 64

Max workgroup size: 256
Max workgroup dimensions: 256 256 256
go nodes 3000     
info depth 1 seldepth 2 time 190 nodes 7 score cp 8 hashfull 0 nps 36 tbhits 0 pv d2d4 g8f6
info depth 2 seldepth 3 time 226 nodes 16 score cp 5 hashfull 0 nps 70 tbhits 0 pv g2g3 d7d5 g1f3
info depth 3 seldepth 4 time 262 nodes 35 score cp 6 hashfull 0 nps 133 tbhits 0 pv g2g3 e7e6 c2c4 d7d5
info depth 3 seldepth 5 time 302 nodes 57 score cp 5 hashfull 0 nps 188 tbhits 0 pv g2g3 e7e6 c2c4 d7d5 f1g2
info depth 3 seldepth 5 time 305 nodes 59 score cp 5 hashfull 0 nps 193 tbhits 0 pv c2c4 e7e5 d2d3 b8c6 g1f3
info depth 4 seldepth 5 time 324 nodes 67 score cp 5 hashfull 0 nps 206 tbhits 0 pv g2g3 d7d5 g1f3 c7c5 f1g2 d5d4
info depth 4 seldepth 6 time 350 nodes 92 score cp 6 hashfull 0 nps 262 tbhits 0 pv g2g3 e7e6 c2c4 d7d5 f1g2 d5d4
info depth 4 seldepth 7 time 395 nodes 128 score cp 6 hashfull 1 nps 324 tbhits 0 pv c2c4 e7e5 d2d3 f8e7 g1f3 b8c6
info depth 4 seldepth 8 time 445 nodes 179 score cp 6 hashfull 1 nps 402 tbhits 0 pv c2c4 e7e5 d2d3 f8e7 g1f3 b8c6
info depth 4 seldepth 8 time 494 nodes 235 score cp 7 hashfull 1 nps 475 tbhits 0 pv g1f3 d7d5 d2d4 g8f6 c1f4 c8f5 e2e3 e7e6
info depth 4 seldepth 9 time 520 nodes 260 score cp 7 hashfull 1 nps 500 tbhits 0 pv g1f3 d7d5 d2d4 g8f6 c1f4 c8f5 e2e3 e7e6
info depth 5 seldepth 9 time 568 nodes 306 score cp 6 hashfull 2 nps 538 tbhits 0 pv g1f3 d7d5 d2d4 g8f6 c1f4 c8f5 c2c4 e7e6 b1c3
info depth 5 seldepth 10 time 640 nodes 415 score cp 6 hashfull 2 nps 648 tbhits 0 pv g1f3 d7d5 d2d4 g8f6 c1f4 c7c5 e2e3 c5d4 e3d4
info depth 5 seldepth 11 time 763 nodes 537 score cp 7 hashfull 3 nps 703 tbhits 0 pv g1f3 d7d5 d2d4 g8f6 c1f4 c7c5 e2e3 c5d4 e3d4
info depth 5 seldepth 12 time 873 nodes 747 score cp 7 hashfull 3 nps 855 tbhits 0 pv g1f3 d7d5 d2d4 g8f6 c1f4 c7c5 e2e3 c5d4 e3d4
info depth 5 seldepth 13 time 1021 nodes 954 score cp 8 hashfull 4 nps 934 tbhits 0 pv g1f3 d7d5 d2d4 g8f6 c1f4 c7c5 e2e3 c5d4 e3d4
info depth 6 seldepth 13 time 1154 nodes 1143 score cp 8 hashfull 5 nps 990 tbhits 0 pv g1f3 d7d5 d2d4 g8f6 c1f4 c7c5 e2e3 c5d4 e3d4 b8c6
info depth 6 seldepth 14 time 1390 nodes 1483 score cp 9 hashfull 6 nps 1066 tbhits 0 pv g1f3 d7d5 d2d4 g8f6 c1f4 c7c5 e2e3 c5d4 e3d4 b8c6
info depth 6 seldepth 14 time 1820 nodes 2205 score cp 13 hashfull 9 nps 1211 tbhits 0 pv d2d4 g8f6 c2c4 e7e6 g2g3 f8b4 b1d2 d7d5 f1g2 e8g8 g1f3
info depth 6 seldepth 14 time 1919 nodes 2291 score cp 9 hashfull 9 nps 1193 tbhits 0 pv g1f3 d7d5 d2d4 g8f6 c1f4 c7c5 e2e3 c5d4 e3d4 b8c6 c2c3
info depth 6 seldepth 14 time 1943 nodes 2402 score cp 13 hashfull 10 nps 1236 tbhits 0 pv d2d4 g8f6 c2c4 e7e6 g2g3 f8b4 b1d2 d7d5 f1g2 e8g8 g1f3
info depth 6 seldepth 15 time 2265 nodes 2951 score cp 12 hashfull 12 nps 1302 tbhits 0 pv d2d4 g8f6 c2c4 e7e6 g2g3 f8b4 b1d2 d7d5 f1g2 e8g8 g1f3 d5c4
info depth 6 seldepth 15 time 2347 nodes 3003 score cp 12 hashfull 12 nps 1279 tbhits 0 pv d2d4 g8f6 c2c4 e7e6 g2g3 f8b4 b1d2 d7d5 f1g2 e8g8 g1f3 d5c4 d4d5
bestmove d2d4 ponder g8f6
Everything works - you still use the weights file if you want, but you can't tune Lc0 with it, it's creating a bad tuning file. So move your weights file to a different folder and place this one there instead.

https://www.dropbox.com/s/6iy7cqeu89ijl34/weights?dl=1
I'm not even sure which it was at this point but it work great ,and you will get much better NPS.

Delete your current leelaz_opencl_tuning file and follow the steps you did before when you first ran LcO.

Also, dkappe has the best nets for those with less than high-end hardware - your nps will increase by 10X easily or more and you will be much more satisfied with lc0.

https://github.com/dkappe/leela-chess-w ... d-Networks

Here's the crush of me by the new version per your inputs.

Stuarts-MacBook-Pro:release stuartcracraft$ ./lc0
_
| _ | |
|_ |_ |_| v0.22.0-dev built May 25 2019
ucinewgame
Found pb network file: ./weights
Creating backend [opencl]...
OpenCL, maximum batch size set to 16.
Initializing OpenCL.
Detected 1 OpenCL platforms.
Platform version: OpenCL 1.2 (Feb 22 2019 20:16:07)
Platform profile: FULL_PROFILE
Platform name: Apple
Platform vendor: Apple
Device ID: 0
Device name: Intel(R) Core(TM) i7-7920HQ CPU @ 3.10GHz
Device type: CPU
Device vendor: Intel
Device driver: 1.1
Device speed: 3100 MHZ
Device cores: 8 CU
Device score: 512
Device ID: 1
Device name: Intel(R) HD Graphics 630
Device type: GPU
Device vendor: Intel Inc.
Device driver: 1.2(Mar 11 2019 21:25:40)
Device speed: 1100 MHZ
Device cores: 24 CU
Device score: 612
Device ID: 2
Device name: AMD Radeon Pro 560 Compute Engine
Device type: GPU
Device vendor: AMD
Device driver: 1.2 (Mar 11 2019 21:14:43)
Device speed: 300 MHZ
Device cores: 16 CU
Device score: 1112
Selected platform: Apple
Selected device: AMD Radeon Pro 560 Compute Engine
with OpenCL 1.2 capability.
Loaded existing SGEMM tuning for batch size 16.
Wavefront/Warp size: 64

Max workgroup size: 256
Max workgroup dimensions: 256 256 256
go nodes 3000
info depth 1 seldepth 2 time 116 nodes 7 score cp 8 hashfull 0 nps 60 tbhits 0 pv d2d4 g8f6
info depth 2 seldepth 3 time 151 nodes 16 score cp 5 hashfull 0 nps 105 tbhits 0 pv g2g3 d7d5 g1f3
info depth 3 seldepth 4 time 189 nodes 35 score cp 6 hashfull 0 nps 185 tbhits 0 pv g2g3 e7e6 c2c4 d7d5
info depth 4 seldepth 5 time 258 nodes 65 score cp 5 hashfull 0 nps 251 tbhits 0 pv g2g3 d7d5 g1f3 c7c5 f1g2
info depth 4 seldepth 6 time 318 nodes 79 score cp 5 hashfull 0 nps 248 tbhits 0 pv c2c4 e7e5 d2d3 g8f6 g1f3
info depth 4 seldepth 6 time 318 nodes 100 score cp 6 hashfull 0 nps 314 tbhits 0 pv g2g3 e7e6 c2c4 d7d5 f1g2 d5d4
info depth 4 seldepth 7 time 375 nodes 118 score cp 6 hashfull 1 nps 314 tbhits 0 pv c2c4 e7e5 d2d3 f8e7 g1f3 b8c6
info depth 4 seldepth 8 time 441 nodes 165 score cp 6 hashfull 1 nps 374 tbhits 0 pv c2c4 e7e5 d2d3 f8e7 g1f3 b8c6
info depth 4 seldepth 9 time 505 nodes 226 score cp 7 hashfull 1 nps 447 tbhits 0 pv c2c4 e7e5 g2g3 b8c6 f1g2 g7g6 b1c3 d7d6 c4c5
info depth 4 seldepth 9 time 527 nodes 263 score cp 7 hashfull 1 nps 499 tbhits 0 pv g1f3 d7d5 d2d4 g8f6 c1f4 c8f5 e2e3 e7e6 f4c7
info depth 5 seldepth 9 time 582 nodes 299 score cp 7 hashfull 1 nps 513 tbhits 0 pv c2c4 e7e5 g2g3 b8c6 f1g2 g7g6 b1c3 d7d6 d2d3
info depth 5 seldepth 9 time 583 nodes 335 score cp 6 hashfull 2 nps 574 tbhits 0 pv g1f3 d7d5 d2d4 g8f6 c1f4 c8f5 e2e3 e7e6 f1d3
info depth 5 seldepth 9 time 648 nodes 364 score cp 7 hashfull 2 nps 561 tbhits 0 pv c2c4 e7e5 g2g3 b8c6 f1g2 g7g6 b1c3 d7d6 d2d3
info depth 5 seldepth 9 time 648 nodes 388 score cp 7 hashfull 2 nps 598 tbhits 0 pv g1f3 d7d5 d2d4 g8f6 c1f4 c8f5 e2e3 e7e6 f1d3
info depth 5 seldepth 10 time 736 nodes 444 score cp 6 hashfull 2 nps 603 tbhits 0 pv g1f3 d7d5 d2d4 g8f6 c1f4 c8f5 e2e3 e7e6 f1d3
info depth 5 seldepth 11 time 815 nodes 531 score cp 6 hashfull 2 nps 651 tbhits 0 pv g1f3 d7d5 d2d4 g8f6 c1f4 c8f5 e2e3 e7e6 f1d3
info depth 5 seldepth 12 time 937 nodes 649 score cp 7 hashfull 3 nps 692 tbhits 0 pv g1f3 d7d5 d2d4 g8f6 c1f4 c8f5 e2e3 e7e6 f1d3
info depth 5 seldepth 13 time 1092 nodes 809 score cp 7 hashfull 4 nps 740 tbhits 0 pv g1f3 d7d5 d2d4 g8f6 c1f4 c8f5 e2e3 e7e6 f1d3
info depth 6 seldepth 13 time 1446 nodes 1223 score cp 9 hashfull 5 nps 845 tbhits 0 pv g1f3 d7d5 d2d4 g8f6 c1f4 c8f5 e2e3 e7e6 f1d3
info depth 6 seldepth 14 time 1620 nodes 1486 score cp 8 hashfull 6 nps 917 tbhits 0 pv g1f3 d7d5 d2d4 g8f6 c1f4 c8f5 c2c4 e7e6 b1c3 f8b4
info depth 6 seldepth 14 time 2135 nodes 2121 score cp 12 hashfull 9 nps 993 tbhits 0 pv d2d4 g8f6 c2c4 e7e6 g2g3 f8b4 b1d2 d7d5 f1g2 e8g8 g1f3 d5c4
info depth 6 seldepth 15 time 2459 nodes 2608 score cp 13 hashfull 10 nps 1060 tbhits 0 pv d2d4 g8f6 c2c4 e7e6 g2g3 f8b4 b1d2 d7d5 f1g2 e8g8 g1f3 d5c4
info depth 6 seldepth 15 time 2885 nodes 3073 score cp 12 hashfull 12 nps 1065 tbhits 0 pv d2d4 g8f6 c2c4 e7e6 g2g3 f8b4 b1d2 d7d5 f1g2 e8g8 g1f3 d5c4 d4d5
bestmove d2d4 ponder g8f6
position startpos moves d2d4 d7d5
go nodes 3000
info depth 6 seldepth 11 time 104 nodes 115 score cp 9 hashfull 13 nps 451 tbhits 0 pv c2c4 e7e6 b1c3 g8f6 c4d5 e6d5 c1g5 c7c6 e2e3 h7h6 g5h4 c8f5 d1f3 f5b1
info depth 6 seldepth 14 time 108 nodes 136 score cp 9 hashfull 13 nps 629 tbhits 0 pv c2c4 e7e6 b1c3 g8f6 c4d5 e6d5 c1g5 c7c6 e2e3 h7h6 g5h4 c8f5 f1d3 f5d3
info depth 6 seldepth 15 time 851 nodes 985 score cp 12 hashfull 16 nps 1077 tbhits 0 pv c2c4 e7e6 b1c3 g8f6 c4d5 e6d5 c1g5 c7c6 e2e3 h7h6 g5h4 c8f5 f1d3 f5d3 d1d3
info depth 6 seldepth 15 time 1948 nodes 2280 score cp 12 hashfull 21 nps 1135 tbhits 0 pv c2c4 e7e6 b1c3 g8f6 c4d5 e6d5 c1g5 c7c6 e2e3 h7h6 g5h4 c8f5 f1d3 f5d3 d1d3
bestmove c2c4 ponder e7e6
position startpos moves d2d4 d7d5 c2c4 g8f6
go nodes 3000
info depth 7 seldepth 10 time 102 nodes 35 score cp 20 hashfull 22 nps 88 tbhits 0 pv c4d5 c8f5 g1f3 f5b1 d1a4 c7c6 d5c6 b8c6 a1b1 f6g4
info depth 6 seldepth 10 time 110 nodes 83 score cp 22 hashfull 22 nps 518 tbhits 0 pv c4d5 c8f5 g1f3 f5b1 d1a4 c7c6 d5c6 b8c6 a1b1 e7e6
info depth 5 seldepth 10 time 205 nodes 130 score cp 24 hashfull 23 nps 507 tbhits 0 pv c4d5 c8f5 g1f3 f5b1 d1a4 c7c6 d5c6 b8c6 a1b1 e7e6
info depth 5 seldepth 11 time 294 nodes 224 score cp 25 hashfull 23 nps 673 tbhits 0 pv c4d5 c8f5 g1f3 f5b1 d1a4 c7c6 d5c6 b8c6 a1b1 e7e6
info depth 5 seldepth 13 time 410 nodes 343 score cp 25 hashfull 23 nps 773 tbhits 0 pv c4d5 c8f5 g1f3 f5b1 d1a4 c7c6 d5c6 b8c6 a1b1 e7e6
info depth 6 seldepth 13 time 449 nodes 387 score cp 26 hashfull 24 nps 804 tbhits 0 pv c4d5 c8f5 g1f3 f5b1 d1a4 c7c6 d5c6 b8c6 a1b1 e7e6
info depth 6 seldepth 14 time 764 nodes 788 score cp 28 hashfull 25 nps 997 tbhits 0 pv c4d5 c8f5 g1f3 f5b1 d1a4 c7c6 d5c6 b8c6 a1b1 e7e6
info depth 6 seldepth 15 time 905 nodes 949 score cp 28 hashfull 26 nps 1019 tbhits 0 pv c4d5 c8f5 g1f3 f5b1 d1a4 c7c6 d5c6 b8c6 a1b1 e7e6
info depth 6 seldepth 16 time 1027 nodes 1066 score cp 29 hashfull 26 nps 1012 tbhits 0 pv c4d5 g7g6 g1f3 f8g7 d1a4 c7c6 d5c6 b8c6 e2e3 e8g8 f1e2
info depth 6 seldepth 17 time 1147 nodes 1175 score cp 30 hashfull 27 nps 1001 tbhits 0 pv c4d5 c7c6 d5c6 b8c6 g1f3 e7e5 d4e5 d8d1 e1d1 f6g4 d1e1 g4e5 b1c3 c8e6 f3g5
info depth 6 seldepth 18 time 1423 nodes 1549 score cp 28 hashfull 28 nps 1070 tbhits 0 pv c4d5 g7g6 g1f3 f8g7 d1a4 c7c6 d5c6 b8c6 b1c3 e8g8 e2e3 e7e5
info depth 7 seldepth 18 time 1698 nodes 1806 score cp 27 hashfull 30 nps 1048 tbhits 0 pv c4d5 g7g6 g1f3 f8g7 d1a4 c7c6 d5c6 b8c6 b1c3 e8g8 e2e3 e7e5
info depth 7 seldepth 18 time 1843 nodes 1942 score cp 27 hashfull 30 nps 1039 tbhits 0 pv c4d5 c7c6 d5c6 b8c6 g1f3 e7e5 d4e5 d8d1 e1d1 f6g4 d1e1 g4e5 b1c3 c8e6 f3e5 c6e5
bestmove c4d5 ponder c7c6
position startpos moves d2d4 d7d5 c2c4 g8f6 c4d5 f6d5
go nodes 3000
info depth 6 seldepth 8 time 101 nodes 73 score cp 31 hashfull 31 nps 99 tbhits 0 pv g1f3 c8f5 d1b3 b8c6 b1c3 d5b4 d4d5
info depth 6 seldepth 9 time 109 nodes 93 score cp 33 hashfull 31 nps 275 tbhits 0 pv g1f3 c8f5 b1d2 d5b4 e2e4 f5g4 d1b3 e7e6 a2a3
info depth 6 seldepth 10 time 174 nodes 147 score cp 34 hashfull 31 nps 482 tbhits 0 pv g1f3 c8f5 b1d2 d5b4 e2e4 f5g4 d1b3 e7e6 a2a3
info depth 5 seldepth 10 time 294 nodes 231 score cp 34 hashfull 32 nps 571 tbhits 0 pv g1f3 c8f5 b1d2 d5b4 e2e4 f5g4 d1b3 e7e6 a2a3
info depth 5 seldepth 11 time 747 nodes 688 score cp 33 hashfull 34 nps 836 tbhits 0 pv g1f3 c8f5 b1d2 d5f6 d1b3 b8d7 b3b7 e7e6 e2e4
info depth 6 seldepth 11 time 880 nodes 847 score cp 34 hashfull 34 nps 890 tbhits 0 pv g1f3 c8f5 b1d2 d5f6 d1b3 d8d5 b3d5 f6d5 e2e4
info depth 6 seldepth 12 time 1030 nodes 1038 score cp 34 hashfull 35 nps 946 tbhits 0 pv g1f3 c8f5 b1d2 d5f6 d1b3 d8c8 f3e5 f5e6 b3c2 g7g6
info depth 6 seldepth 13 time 1438 nodes 1455 score cp 34 hashfull 37 nps 968 tbhits 0 pv g1f3 c8f5 b1d2 d5f6 d1b3 d8c8 f3e5 f5e6 b3c2 g7g6
info depth 6 seldepth 14 time 2289 nodes 2284 score cp 35 hashfull 40 nps 970 tbhits 0 pv g1f3 c8f5 b1d2 d5f6 d1b3 d8c8 e2e4 f6e4 d2e4 f5e4
info depth 6 seldepth 14 time 2352 nodes 2511 score cp 35 hashfull 41 nps 1040 tbhits 0 pv g1f3 c8f5 b1d2 d5f6 d1b3 d8c8 e2e4 f6e4 d2e4 f5e4
bestmove g1f3 ponder c8f5
position startpos moves d2d4 d7d5 c2c4 g8f6 c4d5 f6d5 g1f3 c8f5
go nodes 3000
info depth 7 seldepth 10 time 103 nodes 585 score cp 44 hashfull 42 nps 126 tbhits 0 pv b1d2 d5f6 d1b3 b8d7 b3b7 a8b8 b7a7 b8a8 a7b7 f5b1
info depth 6 seldepth 12 time 169 nodes 645 score cp 44 hashfull 42 nps 431 tbhits 0 pv b1d2 d5f6 d1b3 b8d7 b3b7 a8b8 b7a7 b8a8 a7b7 f5b1
info depth 6 seldepth 13 time 734 nodes 1173 score cp 42 hashfull 45 nps 818 tbhits 0 pv b1d2 d5f6 d1b3 d8c8 e2e4 f6e4 d2e4 f5e4 f3g5 e4g6
info depth 6 seldepth 14 time 977 nodes 1433 score cp 41 hashfull 46 nps 881 tbhits 0 pv b1d2 d5f6 d1b3 d8c8 e2e4 f6e4 d2e4 f5e4 f3g5 e4g6 f1c4
info depth 6 seldepth 14 time 1745 nodes 2268 score cp 39 hashfull 50 nps 971 tbhits 0 pv b1d2 d5f6 d1b3 d8c8 f3e5 f5e6 b3c2 g7g6 e2e4 f8g7 d4d5 e6d7 f1c4
bestmove b1d2 ponder d5f6
position startpos moves d2d4 d7d5 c2c4 g8f6 c4d5 f6d5 g1f3 c8f5 b1d2 f5g6
go nodes 3000
info depth 6 seldepth 9 time 104 nodes 69 score cp 47 hashfull 50 nps 134 tbhits 0 pv e2e4 d5b6 h2h4 g6h5 a2a4 a7a5 f1e2 e7e6 e1g1 h5f3
info depth 6 seldepth 10 time 105 nodes 94 score cp 46 hashfull 51 nps 371 tbhits 0 pv e2e4 d5b6 h2h4 g6h5 a2a4 a7a5 f1e2 e7e6 e1g1 f8e7
info depth 5 seldepth 10 time 169 nodes 137 score cp 47 hashfull 51 nps 485 tbhits 0 pv e2e4 d5b6 h2h4 g6h5 a2a4 a7a5 f1e2 e7e6 e1g1 f8e7
info depth 5 seldepth 10 time 381 nodes 311 score cp 57 hashfull 52 nps 671 tbhits 0 pv h2h4 h7h6 e2e4 d5b6 a2a4 a7a5 f3e5 d8d4 e5g6
info depth 5 seldepth 11 time 514 nodes 396 score cp 54 hashfull 52 nps 663 tbhits 0 pv h2h4 g6h5 d1b3 d5b6 f3e5 e7e6 g2g4
info depth 6 seldepth 11 time 1061 nodes 975 score cp 51 hashfull 55 nps 867 tbhits 0 pv h2h4 d5f6 d1b3 b8d7 b3b7 c7c5 d4c5 d7c5
info depth 6 seldepth 12 time 1427 nodes 1276 score cp 47 hashfull 56 nps 855 tbhits 0 pv h2h4 d5f6 d1b3 b8d7 b3b7 e7e6 h4h5 g6h5 f3e5
info depth 6 seldepth 13 time 1696 nodes 1682 score cp 46 hashfull 58 nps 959 tbhits 0 pv h2h4 d5f6 d1b3 b8d7 b3b7 e7e6 h4h5 g6h5 f3e5
info depth 6 seldepth 14 time 2078 nodes 2143 score cp 45 hashfull 60 nps 1004 tbhits 0 pv h2h4 d5f6 d1b3 b8d7 b3b7 e7e6 h4h5 g6h5 f3e5
info depth 6 seldepth 14 time 2887 nodes 3001 score cp 42 hashfull 63 nps 1020 tbhits 0 pv h2h4 d5f6 d1b3 b8d7 b3b7 e7e6 h4h5 g6h5 f3e5
bestmove h2h4 ponder d5f6
position startpos moves d2d4 d7d5 c2c4 g8f6 c4d5 f6d5 g1f3 c8f5 b1d2 f5g6 h2h4 d5f6
go nodes 3000
info depth 6 seldepth 9 time 102 nodes 686 score cp 36 hashfull 64 nps 166 tbhits 0 pv d1b3 e7e6 b3b7 b8d7 h4h5 g6h5 f3e5 a8b8 b7c8
info depth 6 seldepth 11 time 262 nodes 791 score cp 36 hashfull 65 nps 465 tbhits 0 pv d1b3 e7e6 b3b7 b8d7 h4h5 g6h5 f3e5 d7e5 d4e5
info depth 7 seldepth 11 time 287 nodes 914 score cp 36 hashfull 65 nps 853 tbhits 0 pv d1b3 e7e6 b3b7 b8d7 h4h5 g6h5 f3e5 d7e5 d4e5
info depth 6 seldepth 11 time 385 nodes 982 score cp 36 hashfull 66 nps 812 tbhits 0 pv d1b3 e7e6 b3b7 b8d7 h4h5 g6h5 f3e5 d7e5 d4e5
info depth 6 seldepth 12 time 478 nodes 1116 score cp 38 hashfull 66 nps 935 tbhits 0 pv d1b3 e7e6 b3b7 b8d7 h4h5 g6h5 f3e5 d7e5 d4e5
info depth 6 seldepth 13 time 996 nodes 1670 score cp 38 hashfull 69 nps 1005 tbhits 0 pv d1b3 b8d7 b3b7 e7e6 h4h5 g6h5 f3e5 a8b8 b7a7 d7e5
info depth 6 seldepth 13 time 1542 nodes 2303 score cp 38 hashfull 71 nps 1059 tbhits 0 pv d1b3 e7e6 b3b7 b8d7 h4h5 g6h5 f3e5 f8d6 e5c6 d8c8
bestmove d1b3 ponder e7e6
position startpos moves d2d4 d7d5 c2c4 g8f6 c4d5 f6d5 g1f3 c8f5 b1d2 f5g6 h2h4 d5f6 d1b3 b7b6
go nodes 3000
info depth 4 seldepth 6 time 101 nodes 36 score cp 71 hashfull 72 nps 118 tbhits 0 pv g2g4 e7e6 h4h5 g6e4 d2e4 f6e4
info depth 5 seldepth 7 time 101 nodes 46 score cp 68 hashfull 72 nps 217 tbhits 0 pv g2g4 e7e6 h4h5 g6e4 d2e4 f6e4
info depth 4 seldepth 7 time 166 nodes 58 score cp 72 hashfull 72 nps 204 tbhits 0 pv g2g4 f6g4 h4h5 g6f5 e2e4 f5d7 f3g5
info depth 4 seldepth 8 time 166 nodes 71 score cp 71 hashfull 72 nps 283 tbhits 0 pv g2g4 f6g4 h4h5 g6f5 e2e4 f5d7 f3g5
info depth 5 seldepth 9 time 233 nodes 110 score cp 67 hashfull 73 nps 369 tbhits 0 pv g2g4 f6g4 h4h5 g6f5 e2e4 f5d7 f3g5 e7e6
info depth 5 seldepth 10 time 297 nodes 142 score cp 69 hashfull 73 nps 397 tbhits 0 pv g2g4 f6g4 h4h5 g6f5 e2e4 f5d7 f3g5 e7e6 h5h6
info depth 5 seldepth 11 time 474 nodes 293 score cp 66 hashfull 73 nps 567 tbhits 0 pv g2g4 f6g4 h4h5 g6f5 e2e4 f5d7 f3g5 e7e6 d2f3 h7h6 g5h3
info depth 5 seldepth 12 time 1081 nodes 926 score cp 61 hashfull 76 nps 834 tbhits 0 pv g2g4 f6g4 h4h5 g6f5 e2e4 f5d7 f3g5 e7e6 d2f3 h7h6 g5h3
info depth 6 seldepth 13 time 1149 nodes 984 score cp 61 hashfull 77 nps 835 tbhits 0 pv g2g4 f6g4 h4h5 g6f5 e2e4 f5d7 f3g5 e7e6 d2f3 h7h6 g5h3
info depth 6 seldepth 13 time 1230 nodes 1061 score cp 84 hashfull 77 nps 843 tbhits 0 pv h4h5 g6h5 h1h5 f6h5 f3e5 e7e6 b3f3 h5f6 f3a8 f8b4 a2a3 b4d2 c1d2
info depth 6 seldepth 14 time 2317 nodes 1980 score cp 82 hashfull 82 nps 844 tbhits 0 pv h4h5 g6f5 e2e4 f6e4 f3e5 e4d6 g2g4 f5e6 d4d5 e6c8 d2f3 f7f6
info depth 6 seldepth 14 time 2470 nodes 2197 score cp 80 hashfull 83 nps 879 tbhits 0 pv h4h5 g6f5 e2e4 f6e4 f3e5 e4d6 g2g4 f5e6 d4d5 e6c8 d2f3 f7f6
bestmove h4h5 ponder g6f5
position startpos moves d2d4 d7d5 c2c4 g8f6 c4d5 f6d5 g1f3 c8f5 b1d2 f5g6 h2h4 d5f6 d1b3 b7b6 h4h5 g6h5
go nodes 3000
info depth 7 seldepth 9 time 105 nodes 218 score cp 150 hashfull 83 nps 19 tbhits 0 pv h1h5 f6h5 f3e5 e7e6 b3f3 h5f6 f3a8 f8b4 a2a3 b4d2 c1d2
info depth 8 seldepth 12 time 108 nodes 241 score cp 155 hashfull 84 nps 231 tbhits 0 pv h1h5 f6h5 f3e5 e7e6 b3f3 h5f6 f3a8 f8b4 a2a3 b4d2 c1d2
info depth 8 seldepth 13 time 173 nodes 285 score cp 166 hashfull 84 nps 398 tbhits 0 pv h1h5 f6h5 f3e5 e7e6 b3f3 h5f6 f3a8 f8b4 e2e3 e8g8 a2a3 b4d6
info depth 7 seldepth 13 time 173 nodes 316 score cp 169 hashfull 84 nps 578 tbhits 0 pv h1h5 f6h5 f3e5 e7e6 b3f3 h5f6 f3a8 f8b4 e2e3 e8g8 a2a3 b4d6
info depth 7 seldepth 14 time 372 nodes 443 score cp 173 hashfull 85 nps 610 tbhits 0 pv h1h5 f6h5 f3e5 e7e6 b3f3 h5f6 f3a8 f8b4 e2e3 e8g8 a2a3 b4d6
info depth 7 seldepth 15 time 466 nodes 519 score cp 177 hashfull 85 nps 650 tbhits 0 pv h1h5 f6h5 f3e5 e7e6 b3f3 h5f6 f3a8 f8b4 e2e3 e8g8 a2a3 b4d6
info depth 8 seldepth 15 time 613 nodes 640 score cp 188 hashfull 86 nps 691 tbhits 0 pv h1h5 f6h5 f3e5 e7e6 b3f3 h5f6 f3a8 f8b4 a2a3 b4d2 c1d2 e8g8
info depth 7 seldepth 16 time 806 nodes 921 score cp 198 hashfull 87 nps 874 tbhits 0 pv h1h5 f6h5 f3e5 e7e6 b3f3 h5f6 f3a8 f8b4 a2a3 b4d2 c1d2 e8g8 e2e3
info depth 7 seldepth 17 time 1245 nodes 1276 score cp 213 hashfull 88 nps 851 tbhits 0 pv h1h5 f6h5 f3e5 e7e6 b3f3 h5f6 f3a8 f8b4 a2a3 b4d2 c1d2 e8g8 e2e3 f6g4
info depth 8 seldepth 17 time 1272 nodes 1430 score cp 220 hashfull 89 nps 954 tbhits 0 pv h1h5 f6h5 f3e5 e7e6 b3f3 f7f6 f3h5 g7g6 e5g6 h7g6 h5h8 d8d4 e2e3 d4e5 d2f3 e5a5
info depth 8 seldepth 17 time 1641 nodes 1774 score cp 222 hashfull 90 nps 949 tbhits 0 pv h1h5 f6h5 f3e5 e7e6 b3f3 h5f6 f3a8 f8b4 a2a3 b4d2 c1d2 e8g8 e2e3 c7c5
bestmove h1h5 ponder f6h5
position startpos moves d2d4 d7d5 c2c4 g8f6 c4d5 f6d5 g1f3 c8f5 b1d2 f5g6 h2h4 d5f6 d1b3 b7b6 h4h5 g6h5 h1h5 f6h5
go nodes 3000
info depth 5 seldepth 13 time 103 nodes 1021 score cp 253 hashfull 91 nps 407 tbhits 0 pv f3e5 e7e6 b3f3 h5f6 f3a8 f8b4 a2a3 b4d2 c1d2 e8g8 e2e3 c7c5
info depth 7 seldepth 13 time 138 nodes 1067 score cp 248 hashfull 92 nps 637 tbhits 0 pv f3e5 e7e6 b3f3 h5f6 f3a8 f8b4 a2a3 b4d2 c1d2 e8g8 e2e3 c7c5
info depth 8 seldepth 13 time 260 nodes 1188 score cp 237 hashfull 92 nps 803 tbhits 0 pv f3e5 e7e6 b3f3 h5f6 f3a8 f8b4 a2a3 b4d2 c1d2 e8g8 e2e3 c7c5
info depth 8 seldepth 14 time 311 nodes 1191 score cp 238 hashfull 92 nps 681 tbhits 0 pv f3e5 e7e6 b3f3 h5f6 f3a8 f8b4 a2a3 b4d2 c1d2 e8g8 e2e3 c7c5
info depth 8 seldepth 15 time 376 nodes 1250 score cp 231 hashfull 93 nps 720 tbhits 0 pv f3e5 e7e6 b3f3 h5f6 f3a8 f8b4 a2a3 b4d2 c1d2 e8g8 e2e3 c7c5
info depth 8 seldepth 16 time 630 nodes 1459 score cp 219 hashfull 94 nps 761 tbhits 0 pv f3e5 e7e6 b3f3 d8d4 f3f7 e8d8 f7h5 f8b4 e2e3 d4d5 f1c4 d5g2 e5f7 d8e7 f7h8
info depth 9 seldepth 16 time 630 nodes 1501 score cp 217 hashfull 94 nps 828 tbhits 0 pv f3e5 e7e6 b3f3 d8d4 f3f7 e8d8 f7h5 f8b4 e2e3 d4d5 f1c4 d5g2 e5f7 d8e7 f7h8
info depth 9 seldepth 17 time 760 nodes 1574 score cp 215 hashfull 94 nps 782 tbhits 0 pv f3e5 e7e6 b3f3 d8d4 f3f7 e8d8 f7h5 d8c8 e2e3 d4d5 f1c4 d5g2 c4e6 c8b7 e5f7 f8b4 f7h8
info depth 9 seldepth 17 time 931 nodes 1759 score cp 213 hashfull 95 nps 837 tbhits 0 pv f3e5 e7e6 b3f3 d8d4 f3f7 e8d8 f7h5 d8c8 e2e3 d4d5 f1c4 d5g2 c4e6 c8b7 e5f7 f8b4 f7h8
bestmove f3e5 ponder e7e6
position startpos moves d2d4 d7d5 c2c4 g8f6 c4d5 f6d5 g1f3 c8f5 b1d2 f5g6 h2h4 d5f6 d1b3 b7b6 h4h5 g6h5 h1h5 f6h5 f3e5 e7e6
go nodes 3000
info depth 7 seldepth 12 time 105 nodes 1507 score cp 216 hashfull 96 nps 247 tbhits 0 pv b3f3 d8d4 f3f7 e8d8 f7h5 d8c8 e2e3 d4d5 f1c4 d5g2 c4e6 c8b7 e5f7 f8b4 f7h8
info depth 8 seldepth 13 time 291 nodes 1727 score cp 206 hashfull 97 nps 845 tbhits 0 pv b3f3 d8d4 f3f7 e8d8 f7h5 d8c8 e2e3 d4d5 f1c4 d5g2 c4e6 c8b7 e5f7 f8b4 f7h8
info depth 8 seldepth 13 time 308 nodes 1733 score cp 207 hashfull 97 nps 818 tbhits 0 pv b3f3 d8d4 f3f7 e8d8 f7h5 d8c8 e2e3 d4d5 f1c4 d5g2 c4e6 c8b7 e5f7 f8b4 f7h8
bestmove b3f3 ponder d8d4

Re: Lc0 - macOS binary requested

Posted: Sun May 26, 2019 5:51 am
by MikeB
smcracraft wrote: Sun May 26, 2019 5:34 am
MikeB wrote: Sun May 26, 2019 2:48 am
smcracraft wrote: Sat May 25, 2019 11:38 pm Also it crashes on the mac.

Stuarts-MacBook-Pro:release stuartcracraft$ ./lc0
_
| _ | |
|_ |_ |_| v0.22.0-dev built May 25 2019
ucinewgame
Found pb network file: ./weights_run1_42372.pb.gz
Creating backend [opencl]...
OpenCL, maximum batch size set to 16.
Initializing OpenCL.
Detected 1 OpenCL platforms.
Platform version: OpenCL 1.2 (Feb 22 2019 20:16:07)
Platform profile: FULL_PROFILE
Platform name: Apple
Platform vendor: Apple
Device ID: 0
Device name: Intel(R) Core(TM) i7-7920HQ CPU @ 3.10GHz
Device type: CPU
Device vendor: Intel
Device driver: 1.1
Device speed: 3100 MHZ
Device cores: 8 CU
Device score: 512
Device ID: 1
Device name: Intel(R) HD Graphics 630
Device type: GPU
Device vendor: Intel Inc.
Device driver: 1.2(Mar 11 2019 21:25:40)
Device speed: 1100 MHZ
Device cores: 24 CU
Device score: 612
Device ID: 2
Device name: AMD Radeon Pro 560 Compute Engine
Device type: GPU
Device vendor: AMD
Device driver: 1.2 (Mar 11 2019 21:14:43)
Device speed: 300 MHZ
Device cores: 16 CU
Device score: 1112
Selected platform: Apple
Selected device: AMD Radeon Pro 560 Compute Engine
with OpenCL 1.2 capability.
Loaded existing SGEMM tuning for batch size 16.
Wavefront/Warp size: 64

Max workgroup size: 256
Max workgroup dimensions: 256 256 256
go nodes 3000
info depth 1 seldepth 2 time 209 nodes 4 score cp 17 hashfull 0 nps 19 tbhits 0 pv e2e4 e7e5
info depth 2 seldepth 3 time 361 nodes 8 score cp 10 hashfull 0 nps 22 tbhits 0 pv g1f3 d7d5 d2d4
info depth 2 seldepth 4 time 705 nodes 10 score cp -71 hashfull 0 nps 14 tbhits 0 pv g1f3 e7e6 c2c4 f8e7
info depth 3 seldepth 4 time 706 nodes 19 score cp -201 hashfull 0 nps 26 tbhits 0 pv g1f3 e7e6 d2d4 h7h6
info depth 3 seldepth 5 time 1039 nodes 34 score cp 130 hashfull 0 nps 32 tbhits 0 pv e2e4 e7e5 b1c3 a7a6 c3d5
info depth 3 seldepth 5 time 1524 nodes 86 score cp -4 hashfull 1 nps 56 tbhits 0 pv e2e3 g8f6 g1f3 h8g8
info depth 3 seldepth 6 time 1879 nodes 127 score cp -71 hashfull 1 nps 67 tbhits 0 pv e2e3 g8f6 g1f3 h8g8
info depth 4 seldepth 7 time 2673 nodes 236 score cp 26 hashfull 2 nps 88 tbhits 0 pv e2e4 c7c6 c2c4 d8c7 g2g3 c6c5
info depth 4 seldepth 7 time 4609 nodes 566 score cp 13 hashfull 4 nps 122 tbhits 0 pv b1c3 d7d5 a1b1 f7f5 a2a3 h7h6
info depth 4 seldepth 7 time 6259 nodes 753 score cp 124 hashfull 4 nps 120 tbhits 0 pv f2f4 a7a5 h2h4 c7c5 a2a3
info depth 5 seldepth 8 time 10000 nodes 1470 score cp 59 hashfull 8 nps 147 tbhits 0 pv f2f4 e7e6 c2c3 h7h6 a2a3 a7a6 h2h3
Segmentation fault: 11
Stuarts-MacBook-Pro:release stuartcracraft$
I was able to duplicate your error:

Code: Select all

Mac-Pro:release michaelbyrne$ lc0
       _
|   _ | |
|_ |_ |_| v0.22.0-dev built May 25 2019

uci
id name Lc0 v0.22.0-dev
id author The LCZero Authors.
option name WeightsFile type string default <autodiscover>
option name Backend type combo default opencl var opencl var blas var check var random var roundrobin var multiplexing var demux
option name BackendOptions type string default 
option name Threads type spin default 2 min 1 max 128
option name NNCacheSize type spin default 200000 min 0 max 999999999
option name MinibatchSize type spin default 256 min 1 max 1024
option name MaxPrefetch type spin default 32 min 0 max 1024
option name CPuct type string default 3.000000
option name CPuctBase type string default 19652.000000
option name CPuctFactor type string default 2.000000
option name Temperature type string default 0.000000
option name TempDecayMoves type spin default 0 min 0 max 100
option name TempCutoffMove type spin default 0 min 0 max 1000
option name TempEndgame type string default 0.000000
option name TempValueCutoff type string default 100.000000
option name TempVisitOffset type string default 0.000000
option name DirichletNoise type check default false
option name VerboseMoveStats type check default false
option name SmartPruningFactor type string default 1.330000
option name FpuStrategy type combo default reduction var reduction var absolute
option name FpuValue type string default 1.200000
option name FpuStrategyAtRoot type combo default same var reduction var absolute var same
option name FpuValueAtRoot type string default 1.000000
option name CacheHistoryLength type spin default 0 min 0 max 7
option name PolicyTemperature type string default 2.200000
option name MaxCollisionEvents type spin default 32 min 1 max 1024
option name MaxCollisionVisits type spin default 9999 min 1 max 1000000
option name OutOfOrderEval type check default true
option name StickyEndgames type check default true
option name SyzygyFastPlay type check default true
option name MultiPV type spin default 1 min 1 max 500
option name ScoreType type combo default centipawn var centipawn var centipawn_2018 var win_percentage var Q
option name HistoryFill type combo default fen_only var no var fen_only var always
option name KLDGainAverageInterval type spin default 100 min 1 max 10000000
option name MinimumKLDGainPerNode type string default 0.000000
option name Slowmover type string default 1.000000
option name MoveOverheadMs type spin default 200 min 0 max 100000000
option name SyzygyPath type string default 
option name Ponder type check default true
option name ImmediateTimeUse type string default 1.000000
option name RamLimitMb type spin default 0 min 0 max 100000000
option name ConfigFile type string default lc0.config
option name LogFile type string default 
uciok
ucinewgame
Found pb network file: ./weights_run1_42372.pb.gz
Creating backend [opencl]...
OpenCL, maximum batch size set to 16.
Initializing OpenCL.
Detected 1 OpenCL platforms.
Platform version: OpenCL 1.2 (Apr  7 2019 18:38:19)
Platform profile: FULL_PROFILE
Platform name:    Apple
Platform vendor:  Apple
Device ID:      0
Device name:    Intel(R) Xeon(R) CPU           X5690  @ 3.47GHz
Device type:    CPU
Device vendor:  Intel
Device driver:  1.1
Device speed:   3460 MHZ
Device cores:   24 CU
Device score:   512
Device ID:      1
Device name:    AMD Radeon HD 7950 Compute Engine
Device type:    GPU
Device vendor:  AMD
Device driver:  1.2 (Apr 29 2019 22:28:38)
Device speed:   150 MHZ
Device cores:   28 CU
Device score:   1112
Selected platform: Apple
Selected device: AMD Radeon HD 7950 Compute Engine
with OpenCL 1.2 capability.
Started OpenCL SGEMM tuner with batch size 16.
Will try 578 valid configurations.
(1/578) KWG=32 KWI=2 MDIMA=8 MDIMC=8 MWG=16 NDIMB=8 NDIMC=8 NWG=16 SA=0 SB=0 STRM=0 STRN=0 VWM=1 VWN=1 2234.1 us (240.3 GFLOPS)
(2/578) KWG=32 KWI=2 MDIMA=8 MDIMC=8 MWG=32 NDIMB=8 NDIMC=8 NWG=16 SA=0 SB=0 STRM=0 STRN=0 VWM=1 VWN=1 874.5 us (613.9 GFLOPS)
(4/578) KWG=32 KWI=2 MDIMA=8 MDIMC=8 MWG=16 NDIMB=8 NDIMC=8 NWG=32 SA=0 SB=0 STRM=0 STRN=0 VWM=1 VWN=1 867.5 us (618.9 GFLOPS)
(5/578) KWG=32 KWI=2 MDIMA=8 MDIMC=8 MWG=32 NDIMB=8 NDIMC=8 NWG=32 SA=0 SB=0 STRM=0 STRN=0 VWM=1 VWN=1 436.1 us (1231.0 GFLOPS)
(15/578) KWG=32 KWI=2 MDIMA=16 MDIMC=16 MWG=64 NDIMB=8 NDIMC=8 NWG=32 SA=0 SB=0 STRM=0 STRN=0 VWM=1 VWN=1 431.5 us (1244.2 GFLOPS)
(115/578) KWG=32 KWI=2 MDIMA=8 MDIMC=8 MWG=32 NDIMB=8 NDIMC=8 NWG=32 SA=0 SB=0 STRM=0 STRN=0 VWM=4 VWN=1 430.8 us (1246.1 GFLOPS)
(117/578) KWG=32 KWI=2 MDIMA=8 MDIMC=8 MWG=32 NDIMB=8 NDIMC=8 NWG=64 SA=0 SB=0 STRM=0 STRN=0 VWM=4 VWN=1 362.3 us (1482.0 GFLOPS)
(283/578) KWG=32 KWI=2 MDIMA=8 MDIMC=8 MWG=32 NDIMB=8 NDIMC=8 NWG=64 SA=0 SB=0 STRM=0 STRN=0 VWM=4 VWN=4 360.3 us (1489.9 GFLOPS)
(478/578) KWG=32 KWI=2 MDIMA=8 MDIMC=8 MWG=32 NDIMB=8 NDIMC=8 NWG=32 SA=1 SB=1 STRM=0 STRN=0 VWM=2 VWN=2 354.0 us (1516.5 GFLOPS)
Wavefront/Warp size: 64

Max workgroup size: 256
Max workgroup dimensions: 256 256 256
go nodes 3000
info depth 1 seldepth 2 time 514 nodes 4 score cp 17 hashfull 0 nps 7 tbhits 0 pv e2e4 e7e5
info depth 2 seldepth 3 time 591 nodes 8 score cp 10 hashfull 0 nps 13 tbhits 0 pv g1f3 d7d5 d2d4
info depth 3 seldepth 4 time 744 nodes 17 score cp 11 hashfull 0 nps 22 tbhits 0 pv g1f3 d7d5 d2d4 g8f6
info depth 3 seldepth 5 time 834 nodes 34 score cp 20 hashfull 0 nps 40 tbhits 0 pv g1f3 e7e6 d2d4 d7d5 f3g5
info depth 3 seldepth 6 time 1000 nodes 38 score cp 20 hashfull 0 nps 38 tbhits 0 pv g1f3 e7e6 d2d4 d7d5 f3g5
info depth 4 seldepth 6 time 1067 nodes 94 score cp -131 hashfull 1 nps 88 tbhits 0 pv e2e3 b7b6 e3e4 h7h6
info depth 4 seldepth 6 time 1388 nodes 162 score cp -5 hashfull 1 nps 116 tbhits 0 pv g1f3 e7e6 d2d4 d7d5 e2e3 h7h6
info depth 4 seldepth 7 time 1564 nodes 227 score cp -68 hashfull 2 nps 145 tbhits 0 pv g1f3 e7e6 d2d4 d7d5 e2e3 h7h6
info depth 4 seldepth 7 time 1847 nodes 328 score cp -75 hashfull 2 nps 177 tbhits 0 pv c2c3 e7e5 g2g3 a7a6 h2h3
info depth 4 seldepth 8 time 2069 nodes 362 score cp -75 hashfull 2 nps 174 tbhits 0 pv c2c3 e7e5 g2g3 a7a6 h2h3
info depth 4 seldepth 9 time 2515 nodes 462 score cp -75 hashfull 3 nps 183 tbhits 0 pv c2c3 e7e5 g2g3 a7a6 h2h3
info depth 4 seldepth 9 time 3168 nodes 644 score cp 50 hashfull 4 nps 203 tbhits 0 pv e2e4 e7e5 g1f3 b8c6 d1e2 h7h6 b1a3 a7a6 c2c3
info depth 4 seldepth 9 time 6636 nodes 1720 score cp 59 hashfull 11 nps 259 tbhits 0 pv a2a4 e7e6 b1c3 g8f6 a4a5
info depth 5 seldepth 9 time 7534 nodes 1976 score cp 35 hashfull 12 nps 262 tbhits 0 pv a2a4 g7g6 e2e4 a7a5 f1c4 b7b6 h2h3
info depth 5 seldepth 9 time 10873 nodes 2904 score cp -2147483648 hashfull 18 nps 267 tbhits 0 pv e2e4 a7a6 b2b3 e7e6 d1e2 h7h6 f2f3 h6h5
Segmentation fault: 11
My initial thought that there was something wrong with the NN you downloaded , so I downloaded the NN you used and tried it. There were no issues.

Code: Select all

Mac-Pro:release michaelbyrne$ lc0
       _
|   _ | |
|_ |_ |_| v0.22.0-dev built May 25 2019
ucinewgame
Found pb network file: ./weights_run1_42372.pb.gz
Creating backend [opencl]...
OpenCL, maximum batch size set to 16.
Initializing OpenCL.
Detected 1 OpenCL platforms.
Platform version: OpenCL 1.2 (Apr  7 2019 18:38:19)
Platform profile: FULL_PROFILE
Platform name:    Apple
Platform vendor:  Apple
Device ID:      0
Device name:    Intel(R) Xeon(R) CPU           X5690  @ 3.47GHz
Device type:    CPU
Device vendor:  Intel
Device driver:  1.1
Device speed:   3460 MHZ
Device cores:   24 CU
Device score:   512
Device ID:      1
Device name:    AMD Radeon HD 7950 Compute Engine
Device type:    GPU
Device vendor:  AMD
Device driver:  1.2 (Apr 29 2019 22:28:38)
Device speed:   150 MHZ
Device cores:   28 CU
Device score:   1112
Selected platform: Apple
Selected device: AMD Radeon HD 7950 Compute Engine
with OpenCL 1.2 capability.
Loaded existing SGEMM tuning for batch size 16.
Wavefront/Warp size: 64

Max workgroup size: 256
Max workgroup dimensions: 256 256 256
go nodes 3000
info depth 1 seldepth 2 time 611 nodes 4 score cp 17 hashfull 0 nps 6 tbhits 0 pv e2e4 e7e5
info depth 2 seldepth 3 time 722 nodes 8 score cp 10 hashfull 0 nps 11 tbhits 0 pv g1f3 d7d5 d2d4
info depth 3 seldepth 4 time 886 nodes 17 score cp 11 hashfull 0 nps 19 tbhits 0 pv g1f3 d7d5 d2d4 g8f6
info depth 3 seldepth 5 time 1050 nodes 25 score cp 9 hashfull 0 nps 23 tbhits 0 pv g1f3 d7d5 g2g3 c7c5 f1g2
info depth 4 seldepth 6 time 1203 nodes 63 score cp 10 hashfull 1 nps 52 tbhits 0 pv g1f3 d7d5 g2g3 c7c5 f1g2
info depth 4 seldepth 7 time 1374 nodes 82 score cp 10 hashfull 1 nps 59 tbhits 0 pv g1f3 d7d5 g2g3 e7e6 f1g2 c7c5
info depth 4 seldepth 8 time 1539 nodes 111 score cp 15 hashfull 1 nps 72 tbhits 0 pv e2e4 e7e5 g1f3 b8c6 b1c3 g8f6 f1b5 f8d6
info depth 4 seldepth 8 time 1641 nodes 132 score cp 11 hashfull 1 nps 80 tbhits 0 pv d2d4 g8f6 c2c4 e7e6 b1c3 f8b4 e2e3 e8g8 d4d5
info depth 5 seldepth 9 time 1706 nodes 146 score cp 16 hashfull 1 nps 85 tbhits 0 pv e2e4 e7e5 g1f3 b8c6 b1c3 g8f6 f1b5 f8b4 e1g1
info depth 5 seldepth 10 time 1872 nodes 205 score cp 16 hashfull 1 nps 109 tbhits 0 pv e2e4 e7e5 g1f3 b8c6 f1b5 g8f6 b1c3 f8d6 e1g1 e8g8
info depth 5 seldepth 11 time 2036 nodes 252 score cp 16 hashfull 2 nps 123 tbhits 0 pv e2e4 e7e5 g1f3 b8c6 b1c3 g8f6 f1b5 f8d6 e1g1 e8g8 b5a4
info depth 6 seldepth 11 time 2037 nodes 267 score cp 16 hashfull 2 nps 131 tbhits 0 pv e2e4 e7e5 g1f3 b8c6 b1c3 g8f6 f1b5 f8d6 e1g1 e8g8 b5a4
info depth 6 seldepth 12 time 2201 nodes 311 score cp 16 hashfull 2 nps 141 tbhits 0 pv e2e4 e7e5 b1c3 g8f6 g1f3 b8c6 f1b5 f8d6 e1g1 e8g8 b5a4 h7h6
info depth 6 seldepth 13 time 2343 nodes 398 score cp 16 hashfull 2 nps 169 tbhits 0 pv e2e4 e7e5 b1c3 g8f6 g1f3 b8c6 f1b5 f8d6 d2d3 e8g8 b5a4 f8e8 e1g1
info depth 6 seldepth 14 time 2506 nodes 463 score cp 17 hashfull 2 nps 184 tbhits 0 pv e2e4 e7e5 b1c3 g8f6 g1f3 b8c6 f1b5 f8d6 d2d3 e8g8 b5a4 f8e8 e1g1 h7h6
info depth 7 seldepth 15 time 2782 nodes 620 score cp 18 hashfull 3 nps 222 tbhits 0 pv e2e4 e7e5 b1c3 g8f6 g1f3 b8c6 f1b5 f8d6 d2d3 e8g8 b5a4 f8e8 e1g1 h7h6 a2a3
info depth 7 seldepth 16 time 3000 nodes 720 score cp 18 hashfull 3 nps 240 tbhits 0 pv e2e4 e7e5 b1c3 g8f6 g1f3 b8c6 f1b5 f8d6 d2d3 e8g8 b5a4 f8e8 e1g1 h7h6 a2a3 h6h5
info depth 7 seldepth 17 time 3436 nodes 912 score cp 18 hashfull 4 nps 265 tbhits 0 pv e2e4 e7e5 b1c3 g8f6 g1f3 b8c6 f1b5 f8d6 d2d3 e8g8 b5a4 f8e8 e1g1 h7h6 a2a3 a7a6 a4c6
info depth 7 seldepth 18 time 3605 nodes 1041 score cp 19 hashfull 4 nps 288 tbhits 0 pv e2e4 e7e5 b1c3 g8f6 g1f3 b8c6 f1b5 f8d6 d2d3 e8g8 b5a4 f8e8 e1g1 h7h6 a2a3 a7a6 h2h3
info depth 7 seldepth 19 time 3901 nodes 1089 score cp 19 hashfull 5 nps 279 tbhits 0 pv e2e4 e7e5 b1c3 g8f6 g1f3 b8c6 f1b5 f8d6 d2d3 e8g8 b5a4 f8e8 e1g1 h7h6 a2a3 a7a6 h2h3
info depth 7 seldepth 20 time 4232 nodes 1268 score cp 19 hashfull 5 nps 299 tbhits 0 pv e2e4 e7e5 b1c3 g8f6 g1f3 b8c6 f1b5 f8d6 d2d3 e8g8 b5a4 f8e8 e1g1 h7h6 a2a3 a7a6 h2h3
info depth 8 seldepth 20 time 4619 nodes 1619 score cp 19 hashfull 6 nps 350 tbhits 0 pv e2e4 e7e5 b1c3 g8f6 g1f3 b8c6 f1b5 f8d6 d2d3 e8g8 b5a4 f8e8 e1g1 h7h6 a2a3 a7a6 h2h3 d6f8
info depth 8 seldepth 21 time 5399 nodes 1921 score cp 19 hashfull 8 nps 355 tbhits 0 pv e2e4 e7e5 b1c3 g8f6 g1f3 b8c6 f1b5 f8d6 d2d3 e8g8 b5a4 f8e8 e1g1 h7h6 a2a3 a7a6 h2h3 d6f8 a4b3
info depth 8 seldepth 22 time 6792 nodes 2800 score cp 17 hashfull 10 nps 412 tbhits 0 pv e2e4 e7e5 b1c3 g8f6 g1f3 b8c6 f1b5 f8d6 d2d3 e8g8 b5a4 f8e8 e1g1 h7h6 a2a3 a7a6 h2h3 d6f8 a4b3
info depth 8 seldepth 22 time 7265 nodes 3052 score cp 16 hashfull 11 nps 420 tbhits 0 pv e2e4 e7e5 b1c3 g8f6 g1f3 b8c6 f1b5 f8d6 d2d3 e8g8 b5a4 f8e8 e1g1 h7h6 a2a3 a7a6 h2h3 d6f8 a4b3
bestmove e2e4 ponder e7e5
quit
SoI took this one step further, deleted the leelaz_opencl_tuning, and used the net you were using to rebuild the leelaz_opencl_tuning file. Bingo, I got the seg 11 error. I deleted the weight file you were using, and the leelaz_opencl_tuning file. I put my weights file back.

Code: Select all

Mac-Pro:release michaelbyrne$ lc0
       _
|   _ | |
|_ |_ |_| v0.22.0-dev built May 25 2019
ucinewgame
Found pb network file: ./weights
Creating backend [opencl]...
OpenCL, maximum batch size set to 16.
Initializing OpenCL.
Detected 1 OpenCL platforms.
Platform version: OpenCL 1.2 (Apr  7 2019 18:38:19)
Platform profile: FULL_PROFILE
Platform name:    Apple
Platform vendor:  Apple
Device ID:      0
Device name:    Intel(R) Xeon(R) CPU           X5690  @ 3.47GHz
Device type:    CPU
Device vendor:  Intel
Device driver:  1.1
Device speed:   3460 MHZ
Device cores:   24 CU
Device score:   512
Device ID:      1
Device name:    AMD Radeon HD 7950 Compute Engine
Device type:    GPU
Device vendor:  AMD
Device driver:  1.2 (Apr 29 2019 22:28:38)
Device speed:   150 MHZ
Device cores:   28 CU
Device score:   1112
Selected platform: Apple
Selected device: AMD Radeon HD 7950 Compute Engine
with OpenCL 1.2 capability.
Started OpenCL SGEMM tuner with batch size 16.
Will try 578 valid configurations.
(2/578) KWG=32 KWI=2 MDIMA=8 MDIMC=8 MWG=32 NDIMB=8 NDIMC=8 NWG=16 SA=0 SB=0 STRM=0 STRN=0 VWM=1 VWN=1 249.9 us (411.1 GFLOPS)
(5/578) KWG=32 KWI=2 MDIMA=8 MDIMC=8 MWG=32 NDIMB=8 NDIMC=8 NWG=32 SA=0 SB=0 STRM=0 STRN=0 VWM=1 VWN=1 143.7 us (715.2 GFLOPS)
(15/578) KWG=32 KWI=2 MDIMA=16 MDIMC=16 MWG=64 NDIMB=8 NDIMC=8 NWG=32 SA=0 SB=0 STRM=0 STRN=0 VWM=1 VWN=1 141.1 us (728.2 GFLOPS)
(42/578) KWG=32 KWI=2 MDIMA=16 MDIMC=16 MWG=64 NDIMB=16 NDIMC=16 NWG=64 SA=0 SB=0 STRM=0 STRN=0 VWM=1 VWN=1 138.5 us (741.8 GFLOPS)
(115/578) KWG=32 KWI=2 MDIMA=8 MDIMC=8 MWG=32 NDIMB=8 NDIMC=8 NWG=32 SA=0 SB=0 STRM=0 STRN=0 VWM=4 VWN=1 126.5 us (812.1 GFLOPS)
(117/578) KWG=32 KWI=2 MDIMA=8 MDIMC=8 MWG=32 NDIMB=8 NDIMC=8 NWG=64 SA=0 SB=0 STRM=0 STRN=0 VWM=4 VWN=1 122.1 us (841.6 GFLOPS)
(121/578) KWG=32 KWI=2 MDIMA=16 MDIMC=16 MWG=64 NDIMB=8 NDIMC=8 NWG=64 SA=0 SB=0 STRM=0 STRN=0 VWM=4 VWN=1 121.2 us (847.8 GFLOPS)
(376/578) KWG=32 KWI=2 MDIMA=8 MDIMC=8 MWG=32 NDIMB=16 NDIMC=16 NWG=32 SA=1 SB=1 STRM=0 STRN=0 VWM=2 VWN=1 116.3 us (883.6 GFLOPS)
(485/578) KWG=32 KWI=2 MDIMA=16 MDIMC=16 MWG=32 NDIMB=8 NDIMC=8 NWG=32 SA=1 SB=1 STRM=0 STRN=0 VWM=2 VWN=2 111.1 us (925.2 GFLOPS)
Wavefront/Warp size: 64

Max workgroup size: 256
Max workgroup dimensions: 256 256 256
go nodes 3000     
info depth 1 seldepth 2 time 190 nodes 7 score cp 8 hashfull 0 nps 36 tbhits 0 pv d2d4 g8f6
info depth 2 seldepth 3 time 226 nodes 16 score cp 5 hashfull 0 nps 70 tbhits 0 pv g2g3 d7d5 g1f3
info depth 3 seldepth 4 time 262 nodes 35 score cp 6 hashfull 0 nps 133 tbhits 0 pv g2g3 e7e6 c2c4 d7d5
info depth 3 seldepth 5 time 302 nodes 57 score cp 5 hashfull 0 nps 188 tbhits 0 pv g2g3 e7e6 c2c4 d7d5 f1g2
info depth 3 seldepth 5 time 305 nodes 59 score cp 5 hashfull 0 nps 193 tbhits 0 pv c2c4 e7e5 d2d3 b8c6 g1f3
info depth 4 seldepth 5 time 324 nodes 67 score cp 5 hashfull 0 nps 206 tbhits 0 pv g2g3 d7d5 g1f3 c7c5 f1g2 d5d4
info depth 4 seldepth 6 time 350 nodes 92 score cp 6 hashfull 0 nps 262 tbhits 0 pv g2g3 e7e6 c2c4 d7d5 f1g2 d5d4
info depth 4 seldepth 7 time 395 nodes 128 score cp 6 hashfull 1 nps 324 tbhits 0 pv c2c4 e7e5 d2d3 f8e7 g1f3 b8c6
info depth 4 seldepth 8 time 445 nodes 179 score cp 6 hashfull 1 nps 402 tbhits 0 pv c2c4 e7e5 d2d3 f8e7 g1f3 b8c6
info depth 4 seldepth 8 time 494 nodes 235 score cp 7 hashfull 1 nps 475 tbhits 0 pv g1f3 d7d5 d2d4 g8f6 c1f4 c8f5 e2e3 e7e6
info depth 4 seldepth 9 time 520 nodes 260 score cp 7 hashfull 1 nps 500 tbhits 0 pv g1f3 d7d5 d2d4 g8f6 c1f4 c8f5 e2e3 e7e6
info depth 5 seldepth 9 time 568 nodes 306 score cp 6 hashfull 2 nps 538 tbhits 0 pv g1f3 d7d5 d2d4 g8f6 c1f4 c8f5 c2c4 e7e6 b1c3
info depth 5 seldepth 10 time 640 nodes 415 score cp 6 hashfull 2 nps 648 tbhits 0 pv g1f3 d7d5 d2d4 g8f6 c1f4 c7c5 e2e3 c5d4 e3d4
info depth 5 seldepth 11 time 763 nodes 537 score cp 7 hashfull 3 nps 703 tbhits 0 pv g1f3 d7d5 d2d4 g8f6 c1f4 c7c5 e2e3 c5d4 e3d4
info depth 5 seldepth 12 time 873 nodes 747 score cp 7 hashfull 3 nps 855 tbhits 0 pv g1f3 d7d5 d2d4 g8f6 c1f4 c7c5 e2e3 c5d4 e3d4
info depth 5 seldepth 13 time 1021 nodes 954 score cp 8 hashfull 4 nps 934 tbhits 0 pv g1f3 d7d5 d2d4 g8f6 c1f4 c7c5 e2e3 c5d4 e3d4
info depth 6 seldepth 13 time 1154 nodes 1143 score cp 8 hashfull 5 nps 990 tbhits 0 pv g1f3 d7d5 d2d4 g8f6 c1f4 c7c5 e2e3 c5d4 e3d4 b8c6
info depth 6 seldepth 14 time 1390 nodes 1483 score cp 9 hashfull 6 nps 1066 tbhits 0 pv g1f3 d7d5 d2d4 g8f6 c1f4 c7c5 e2e3 c5d4 e3d4 b8c6
info depth 6 seldepth 14 time 1820 nodes 2205 score cp 13 hashfull 9 nps 1211 tbhits 0 pv d2d4 g8f6 c2c4 e7e6 g2g3 f8b4 b1d2 d7d5 f1g2 e8g8 g1f3
info depth 6 seldepth 14 time 1919 nodes 2291 score cp 9 hashfull 9 nps 1193 tbhits 0 pv g1f3 d7d5 d2d4 g8f6 c1f4 c7c5 e2e3 c5d4 e3d4 b8c6 c2c3
info depth 6 seldepth 14 time 1943 nodes 2402 score cp 13 hashfull 10 nps 1236 tbhits 0 pv d2d4 g8f6 c2c4 e7e6 g2g3 f8b4 b1d2 d7d5 f1g2 e8g8 g1f3
info depth 6 seldepth 15 time 2265 nodes 2951 score cp 12 hashfull 12 nps 1302 tbhits 0 pv d2d4 g8f6 c2c4 e7e6 g2g3 f8b4 b1d2 d7d5 f1g2 e8g8 g1f3 d5c4
info depth 6 seldepth 15 time 2347 nodes 3003 score cp 12 hashfull 12 nps 1279 tbhits 0 pv d2d4 g8f6 c2c4 e7e6 g2g3 f8b4 b1d2 d7d5 f1g2 e8g8 g1f3 d5c4 d4d5
bestmove d2d4 ponder g8f6
Everything works - you still use the weights file if you want, but you can't tune Lc0 with it, it's creating a bad tuning file. So move your weights file to a different folder and place this one there instead.

https://www.dropbox.com/s/6iy7cqeu89ijl34/weights?dl=1
I'm not even sure which it was at this point but it work great ,and you will get much better NPS.

Delete your current leelaz_opencl_tuning file and follow the steps you did before when you first ran LcO.

Also, dkappe has the best nets for those with less than high-end hardware - your nps will increase by 10X easily or more and you will be much more satisfied with lc0.

https://github.com/dkappe/leela-chess-w ... d-Networks

Here's the crush of me by the new version per your inputs.

Stuarts-MacBook-Pro:release stuartcracraft$ ./lc0
_
| _ | |
|_ |_ |_| v0.22.0-dev built May 25 2019
ucinewgame
Found pb network file: ./weights
Creating backend [opencl]...
OpenCL, maximum batch size set to 16.
Initializing OpenCL.
Detected 1 OpenCL platforms.
Platform version: OpenCL 1.2 (Feb 22 2019 20:16:07)
Platform profile: FULL_PROFILE
Platform name: Apple
Platform vendor: Apple
Device ID: 0
Device name: Intel(R) Core(TM) i7-7920HQ CPU @ 3.10GHz
Device type: CPU
Device vendor: Intel
Device driver: 1.1
Device speed: 3100 MHZ
Device cores: 8 CU
Device score: 512
Device ID: 1
Device name: Intel(R) HD Graphics 630
Device type: GPU
Device vendor: Intel Inc.
Device driver: 1.2(Mar 11 2019 21:25:40)
Device speed: 1100 MHZ
Device cores: 24 CU
Device score: 612
Device ID: 2
Device name: AMD Radeon Pro 560 Compute Engine
Device type: GPU
Device vendor: AMD
Device driver: 1.2 (Mar 11 2019 21:14:43)
Device speed: 300 MHZ
Device cores: 16 CU
Device score: 1112
Selected platform: Apple
Selected device: AMD Radeon Pro 560 Compute Engine
with OpenCL 1.2 capability.
Loaded existing SGEMM tuning for batch size 16.
Wavefront/Warp size: 64

Max workgroup size: 256
Max workgroup dimensions: 256 256 256
go nodes 3000
info depth 1 seldepth 2 time 116 nodes 7 score cp 8 hashfull 0 nps 60 tbhits 0 pv d2d4 g8f6
info depth 2 seldepth 3 time 151 nodes 16 score cp 5 hashfull 0 nps 105 tbhits 0 pv g2g3 d7d5 g1f3
info depth 3 seldepth 4 time 189 nodes 35 score cp 6 hashfull 0 nps 185 tbhits 0 pv g2g3 e7e6 c2c4 d7d5
info depth 4 seldepth 5 time 258 nodes 65 score cp 5 hashfull 0 nps 251 tbhits 0 pv g2g3 d7d5 g1f3 c7c5 f1g2
info depth 4 seldepth 6 time 318 nodes 79 score cp 5 hashfull 0 nps 248 tbhits 0 pv c2c4 e7e5 d2d3 g8f6 g1f3
info depth 4 seldepth 6 time 318 nodes 100 score cp 6 hashfull 0 nps 314 tbhits 0 pv g2g3 e7e6 c2c4 d7d5 f1g2 d5d4
info depth 4 seldepth 7 time 375 nodes 118 score cp 6 hashfull 1 nps 314 tbhits 0 pv c2c4 e7e5 d2d3 f8e7 g1f3 b8c6
info depth 4 seldepth 8 time 441 nodes 165 score cp 6 hashfull 1 nps 374 tbhits 0 pv c2c4 e7e5 d2d3 f8e7 g1f3 b8c6
info depth 4 seldepth 9 time 505 nodes 226 score cp 7 hashfull 1 nps 447 tbhits 0 pv c2c4 e7e5 g2g3 b8c6 f1g2 g7g6 b1c3 d7d6 c4c5
info depth 4 seldepth 9 time 527 nodes 263 score cp 7 hashfull 1 nps 499 tbhits 0 pv g1f3 d7d5 d2d4 g8f6 c1f4 c8f5 e2e3 e7e6 f4c7
info depth 5 seldepth 9 time 582 nodes 299 score cp 7 hashfull 1 nps 513 tbhits 0 pv c2c4 e7e5 g2g3 b8c6 f1g2 g7g6 b1c3 d7d6 d2d3
info depth 5 seldepth 9 time 583 nodes 335 score cp 6 hashfull 2 nps 574 tbhits 0 pv g1f3 d7d5 d2d4 g8f6 c1f4 c8f5 e2e3 e7e6 f1d3
info depth 5 seldepth 9 time 648 nodes 364 score cp 7 hashfull 2 nps 561 tbhits 0 pv c2c4 e7e5 g2g3 b8c6 f1g2 g7g6 b1c3 d7d6 d2d3
info depth 5 seldepth 9 time 648 nodes 388 score cp 7 hashfull 2 nps 598 tbhits 0 pv g1f3 d7d5 d2d4 g8f6 c1f4 c8f5 e2e3 e7e6 f1d3
info depth 5 seldepth 10 time 736 nodes 444 score cp 6 hashfull 2 nps 603 tbhits 0 pv g1f3 d7d5 d2d4 g8f6 c1f4 c8f5 e2e3 e7e6 f1d3
info depth 5 seldepth 11 time 815 nodes 531 score cp 6 hashfull 2 nps 651 tbhits 0 pv g1f3 d7d5 d2d4 g8f6 c1f4 c8f5 e2e3 e7e6 f1d3
info depth 5 seldepth 12 time 937 nodes 649 score cp 7 hashfull 3 nps 692 tbhits 0 pv g1f3 d7d5 d2d4 g8f6 c1f4 c8f5 e2e3 e7e6 f1d3
info depth 5 seldepth 13 time 1092 nodes 809 score cp 7 hashfull 4 nps 740 tbhits 0 pv g1f3 d7d5 d2d4 g8f6 c1f4 c8f5 e2e3 e7e6 f1d3
info depth 6 seldepth 13 time 1446 nodes 1223 score cp 9 hashfull 5 nps 845 tbhits 0 pv g1f3 d7d5 d2d4 g8f6 c1f4 c8f5 e2e3 e7e6 f1d3
info depth 6 seldepth 14 time 1620 nodes 1486 score cp 8 hashfull 6 nps 917 tbhits 0 pv g1f3 d7d5 d2d4 g8f6 c1f4 c8f5 c2c4 e7e6 b1c3 f8b4
info depth 6 seldepth 14 time 2135 nodes 2121 score cp 12 hashfull 9 nps 993 tbhits 0 pv d2d4 g8f6 c2c4 e7e6 g2g3 f8b4 b1d2 d7d5 f1g2 e8g8 g1f3 d5c4
info depth 6 seldepth 15 time 2459 nodes 2608 score cp 13 hashfull 10 nps 1060 tbhits 0 pv d2d4 g8f6 c2c4 e7e6 g2g3 f8b4 b1d2 d7d5 f1g2 e8g8 g1f3 d5c4
info depth 6 seldepth 15 time 2885 nodes 3073 score cp 12 hashfull 12 nps 1065 tbhits 0 pv d2d4 g8f6 c2c4 e7e6 g2g3 f8b4 b1d2 d7d5 f1g2 e8g8 g1f3 d5c4 d4d5
bestmove d2d4 ponder g8f6
position startpos moves d2d4 d7d5
go nodes 3000
info depth 6 seldepth 11 time 104 nodes 115 score cp 9 hashfull 13 nps 451 tbhits 0 pv c2c4 e7e6 b1c3 g8f6 c4d5 e6d5 c1g5 c7c6 e2e3 h7h6 g5h4 c8f5 d1f3 f5b1
info depth 6 seldepth 14 time 108 nodes 136 score cp 9 hashfull 13 nps 629 tbhits 0 pv c2c4 e7e6 b1c3 g8f6 c4d5 e6d5 c1g5 c7c6 e2e3 h7h6 g5h4 c8f5 f1d3 f5d3
info depth 6 seldepth 15 time 851 nodes 985 score cp 12 hashfull 16 nps 1077 tbhits 0 pv c2c4 e7e6 b1c3 g8f6 c4d5 e6d5 c1g5 c7c6 e2e3 h7h6 g5h4 c8f5 f1d3 f5d3 d1d3
info depth 6 seldepth 15 time 1948 nodes 2280 score cp 12 hashfull 21 nps 1135 tbhits 0 pv c2c4 e7e6 b1c3 g8f6 c4d5 e6d5 c1g5 c7c6 e2e3 h7h6 g5h4 c8f5 f1d3 f5d3 d1d3
bestmove c2c4 ponder e7e6
position startpos moves d2d4 d7d5 c2c4 g8f6
go nodes 3000
info depth 7 seldepth 10 time 102 nodes 35 score cp 20 hashfull 22 nps 88 tbhits 0 pv c4d5 c8f5 g1f3 f5b1 d1a4 c7c6 d5c6 b8c6 a1b1 f6g4
info depth 6 seldepth 10 time 110 nodes 83 score cp 22 hashfull 22 nps 518 tbhits 0 pv c4d5 c8f5 g1f3 f5b1 d1a4 c7c6 d5c6 b8c6 a1b1 e7e6
info depth 5 seldepth 10 time 205 nodes 130 score cp 24 hashfull 23 nps 507 tbhits 0 pv c4d5 c8f5 g1f3 f5b1 d1a4 c7c6 d5c6 b8c6 a1b1 e7e6
info depth 5 seldepth 11 time 294 nodes 224 score cp 25 hashfull 23 nps 673 tbhits 0 pv c4d5 c8f5 g1f3 f5b1 d1a4 c7c6 d5c6 b8c6 a1b1 e7e6
info depth 5 seldepth 13 time 410 nodes 343 score cp 25 hashfull 23 nps 773 tbhits 0 pv c4d5 c8f5 g1f3 f5b1 d1a4 c7c6 d5c6 b8c6 a1b1 e7e6
info depth 6 seldepth 13 time 449 nodes 387 score cp 26 hashfull 24 nps 804 tbhits 0 pv c4d5 c8f5 g1f3 f5b1 d1a4 c7c6 d5c6 b8c6 a1b1 e7e6
info depth 6 seldepth 14 time 764 nodes 788 score cp 28 hashfull 25 nps 997 tbhits 0 pv c4d5 c8f5 g1f3 f5b1 d1a4 c7c6 d5c6 b8c6 a1b1 e7e6
info depth 6 seldepth 15 time 905 nodes 949 score cp 28 hashfull 26 nps 1019 tbhits 0 pv c4d5 c8f5 g1f3 f5b1 d1a4 c7c6 d5c6 b8c6 a1b1 e7e6
info depth 6 seldepth 16 time 1027 nodes 1066 score cp 29 hashfull 26 nps 1012 tbhits 0 pv c4d5 g7g6 g1f3 f8g7 d1a4 c7c6 d5c6 b8c6 e2e3 e8g8 f1e2
info depth 6 seldepth 17 time 1147 nodes 1175 score cp 30 hashfull 27 nps 1001 tbhits 0 pv c4d5 c7c6 d5c6 b8c6 g1f3 e7e5 d4e5 d8d1 e1d1 f6g4 d1e1 g4e5 b1c3 c8e6 f3g5
info depth 6 seldepth 18 time 1423 nodes 1549 score cp 28 hashfull 28 nps 1070 tbhits 0 pv c4d5 g7g6 g1f3 f8g7 d1a4 c7c6 d5c6 b8c6 b1c3 e8g8 e2e3 e7e5
info depth 7 seldepth 18 time 1698 nodes 1806 score cp 27 hashfull 30 nps 1048 tbhits 0 pv c4d5 g7g6 g1f3 f8g7 d1a4 c7c6 d5c6 b8c6 b1c3 e8g8 e2e3 e7e5
info depth 7 seldepth 18 time 1843 nodes 1942 score cp 27 hashfull 30 nps 1039 tbhits 0 pv c4d5 c7c6 d5c6 b8c6 g1f3 e7e5 d4e5 d8d1 e1d1 f6g4 d1e1 g4e5 b1c3 c8e6 f3e5 c6e5
bestmove c4d5 ponder c7c6
position startpos moves d2d4 d7d5 c2c4 g8f6 c4d5 f6d5
go nodes 3000
info depth 6 seldepth 8 time 101 nodes 73 score cp 31 hashfull 31 nps 99 tbhits 0 pv g1f3 c8f5 d1b3 b8c6 b1c3 d5b4 d4d5
info depth 6 seldepth 9 time 109 nodes 93 score cp 33 hashfull 31 nps 275 tbhits 0 pv g1f3 c8f5 b1d2 d5b4 e2e4 f5g4 d1b3 e7e6 a2a3
info depth 6 seldepth 10 time 174 nodes 147 score cp 34 hashfull 31 nps 482 tbhits 0 pv g1f3 c8f5 b1d2 d5b4 e2e4 f5g4 d1b3 e7e6 a2a3
info depth 5 seldepth 10 time 294 nodes 231 score cp 34 hashfull 32 nps 571 tbhits 0 pv g1f3 c8f5 b1d2 d5b4 e2e4 f5g4 d1b3 e7e6 a2a3
info depth 5 seldepth 11 time 747 nodes 688 score cp 33 hashfull 34 nps 836 tbhits 0 pv g1f3 c8f5 b1d2 d5f6 d1b3 b8d7 b3b7 e7e6 e2e4
info depth 6 seldepth 11 time 880 nodes 847 score cp 34 hashfull 34 nps 890 tbhits 0 pv g1f3 c8f5 b1d2 d5f6 d1b3 d8d5 b3d5 f6d5 e2e4
info depth 6 seldepth 12 time 1030 nodes 1038 score cp 34 hashfull 35 nps 946 tbhits 0 pv g1f3 c8f5 b1d2 d5f6 d1b3 d8c8 f3e5 f5e6 b3c2 g7g6
info depth 6 seldepth 13 time 1438 nodes 1455 score cp 34 hashfull 37 nps 968 tbhits 0 pv g1f3 c8f5 b1d2 d5f6 d1b3 d8c8 f3e5 f5e6 b3c2 g7g6
info depth 6 seldepth 14 time 2289 nodes 2284 score cp 35 hashfull 40 nps 970 tbhits 0 pv g1f3 c8f5 b1d2 d5f6 d1b3 d8c8 e2e4 f6e4 d2e4 f5e4
info depth 6 seldepth 14 time 2352 nodes 2511 score cp 35 hashfull 41 nps 1040 tbhits 0 pv g1f3 c8f5 b1d2 d5f6 d1b3 d8c8 e2e4 f6e4 d2e4 f5e4
bestmove g1f3 ponder c8f5
position startpos moves d2d4 d7d5 c2c4 g8f6 c4d5 f6d5 g1f3 c8f5
go nodes 3000
info depth 7 seldepth 10 time 103 nodes 585 score cp 44 hashfull 42 nps 126 tbhits 0 pv b1d2 d5f6 d1b3 b8d7 b3b7 a8b8 b7a7 b8a8 a7b7 f5b1
info depth 6 seldepth 12 time 169 nodes 645 score cp 44 hashfull 42 nps 431 tbhits 0 pv b1d2 d5f6 d1b3 b8d7 b3b7 a8b8 b7a7 b8a8 a7b7 f5b1
info depth 6 seldepth 13 time 734 nodes 1173 score cp 42 hashfull 45 nps 818 tbhits 0 pv b1d2 d5f6 d1b3 d8c8 e2e4 f6e4 d2e4 f5e4 f3g5 e4g6
info depth 6 seldepth 14 time 977 nodes 1433 score cp 41 hashfull 46 nps 881 tbhits 0 pv b1d2 d5f6 d1b3 d8c8 e2e4 f6e4 d2e4 f5e4 f3g5 e4g6 f1c4
info depth 6 seldepth 14 time 1745 nodes 2268 score cp 39 hashfull 50 nps 971 tbhits 0 pv b1d2 d5f6 d1b3 d8c8 f3e5 f5e6 b3c2 g7g6 e2e4 f8g7 d4d5 e6d7 f1c4
bestmove b1d2 ponder d5f6
position startpos moves d2d4 d7d5 c2c4 g8f6 c4d5 f6d5 g1f3 c8f5 b1d2 f5g6
go nodes 3000
info depth 6 seldepth 9 time 104 nodes 69 score cp 47 hashfull 50 nps 134 tbhits 0 pv e2e4 d5b6 h2h4 g6h5 a2a4 a7a5 f1e2 e7e6 e1g1 h5f3
info depth 6 seldepth 10 time 105 nodes 94 score cp 46 hashfull 51 nps 371 tbhits 0 pv e2e4 d5b6 h2h4 g6h5 a2a4 a7a5 f1e2 e7e6 e1g1 f8e7
info depth 5 seldepth 10 time 169 nodes 137 score cp 47 hashfull 51 nps 485 tbhits 0 pv e2e4 d5b6 h2h4 g6h5 a2a4 a7a5 f1e2 e7e6 e1g1 f8e7
info depth 5 seldepth 10 time 381 nodes 311 score cp 57 hashfull 52 nps 671 tbhits 0 pv h2h4 h7h6 e2e4 d5b6 a2a4 a7a5 f3e5 d8d4 e5g6
info depth 5 seldepth 11 time 514 nodes 396 score cp 54 hashfull 52 nps 663 tbhits 0 pv h2h4 g6h5 d1b3 d5b6 f3e5 e7e6 g2g4
info depth 6 seldepth 11 time 1061 nodes 975 score cp 51 hashfull 55 nps 867 tbhits 0 pv h2h4 d5f6 d1b3 b8d7 b3b7 c7c5 d4c5 d7c5
info depth 6 seldepth 12 time 1427 nodes 1276 score cp 47 hashfull 56 nps 855 tbhits 0 pv h2h4 d5f6 d1b3 b8d7 b3b7 e7e6 h4h5 g6h5 f3e5
info depth 6 seldepth 13 time 1696 nodes 1682 score cp 46 hashfull 58 nps 959 tbhits 0 pv h2h4 d5f6 d1b3 b8d7 b3b7 e7e6 h4h5 g6h5 f3e5
info depth 6 seldepth 14 time 2078 nodes 2143 score cp 45 hashfull 60 nps 1004 tbhits 0 pv h2h4 d5f6 d1b3 b8d7 b3b7 e7e6 h4h5 g6h5 f3e5
info depth 6 seldepth 14 time 2887 nodes 3001 score cp 42 hashfull 63 nps 1020 tbhits 0 pv h2h4 d5f6 d1b3 b8d7 b3b7 e7e6 h4h5 g6h5 f3e5
bestmove h2h4 ponder d5f6
position startpos moves d2d4 d7d5 c2c4 g8f6 c4d5 f6d5 g1f3 c8f5 b1d2 f5g6 h2h4 d5f6
go nodes 3000
info depth 6 seldepth 9 time 102 nodes 686 score cp 36 hashfull 64 nps 166 tbhits 0 pv d1b3 e7e6 b3b7 b8d7 h4h5 g6h5 f3e5 a8b8 b7c8
info depth 6 seldepth 11 time 262 nodes 791 score cp 36 hashfull 65 nps 465 tbhits 0 pv d1b3 e7e6 b3b7 b8d7 h4h5 g6h5 f3e5 d7e5 d4e5
info depth 7 seldepth 11 time 287 nodes 914 score cp 36 hashfull 65 nps 853 tbhits 0 pv d1b3 e7e6 b3b7 b8d7 h4h5 g6h5 f3e5 d7e5 d4e5
info depth 6 seldepth 11 time 385 nodes 982 score cp 36 hashfull 66 nps 812 tbhits 0 pv d1b3 e7e6 b3b7 b8d7 h4h5 g6h5 f3e5 d7e5 d4e5
info depth 6 seldepth 12 time 478 nodes 1116 score cp 38 hashfull 66 nps 935 tbhits 0 pv d1b3 e7e6 b3b7 b8d7 h4h5 g6h5 f3e5 d7e5 d4e5
info depth 6 seldepth 13 time 996 nodes 1670 score cp 38 hashfull 69 nps 1005 tbhits 0 pv d1b3 b8d7 b3b7 e7e6 h4h5 g6h5 f3e5 a8b8 b7a7 d7e5
info depth 6 seldepth 13 time 1542 nodes 2303 score cp 38 hashfull 71 nps 1059 tbhits 0 pv d1b3 e7e6 b3b7 b8d7 h4h5 g6h5 f3e5 f8d6 e5c6 d8c8
bestmove d1b3 ponder e7e6
position startpos moves d2d4 d7d5 c2c4 g8f6 c4d5 f6d5 g1f3 c8f5 b1d2 f5g6 h2h4 d5f6 d1b3 b7b6
go nodes 3000
info depth 4 seldepth 6 time 101 nodes 36 score cp 71 hashfull 72 nps 118 tbhits 0 pv g2g4 e7e6 h4h5 g6e4 d2e4 f6e4
info depth 5 seldepth 7 time 101 nodes 46 score cp 68 hashfull 72 nps 217 tbhits 0 pv g2g4 e7e6 h4h5 g6e4 d2e4 f6e4
info depth 4 seldepth 7 time 166 nodes 58 score cp 72 hashfull 72 nps 204 tbhits 0 pv g2g4 f6g4 h4h5 g6f5 e2e4 f5d7 f3g5
info depth 4 seldepth 8 time 166 nodes 71 score cp 71 hashfull 72 nps 283 tbhits 0 pv g2g4 f6g4 h4h5 g6f5 e2e4 f5d7 f3g5
info depth 5 seldepth 9 time 233 nodes 110 score cp 67 hashfull 73 nps 369 tbhits 0 pv g2g4 f6g4 h4h5 g6f5 e2e4 f5d7 f3g5 e7e6
info depth 5 seldepth 10 time 297 nodes 142 score cp 69 hashfull 73 nps 397 tbhits 0 pv g2g4 f6g4 h4h5 g6f5 e2e4 f5d7 f3g5 e7e6 h5h6
info depth 5 seldepth 11 time 474 nodes 293 score cp 66 hashfull 73 nps 567 tbhits 0 pv g2g4 f6g4 h4h5 g6f5 e2e4 f5d7 f3g5 e7e6 d2f3 h7h6 g5h3
info depth 5 seldepth 12 time 1081 nodes 926 score cp 61 hashfull 76 nps 834 tbhits 0 pv g2g4 f6g4 h4h5 g6f5 e2e4 f5d7 f3g5 e7e6 d2f3 h7h6 g5h3
info depth 6 seldepth 13 time 1149 nodes 984 score cp 61 hashfull 77 nps 835 tbhits 0 pv g2g4 f6g4 h4h5 g6f5 e2e4 f5d7 f3g5 e7e6 d2f3 h7h6 g5h3
info depth 6 seldepth 13 time 1230 nodes 1061 score cp 84 hashfull 77 nps 843 tbhits 0 pv h4h5 g6h5 h1h5 f6h5 f3e5 e7e6 b3f3 h5f6 f3a8 f8b4 a2a3 b4d2 c1d2
info depth 6 seldepth 14 time 2317 nodes 1980 score cp 82 hashfull 82 nps 844 tbhits 0 pv h4h5 g6f5 e2e4 f6e4 f3e5 e4d6 g2g4 f5e6 d4d5 e6c8 d2f3 f7f6
info depth 6 seldepth 14 time 2470 nodes 2197 score cp 80 hashfull 83 nps 879 tbhits 0 pv h4h5 g6f5 e2e4 f6e4 f3e5 e4d6 g2g4 f5e6 d4d5 e6c8 d2f3 f7f6
bestmove h4h5 ponder g6f5
position startpos moves d2d4 d7d5 c2c4 g8f6 c4d5 f6d5 g1f3 c8f5 b1d2 f5g6 h2h4 d5f6 d1b3 b7b6 h4h5 g6h5
go nodes 3000
info depth 7 seldepth 9 time 105 nodes 218 score cp 150 hashfull 83 nps 19 tbhits 0 pv h1h5 f6h5 f3e5 e7e6 b3f3 h5f6 f3a8 f8b4 a2a3 b4d2 c1d2
info depth 8 seldepth 12 time 108 nodes 241 score cp 155 hashfull 84 nps 231 tbhits 0 pv h1h5 f6h5 f3e5 e7e6 b3f3 h5f6 f3a8 f8b4 a2a3 b4d2 c1d2
info depth 8 seldepth 13 time 173 nodes 285 score cp 166 hashfull 84 nps 398 tbhits 0 pv h1h5 f6h5 f3e5 e7e6 b3f3 h5f6 f3a8 f8b4 e2e3 e8g8 a2a3 b4d6
info depth 7 seldepth 13 time 173 nodes 316 score cp 169 hashfull 84 nps 578 tbhits 0 pv h1h5 f6h5 f3e5 e7e6 b3f3 h5f6 f3a8 f8b4 e2e3 e8g8 a2a3 b4d6
info depth 7 seldepth 14 time 372 nodes 443 score cp 173 hashfull 85 nps 610 tbhits 0 pv h1h5 f6h5 f3e5 e7e6 b3f3 h5f6 f3a8 f8b4 e2e3 e8g8 a2a3 b4d6
info depth 7 seldepth 15 time 466 nodes 519 score cp 177 hashfull 85 nps 650 tbhits 0 pv h1h5 f6h5 f3e5 e7e6 b3f3 h5f6 f3a8 f8b4 e2e3 e8g8 a2a3 b4d6
info depth 8 seldepth 15 time 613 nodes 640 score cp 188 hashfull 86 nps 691 tbhits 0 pv h1h5 f6h5 f3e5 e7e6 b3f3 h5f6 f3a8 f8b4 a2a3 b4d2 c1d2 e8g8
info depth 7 seldepth 16 time 806 nodes 921 score cp 198 hashfull 87 nps 874 tbhits 0 pv h1h5 f6h5 f3e5 e7e6 b3f3 h5f6 f3a8 f8b4 a2a3 b4d2 c1d2 e8g8 e2e3
info depth 7 seldepth 17 time 1245 nodes 1276 score cp 213 hashfull 88 nps 851 tbhits 0 pv h1h5 f6h5 f3e5 e7e6 b3f3 h5f6 f3a8 f8b4 a2a3 b4d2 c1d2 e8g8 e2e3 f6g4
info depth 8 seldepth 17 time 1272 nodes 1430 score cp 220 hashfull 89 nps 954 tbhits 0 pv h1h5 f6h5 f3e5 e7e6 b3f3 f7f6 f3h5 g7g6 e5g6 h7g6 h5h8 d8d4 e2e3 d4e5 d2f3 e5a5
info depth 8 seldepth 17 time 1641 nodes 1774 score cp 222 hashfull 90 nps 949 tbhits 0 pv h1h5 f6h5 f3e5 e7e6 b3f3 h5f6 f3a8 f8b4 a2a3 b4d2 c1d2 e8g8 e2e3 c7c5
bestmove h1h5 ponder f6h5
position startpos moves d2d4 d7d5 c2c4 g8f6 c4d5 f6d5 g1f3 c8f5 b1d2 f5g6 h2h4 d5f6 d1b3 b7b6 h4h5 g6h5 h1h5 f6h5
go nodes 3000
info depth 5 seldepth 13 time 103 nodes 1021 score cp 253 hashfull 91 nps 407 tbhits 0 pv f3e5 e7e6 b3f3 h5f6 f3a8 f8b4 a2a3 b4d2 c1d2 e8g8 e2e3 c7c5
info depth 7 seldepth 13 time 138 nodes 1067 score cp 248 hashfull 92 nps 637 tbhits 0 pv f3e5 e7e6 b3f3 h5f6 f3a8 f8b4 a2a3 b4d2 c1d2 e8g8 e2e3 c7c5
info depth 8 seldepth 13 time 260 nodes 1188 score cp 237 hashfull 92 nps 803 tbhits 0 pv f3e5 e7e6 b3f3 h5f6 f3a8 f8b4 a2a3 b4d2 c1d2 e8g8 e2e3 c7c5
info depth 8 seldepth 14 time 311 nodes 1191 score cp 238 hashfull 92 nps 681 tbhits 0 pv f3e5 e7e6 b3f3 h5f6 f3a8 f8b4 a2a3 b4d2 c1d2 e8g8 e2e3 c7c5
info depth 8 seldepth 15 time 376 nodes 1250 score cp 231 hashfull 93 nps 720 tbhits 0 pv f3e5 e7e6 b3f3 h5f6 f3a8 f8b4 a2a3 b4d2 c1d2 e8g8 e2e3 c7c5
info depth 8 seldepth 16 time 630 nodes 1459 score cp 219 hashfull 94 nps 761 tbhits 0 pv f3e5 e7e6 b3f3 d8d4 f3f7 e8d8 f7h5 f8b4 e2e3 d4d5 f1c4 d5g2 e5f7 d8e7 f7h8
info depth 9 seldepth 16 time 630 nodes 1501 score cp 217 hashfull 94 nps 828 tbhits 0 pv f3e5 e7e6 b3f3 d8d4 f3f7 e8d8 f7h5 f8b4 e2e3 d4d5 f1c4 d5g2 e5f7 d8e7 f7h8
info depth 9 seldepth 17 time 760 nodes 1574 score cp 215 hashfull 94 nps 782 tbhits 0 pv f3e5 e7e6 b3f3 d8d4 f3f7 e8d8 f7h5 d8c8 e2e3 d4d5 f1c4 d5g2 c4e6 c8b7 e5f7 f8b4 f7h8
info depth 9 seldepth 17 time 931 nodes 1759 score cp 213 hashfull 95 nps 837 tbhits 0 pv f3e5 e7e6 b3f3 d8d4 f3f7 e8d8 f7h5 d8c8 e2e3 d4d5 f1c4 d5g2 c4e6 c8b7 e5f7 f8b4 f7h8
bestmove f3e5 ponder e7e6
position startpos moves d2d4 d7d5 c2c4 g8f6 c4d5 f6d5 g1f3 c8f5 b1d2 f5g6 h2h4 d5f6 d1b3 b7b6 h4h5 g6h5 h1h5 f6h5 f3e5 e7e6
go nodes 3000
info depth 7 seldepth 12 time 105 nodes 1507 score cp 216 hashfull 96 nps 247 tbhits 0 pv b3f3 d8d4 f3f7 e8d8 f7h5 d8c8 e2e3 d4d5 f1c4 d5g2 c4e6 c8b7 e5f7 f8b4 f7h8
info depth 8 seldepth 13 time 291 nodes 1727 score cp 206 hashfull 97 nps 845 tbhits 0 pv b3f3 d8d4 f3f7 e8d8 f7h5 d8c8 e2e3 d4d5 f1c4 d5g2 c4e6 c8b7 e5f7 f8b4 f7h8
info depth 8 seldepth 13 time 308 nodes 1733 score cp 207 hashfull 97 nps 818 tbhits 0 pv b3f3 d8d4 f3f7 e8d8 f7h5 d8c8 e2e3 d4d5 f1c4 d5g2 c4e6 c8b7 e5f7 f8b4 f7h8
bestmove b3f3 ponder d8d4
+1
Always glad to help one of the early pioneers.

Re: Lc0 - macOS binary requested

Posted: Thu Dec 05, 2019 7:48 pm
by littlecapa
MikeB wrote: Sat May 25, 2019 7:05 am
Steppenwolf wrote: Fri May 24, 2019 4:54 pm I did all the things you mentioned.
Everything worked properly to my feeling.

Please go ahead...

awesome - next we will grab the source , but before grabbing the source, let's create a folder for it - sincere we will be pulling the source from GitHub and at some point you may want to pull other chess engine sources from Github , will make a Github directory off your home and sub directory lc0 in the Github folder..

typically, when you start terminal window under macOS, it puts you right in the home directory

Mac-Pro:~ your_ user_name_here$

so lets begin by typing at the command prompt in terminal all one one line You can copy and paste:

Code: Select all

mkdir Github && cd Github && mkdir Lc0 && cd Lc0
type "pwd" and you will see you created the path of "Mac-Pro:~ your_ user_name_here/Github/Lc0$
perfect

now lets grab the latest stable release source 21.1 - copy and paste this into terminal:

Code: Select all

git clone  --recurse-submodules https://github.com/LeelaChessZero/lc0.git v0.21.1
it will create another subfolder "v0.21.1" under Lc0. so let go to that folder in terminal by typing "cd v0.21.1" ( or copy and paste)

now if you type "pwd" ( Path of Working Directory) you will see this:

Code: Select all

/Users/your_user_name_here/Github/LcO/v0.21.1
copy that line and click on a finder window toactivtate its menu , select on the menu >Go>Go To Folder and paste the line you just copied.

You will ow see a finder window with all of the sources of LcO which you had just downloaded. You will see a file called "build.sh".

Just to be safe , we need to take sure we build this with clang and not gcc.

copy and paste in the terminal (under the v0.21.1 folder):

Code: Select all

CC=clang CXX=clang++ ./build.sh
As I was typing this, I was replicating the steps on my computer to make sure it will work. I obtained working copy of lc) - you may not, your MacBok Pro might be missing some dependancies - give it a shot and if it errors out , post the entire log message

If you see this

Code: Select all

21 warnings generated.
[52/52] Linking target lc0.
~/Github/Lc0/v0.21.1
you're personal copy of LcO is under ~/Github/Lc0/v0.21.1/build/release folder "lc0" I would start with this NN first :

https://github.com/dkappe/leela-chess-w ... 58-32x4-se - it's not too big, place it in the same folder you are going to run lc0 - for starters , I would run lc0 in the release folder first ( to work out any bugs - before moving it to another folder) via terminal simply type "lc0" ( BUT make sure you copied the weights over to this foilder. No need to unzip the weights file or anything, changing the name to just "weights" and lc0 will find it instantly. Once you start lc0, type uci to see what type of backend you can run whether it's OpenCL , blas or both. If you don't see OpenCL at first , there are other steps we can try to get that for you. OpenCL will be about 100 times faster than blas.
Hi MikeB!

I followed all your steps but it does not work on my Mac Mini 2018 (macOS 10.15.1). The linker always throws this error:

ld: can't map file, errno=22 file '/Applications/Xcode.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX10.15.sdk/System/Library/Frameworks/OpenCL.framework' for architecture x86_64

Any idea how to fix this?

Re: Lc0 - macOS binary requested

Posted: Fri Dec 06, 2019 4:46 am
by MikeB
littlecapa wrote: Thu Dec 05, 2019 7:48 pm
MikeB wrote: Sat May 25, 2019 7:05 am
Steppenwolf wrote: Fri May 24, 2019 4:54 pm I did all the things you mentioned.
Everything worked properly to my feeling.

Please go ahead...

awesome - next we will grab the source , but before grabbing the source, let's create a folder for it - sincere we will be pulling the source from GitHub and at some point you may want to pull other chess engine sources from Github , will make a Github directory off your home and sub directory lc0 in the Github folder..

typically, when you start terminal window under macOS, it puts you right in the home directory

Mac-Pro:~ your_ user_name_here$

so lets begin by typing at the command prompt in terminal all one one line You can copy and paste:

Code: Select all

mkdir Github && cd Github && mkdir Lc0 && cd Lc0
type "pwd" and you will see you created the path of "Mac-Pro:~ your_ user_name_here/Github/Lc0$
perfect

now lets grab the latest stable release source 21.1 - copy and paste this into terminal:

Code: Select all

git clone  --recurse-submodules https://github.com/LeelaChessZero/lc0.git v0.21.1
it will create another subfolder "v0.21.1" under Lc0. so let go to that folder in terminal by typing "cd v0.21.1" ( or copy and paste)

now if you type "pwd" ( Path of Working Directory) you will see this:

Code: Select all

/Users/your_user_name_here/Github/LcO/v0.21.1
copy that line and click on a finder window toactivtate its menu , select on the menu >Go>Go To Folder and paste the line you just copied.

You will ow see a finder window with all of the sources of LcO which you had just downloaded. You will see a file called "build.sh".

Just to be safe , we need to take sure we build this with clang and not gcc.

copy and paste in the terminal (under the v0.21.1 folder):

Code: Select all

CC=clang CXX=clang++ ./build.sh
As I was typing this, I was replicating the steps on my computer to make sure it will work. I obtained working copy of lc) - you may not, your MacBok Pro might be missing some dependancies - give it a shot and if it errors out , post the entire log message

If you see this

Code: Select all

21 warnings generated.
[52/52] Linking target lc0.
~/Github/Lc0/v0.21.1
you're personal copy of LcO is under ~/Github/Lc0/v0.21.1/build/release folder "lc0" I would start with this NN first :

https://github.com/dkappe/leela-chess-w ... 58-32x4-se - it's not too big, place it in the same folder you are going to run lc0 - for starters , I would run lc0 in the release folder first ( to work out any bugs - before moving it to another folder) via terminal simply type "lc0" ( BUT make sure you copied the weights over to this foilder. No need to unzip the weights file or anything, changing the name to just "weights" and lc0 will find it instantly. Once you start lc0, type uci to see what type of backend you can run whether it's OpenCL , blas or both. If you don't see OpenCL at first , there are other steps we can try to get that for you. OpenCL will be about 100 times faster than blas.
Hi MikeB!

I followed all your steps but it does not work on my Mac Mini 2018 (macOS 10.15.1). The linker always throws this error:

ld: can't map file, errno=22 file '/Applications/Xcode.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX10.15.sdk/System/Library/Frameworks/OpenCL.framework' for architecture x86_64

Any idea how to fix this?
Unchartered waters for me since I see you are on Catalina. Did you pull the most recent committed source or the last stable release? Generally for Lc0, they are different.

As a fyi, Catalina has caused some problems from some users.
Generally speaking, unless you are quite adept at solving Apple issues on your own, if you are compiling your own programs , best to wait ( I would say 6 months) for bugs to be fixed on major releases.

There are few things we can try. First I need to know the source you are trying to compile( date) and the compiler - gcc or clang and which version.

The error description "erron==22" is not very precise, it basically say we can not open a file that is needed, suggesting that a file is missing or is in a different location than expected, or it may be corrupt. In this case, we know it's looking for '/Applications/Xcode.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX10.15.sdk/System/Library/Frameworks/OpenCL.framework' so , in finder , if you go to folder ( Finder>>Go>>Go to Folder...)
'/Applications/Xcode.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX10.15.sdk/System/Library/Frameworks/" is that a valid path on your Mac Mini? If it is, does it include the file OpenCL.framework file?

Things we can try , if you complied a source correctly before without issues., try that "older" source on your Mac Mini. Compare the Mac Mini compiler and version to the compiler version you successfully compiled. We want to rule out differences between the successful compilation and the unsuccessful compilation as much as possible - so try to make everything on your Mac Mini the same as your previous Mac - date of source, complier version etc.

Re: Lc0 - macOS binary requested

Posted: Fri Dec 06, 2019 7:13 am
by MikeB
MikeB wrote: Fri Dec 06, 2019 4:46 am
littlecapa wrote: Thu Dec 05, 2019 7:48 pm
MikeB wrote: Sat May 25, 2019 7:05 am
Steppenwolf wrote: Fri May 24, 2019 4:54 pm I did all the things you mentioned.
Everything worked properly to my feeling.

Please go ahead...

awesome - next we will grab the source , but before grabbing the source, let's create a folder for it - sincere we will be pulling the source from GitHub and at some point you may want to pull other chess engine sources from Github , will make a Github directory off your home and sub directory lc0 in the Github folder..

typically, when you start terminal window under macOS, it puts you right in the home directory

Mac-Pro:~ your_ user_name_here$

so lets begin by typing at the command prompt in terminal all one one line You can copy and paste:

Code: Select all

mkdir Github && cd Github && mkdir Lc0 && cd Lc0
type "pwd" and you will see you created the path of "Mac-Pro:~ your_ user_name_here/Github/Lc0$
perfect

now lets grab the latest stable release source 21.1 - copy and paste this into terminal:

Code: Select all

git clone  --recurse-submodules https://github.com/LeelaChessZero/lc0.git v0.21.1
it will create another subfolder "v0.21.1" under Lc0. so let go to that folder in terminal by typing "cd v0.21.1" ( or copy and paste)

now if you type "pwd" ( Path of Working Directory) you will see this:

Code: Select all

/Users/your_user_name_here/Github/LcO/v0.21.1
copy that line and click on a finder window toactivtate its menu , select on the menu >Go>Go To Folder and paste the line you just copied.

You will ow see a finder window with all of the sources of LcO which you had just downloaded. You will see a file called "build.sh".

Just to be safe , we need to take sure we build this with clang and not gcc.

copy and paste in the terminal (under the v0.21.1 folder):

Code: Select all

CC=clang CXX=clang++ ./build.sh
As I was typing this, I was replicating the steps on my computer to make sure it will work. I obtained working copy of lc) - you may not, your MacBok Pro might be missing some dependancies - give it a shot and if it errors out , post the entire log message

If you see this

Code: Select all

21 warnings generated.
[52/52] Linking target lc0.
~/Github/Lc0/v0.21.1
you're personal copy of LcO is under ~/Github/Lc0/v0.21.1/build/release folder "lc0" I would start with this NN first :

https://github.com/dkappe/leela-chess-w ... 58-32x4-se - it's not too big, place it in the same folder you are going to run lc0 - for starters , I would run lc0 in the release folder first ( to work out any bugs - before moving it to another folder) via terminal simply type "lc0" ( BUT make sure you copied the weights over to this foilder. No need to unzip the weights file or anything, changing the name to just "weights" and lc0 will find it instantly. Once you start lc0, type uci to see what type of backend you can run whether it's OpenCL , blas or both. If you don't see OpenCL at first , there are other steps we can try to get that for you. OpenCL will be about 100 times faster than blas.
Hi MikeB!

I followed all your steps but it does not work on my Mac Mini 2018 (macOS 10.15.1). The linker always throws this error:

ld: can't map file, errno=22 file '/Applications/Xcode.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX10.15.sdk/System/Library/Frameworks/OpenCL.framework' for architecture x86_64

Any idea how to fix this?
Unchartered waters for me since I see you are on Catalina. Did you pull the most recent committed source or the last stable release? Generally for Lc0, they are different.

As a fyi, Catalina has caused some problems from some users.
Generally speaking, unless you are quite adept at solving Apple issues on your own, if you are compiling your own programs , best to wait ( I would say 6 months) for bugs to be fixed on major releases.

There are few things we can try. First I need to know the source you are trying to compile( date) and the compiler - gcc or clang and which version.

The error description "erron==22" is not very precise, it basically say we can not open a file that is needed, suggesting that a file is missing or is in a different location than expected, or it may be corrupt. In this case, we know it's looking for '/Applications/Xcode.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX10.15.sdk/System/Library/Frameworks/OpenCL.framework' so , in finder , if you go to folder ( Finder>>Go>>Go to Folder...)
'/Applications/Xcode.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX10.15.sdk/System/Library/Frameworks/" is that a valid path on your Mac Mini? If it is, does it include the file OpenCL.framework file?

Things we can try , if you complied a source correctly before without issues., try that "older" source on your Mac Mini. Compare the Mac Mini compiler and version to the compiler version you successfully compiled. We want to rule out differences between the successful compilation and the unsuccessful compilation as much as possible - so try to make everything on your Mac Mini the same as your previous Mac - date of source, complier version etc.
Through the process of elimination, as I was describing above ,I was able to locate the source of the error message - first thing I tried was to build lc0 using the source which built fine before. I got the same error you got above. I then restored the prior OpenCL framework, and to my surprise that also failed. The fact that BOTH gcc and clang were creating the same error , ruled out the possibility it was the compilers at fault, the error was not to due to any changes in the source - so it had to be one of other tools that are using being used to build Lc0. I took a wild guess it was meson ;>) , reactivated the old version ( I use macports, if you use home-brew, you will have to check with their instructions to reactive prior ports)

Code: Select all

  sudo port activate meson @0.51.2_0
--->  Computing dependencies for meson
--->  Activating meson @0.51.2_0
--->  Cleaning meson
 
bingo it worked on the old source. meson 5.2 is the culprit here.

update to the latest source - and you will get this error

../../src/neural/encoder.cc:67:37: error: no member named 'INPUT_112_WITH_CASTLING_PLANE' in 'pblczero::NetworkFormat'
case pblczero::NetworkFormat::INPUT_112_WITH_CASTLING_PLANE: {

an error like this should never happen, it tells me that this code is not being used, how it got comitted, I don't know.

Comment out lines 67 through 82 ( I used /* and */ below in encoder.cc , making the code inactive , my guess is this is related to FRC castling code without taking a real hard look at it.
code above references , "INPUT_CLASSICAL_112_PLANE" which is probably classical/standard chess.

Code: Select all

/*
      case pblczero::NetworkFormat::INPUT_112_WITH_CASTLING_PLANE: {
        // - Plane 104 for positions of rooks (both white and black) which have
        // a-side (queenside) castling right.
        // - Plane 105 for positions of rooks (both white and black) which have
        // h-side (kingside) castling right.
        const auto& cast = board.castlings();
        result[kAuxPlaneBase + 0].mask =
            ((cast.we_can_000() ? ChessBoard::A1 : 0) |
             (cast.they_can_000() ? ChessBoard::A8 : 0))
            << cast.queenside_rook();
        result[kAuxPlaneBase + 1].mask =
            ((cast.we_can_00() ? ChessBoard::A1 : 0) |
             (cast.they_can_00() ? ChessBoard::A8 : 0))
            << cast.kingside_rook();
        break;
      }
*/
bingo again, it compiles

Code: Select all

[100/100] Linking target encoder_test.
~/Github/LeelaChessZero
good luck!