Help with Scorpio 3.0.8

[Dann Corbit]

Look at how frugal it is with resources:


Or maybe it is not running properly?

[Terje]

Why does scorpio not report depths in the same way leela or allie does?

I guess because depths are relatively meaningless with MCTS programs. It's already iffy with alpha-beta (Does a single reply extension count? (for instance))

With alpha-beta, you at least have a depth measure that is passed into the search algorithm. The MCTS searches kind of have an approximation of that, but it is not the same as an alpha-beta depth.

A 20 ply MCTS search is not commensurate with a 20 ply alpha-beta search.

So I guess he was being funny and making a joke about depth in the PGN standard for EPD and FEN.

When he answers, I'll see if my guess is right.

The depths aren't particularly useful, but they are well defined in AB. And sure, MCTS depths aren't comparable to AB in the slightest, but trying to keep them somewhat similar between MCTS engines would be nice..

[Dann Corbit]

Why does scorpio not report depths in the same way leela or allie does?

I guess because depths are relatively meaningless with MCTS programs. It's already iffy with alpha-beta (Does a single reply extension count? (for instance))

With alpha-beta, you at least have a depth measure that is passed into the search algorithm. The MCTS searches kind of have an approximation of that, but it is not the same as an alpha-beta depth.

A 20 ply MCTS search is not commensurate with a 20 ply alpha-beta search.

So I guess he was being funny and making a joke about depth in the PGN standard for EPD and FEN.

When he answers, I'll see if my guess is right.

The depths aren't particularly useful, but they are well defined in AB. And sure, MCTS depths aren't comparable to AB in the slightest, but trying to keep them somewhat similar between MCTS engines would be nice..

He is doing something different (e.g. part of the time he does AB search). But his odd depths were causing me grief so I asked him to change it and he did. So starting with the next contest (I imagine) the depths will be more like what we are used to.

[Dann Corbit]

I see that Scorpio is now using 100% of my CPU.
Still nothing is happening with my GPU.
That seems backwards.

[Daniel Shawul]

At first I thought that it was a mistake it was launching only 10 threads, but it isn't.
The GTX 1050 has only 5 SMs, so it allocates 2x the number of threads.

I am curious though how much pps it will get with mt set to 20 or 40.
Could you set mt to these values in scoprio.ini and run scorpio.bat go quit again, and post the results here?
My estimate of using 2x number of SMs of GPU could be wrong on low end GPUs.

Thanks.
Daniel

This is mt 20:

Code: Select all

C:\Engines\ScorpioNN\Scorpio\bin\Windows>scorpio.bat go quit
feature done=0
Number of cores 8 of 8
treeht 83886080 X 320 = 25600.0 MB
processors [20]
ht 67108864 X 16 = 1024.0 MB
eht 262144 X 8 X 20 = 80.0 MB
pht 32768 X 24 X 20 = 15.0 MB
EgbbProbe 4.3 by Daniel Shawul
egbb_cache 4084 X 8216 = 32.0 MB
180 egbbs loaded !
nn_cache 131072 X 1568 = 196.0 MB
Loading neural network : C:\Engines\ScorpioNN\Scorpio\nets-scorpio/ens-net-20x256.uff
With "INT8" precision
Loading graph on /gpu:0
0. main_input 2048 = 32 8 810
1. policy/Reshape 1024 = 1024 1 1
2. value/BiasAdd 3 = 3 1 1
Neural network loaded !
nn_cache 131072 X 1568 = 196.0 MB
Loading neural network : C:\Engines\ScorpioNN\Scorpio\nets-maddex/ME.uff
With "INT8" precision
Loading graph on /gpu:0
0. main_input 7168 = 112 8 80
1. value_head 3 = 3 1 1
2. policy_head 1858 = 1858 1 1
Neural network loaded !
loading_time = 1400s
# rnbqkbnr/pppppppp/8/8/8/8/PPPPPPPP/RNBQKBNR w KQkq - 0 1
# [st = 12502ms, mt = 29250ms , hply = 0 , moves_left 10]
63 32 87 2095 63 2394 0 e2-e4 e7-e5 Ng1-f3 Nb8-c6 Bf1-b5 Ng8-f6 Ke1-g1 Nf6xe4 Rf1-e1 Ne4-d6 Nf3xe5 Bf8-e7 Bb5-f1 Nc6xe5 Re1xe5 Ke8-g8 d2-d4 Be7-f6
64 32 175 4342 64 2481 0 e2-e4 e7-e5 Ng1-f3 Nb8-c6 Bf1-b5 Ng8-f6 Ke1-g1 Nf6xe4 Rf1-e1 Ne4-d6 Nf3xe5 Bf8-e7 Bb5-f1 Nc6xe5 Re1xe5 Ke8-g8 d2-d4 Be7-f6 Re5-e1 Rf8-e8 c2-c3 Re8xe1 Qd1xe1 Nd6-e8 Bc1-f4 d7-d5 Nb1-d2 c7-c6 Bf1-d3 g7-g6 Nd2-f3
65 31 264 6663 65 2516 0 e2-e4 e7-e5 Ng1-f3 Nb8-c6 Bf1-b5 Ng8-f6 Ke1-g1 Nf6xe4 Rf1-e1 Ne4-d6 Nf3xe5 Bf8-e7 Bb5-f1 Nc6xe5 Re1xe5 Ke8-g8 d2-d4 Be7-f6 Re5-e1 Rf8-e8 c2-c3 Re8xe1 Qd1xe1 Nd6-e8 Bc1-f4 d7-d5 Nb1-d2 c7-c6 Bf1-d3 g7-g6 Nd2-f3
66 31 352 9056 66 2570 0 e2-e4 e7-e5 Ng1-f3 Nb8-c6 Bf1-b5 a7-a6 Bb5-a4 Ng8-f6 Ke1-g1 Bf8-e7 Rf1-e1 b7-b5 Ba4-b3 d7-d6 c2-c3 Ke8-g8 h2-h3 Bc8-b7 d2-d4 Rf8-e8 Nb1-d2 Be7-f8 a2-a4 h7-h6 Bb3-c2 Ra8-b8
67 30 439 11525 67 2620 0 e2-e4 e7-e5 Ng1-f3 Nb8-c6 Bf1-b5 a7-a6 Bb5-a4 Ng8-f6 Ke1-g1 Bf8-e7 Rf1-e1 b7-b5 Ba4-b3 d7-d6 c2-c3 Ke8-g8 h2-h3 Bc8-b7 d2-d4 Rf8-e8 Nb1-d2 Be7-f8 a2-a4 h7-h6 Bb3-c2 Ra8-b8
68 30 528 14276 68 2699 0 e2-e4 e7-e5 Ng1-f3 Nb8-c6 Bf1-b5 a7-a6 Bb5-a4 Ng8-f6 Ke1-g1 Bf8-e7 Rf1-e1 b7-b5 Ba4-b3 d7-d6 c2-c3 Ke8-g8 h2-h3 Bc8-b7 d2-d4 Rf8-e8 Nb1-d2 Be7-f8 a2-a4 h7-h6 Bb3-c2 Ra8-b8
69 30 616 16729 69 2714 0 e2-e4 e7-e5 Ng1-f3 Nb8-c6 Bf1-b5 a7-a6 Bb5-a4 Ng8-f6 Ke1-g1 Bf8-e7 Rf1-e1 b7-b5 Ba4-b3 d7-d6 c2-c3 Ke8-g8 h2-h3 Bc8-b7 d2-d4 Rf8-e8 Nb1-d2 Be7-f8 a2-a4 h7-h6 Bb3-c2 Ra8-b8 Bc2-b1
70 30 703 19160 70 2722 0 e2-e4 e7-e5 Ng1-f3 Nb8-c6 Bf1-b5 a7-a6 Bb5-a4 Ng8-f6 Ke1-g1 Nf6xe4 d2-d4 b7-b5 Ba4-b3 d7-d5 d4xe5 Bc8-e6 c2-c3 Bf8-e7 Nb1-d2 Ne4-c5 Bb3-c2 d5-d4 Nd2-b3 d4xc3 Nb3xc5 Be7xc5 Bc2-e4 Qd8-d7 b2xc3 Ra8-d8
71 31 791 21572 71 2726 0 e2-e4 e7-e5 Ng1-f3 Nb8-c6 Bf1-b5 a7-a6 Bb5-a4 Ng8-f6 Ke1-g1 Nf6xe4 d2-d4 b7-b5 Ba4-b3 d7-d5 d4xe5 Bc8-e6 c2-c3 Bf8-e7 Nb1-d2 Ne4-c5 Bb3-c2 d5-d4 Nd2-b3 d4xc3 Nb3xc5 Be7xc5 Bc2-e4 Qd8-d7 b2xc3 Ra8-d8 Qd1xd7 Be6xd7
72 31 878 24040 72 2735 0 e2-e4 e7-e5 Ng1-f3 Nb8-c6 Bf1-b5 a7-a6 Bb5-a4 Ng8-f6 Ke1-g1 Nf6xe4 d2-d4 b7-b5 Ba4-b3 d7-d5 d4xe5 Bc8-e6 c2-c3 Bf8-e7 Nb1-d2 Ne4-c5 Bb3-c2 d5-d4 Nd2-b3 d4xc3 Nb3xc5 Be7xc5 Bc2-e4 Qd8-d7 b2xc3 Ra8-d8 Qd1xd7 Be6xd7
73 31 946 25919 73 2738 0 e2-e4 e7-e5 Ng1-f3 Nb8-c6 Bf1-b5 a7-a6 Bb5-a4 Ng8-f6 Ke1-g1 Nf6xe4 d2-d4 b7-b5 Ba4-b3 d7-d5 d4xe5 Bc8-e6 c2-c3 Bf8-e7 Nb1-d2 Ne4-c5 Bb3-c2 d5-d4 Nd2-b3 d4xc3 Nb3xc5 Be7xc5 Bc2-e4 Qd8-d7 b2xc3 Ra8-d8 Qd1xd7 Be6xd7
# nodes 267354 <0% qnodes> time 9550ms nps 27995 eps 0 nneps 2733 tbhits 0
# Tree: nodes 33428 depth 33 pps 2732 visits 25927 qcalls 0 scalls 0
move e2e4
Bye Bye

Thanks could you please use mt 20 then, as it seems to have improved a bit!

[Daniel Shawul]

Why does scorpio not report depths in the same way leela or allie does?

Dann, answered it that there is no clear iterative search depth in mcts search.
My odd choice of starting depth from 63 is because my AB search runs from depth = 1 to 63. Scorpio can do hybrid AB
and MCTS, running the AB first, say the first 30% of the time, so 63 is a safe option that avoids depth overlap.

Assuming an effective branching of 2 or something similar, I could calculate an average depth like leela i guess.

[Daniel Shawul]

I see that Scorpio is now using 100% of my CPU.
Still nothing is happening with my GPU.
That seems backwards.

Sorry Dann, I was asleep. Can you post the result of "scorpio.bat go quit" from the bin/Windows directory ?
That will tell me what settings it intalled scorpio with and how many pps it is able to achieve.

You are already contributing games as shown here http://scorpiozero.ddns.net/contrib.php !

Scorpio is rather hungry for CPU cores, so depends on how many you have whether to run two separate clients or just one.
I have a relatively new CPU (ryzen 9) with a modest rtx 2070 super, and i have not found that beats this combo on any of my vast.ai instances interms
of game generation rate -- even with an RTX 2080 ti. Most have very few few cores per GPU there.

[Terje]

Why does scorpio not report depths in the same way leela or allie does?

Dann, answered it that there is no clear iterative search depth in mcts search.
My odd choice of starting depth from 63 is because my AB search runs from depth = 1 to 63. Scorpio can do hybrid AB
and MCTS, running the AB first, say the first 30% of the time, so 63 is a safe option that avoids depth overlap.

Assuming an effective branching of 2 or something similar, I could calculate an average depth like leela i guess.

I see. In most parts of the engine originality is great, but just for the UCI/WB communication it would be nice if everyone was on the same page as much as possible

[Daniel Shawul]

Why does scorpio not report depths in the same way leela or allie does?

Dann, answered it that there is no clear iterative search depth in mcts search.
My odd choice of starting depth from 63 is because my AB search runs from depth = 1 to 63. Scorpio can do hybrid AB
and MCTS, running the AB first, say the first 30% of the time, so 63 is a safe option that avoids depth overlap.

Assuming an effective branching of 2 or something similar, I could calculate an average depth like leela i guess.

I see. In most parts of the engine originality is great, but just for the UCI/WB communication it would be nice if everyone was on the same page as much as possible

Well tell me what depth to report when the AB search first did search up to depth of 30. AFAIK lc0 depths are maximum 30.
If I want to report depths just like leela the two will overlap, and I know for instance xboard sorts your PVs based on your depth creating a mess i guess.
An example if I do AB all the time and how Scorpio avoid this. It offsets the AB depth by mcts_strategy_depth=30 when it sarts alpha-beta search.

Code: Select all

$ ./scorpio.sh alphabeta_man_c 32 go quit
$ ./scorpio.sh alphabeta_man_c 32 go quit
feature done=0
Number of cores 4 of 4
treeht 83886080 X 320 = 25600.0 MB
processors [16]
ht 67108864 X 16 = 1024.0 MB
eht 262144 X 8 X 16 = 64.0 MB
pht 32768 X 24 X 16 = 12.0 MB
EgbbProbe 4.3 by Daniel Shawul
egbb_cache 4084 X 8216 = 32.0 MB
180 egbbs loaded !      
nn_cache 131072 X 1568 = 196.0 MB
Loading neural network : ../nets-scorpio/ens-net-12x128.pb
With "HALF" precision
Loading graph on /cpu:0
0. main_input = -1 -1 32 
1. policy/Reshape
0. value/BiasAdd
Neural network loaded !	
loading_time = 0s
# rnbqkbnr/pppppppp/8/8/8/8/PPPPPPPP/RNBQKBNR w KQkq - 0 1
# [st = 12502ms, mt = 29250ms , hply = 0 , moves_left 10]
2 48 0 21 2 21000 0 e2-e4 e7-e5
3 29 0 224 3 112000 0 e2-e4 e7-e5 Nb1-c3
4 48 0 1005 4 335000 0 e2-e4 e7-e5 Nb1-c3 Nb8-c6
5 13 0 2291 5 458200 0 e2-e4 e7-e5 Nb1-c3 Nb8-c6 Ng1-f3
5 15 1 3625 5 362500 0 Ng1-f3 d7-d5 d2-d3 Ng8-f6 Nb1-c3
6 43 2 14773 6 671500 0 Ng1-f3 d7-d5 d2-d3 Ng8-f6 Nb1-c3 Nb8-c6
6 48 2 15737 6 655708 0 e2-e4 e7-e5 Nb1-c3 Nb8-c6 Ng1-f3 Ng8-f6
7 16 2 26601 7 985222 0 e2-e4 e7-e5 Nb1-c3 Nb8-c6 Ng1-f3 Ng8-f6 d2-d3
7 17 3 28173 7 939100 0 Ng1-f3 d7-d5 Nb1-c3 Ng8-f6 d2-d4 Nb8-c6 e2-e3
8 48 4 54526 8 1239227 0 Ng1-f3 d7-d5 d2-d4 Ng8-f6 Nb1-c3 Nb8-c6 e2-e3 e7-e6
9 33 6 79241 9 1257793 0 Ng1-f3 d7-d5 d2-d4 Ng8-f6 e2-e3 Nb8-c6 Bf1-e2 e7-e6 Ke1-g1
10 53 10 153548 10 1520277 0 Ng1-f3 d7-d5 d2-d4 Ng8-f6 e2-e3 e7-e6 Bf1-e2 Bf8-d6 Ke1-g1 Ke8-g8
11 17 14 250806 11 1706163 0 Ng1-f3 Ng8-f6 d2-d4 d7-d5 e2-e3 e7-e6 Bf1-e2 Bf8-d6 Ke1-g1 Ke8-g8 Nb1-c3
11 32 19 280744 11 1410773 0 e2-e4 e7-e5 Ng1-f3 Nb8-c6 d2-d4 e5xd4 Nf3xd4 Ng8-f6 Nb1-c3 Nc6xd4 Qd1xd4 d7-d6
12 57 27 602402 12 2166913 0 e2-e4 e7-e5 Nb1-c3 Bf8-e7 Ng1-f3 Nb8-c6 Bf1-b5 Ng8-f6 Ke1-g1 Ke8-g8 d2-d3 d7-d6
13 25 33 718671 13 2132554 0 e2-e4 e7-e5 Ng1-f3 Nb8-c6 Bf1-c4 Ng8-f6 Ke1-g1 Bf8-d6 Nb1-c3 Ke8-g8 d2-d3 a7-a6 Bc1-e3
14 43 60 1159072 14 1906368 0 e2-e4 e7-e6 d2-d4 d7-d5 Nb1-c3 d5xe4 Nc3xe4 Bf8-e7 Ng1-f3 Ng8-f6 Ne4xf6 Be7xf6 Bf1-d3 Bf6xd4 Ke1-g1
14 48 66 1299566 14 1969039 0 Ng1-f3 Ng8-f6 e2-e3 d7-d5 d2-d4 e7-e6 Bf1-d3 Nb8-c6 Ke1-g1 Bf8-d6 Nb1-c3 Ke8-g8 a2-a3 a7-a6
15 14 79 1779451 15 2246781 0 Ng1-f3 Ng8-f6 a2-a3 Nb8-c6 d2-d4 d7-d5 Nb1-c3 e7-e6 e2-e3 Bf8-d6 Bf1-d3 Ke8-g8 Ke1-g1 Rf8-e8 Bc1-d2 a7-a6 h2-h3
15 55 85 1825655 15 2145305 0 e2-e4 e7-e6 d2-d4 d7-d5 Nb1-c3 d5xe4 Nc3xe4 Bf8-e7 Ng1-f3 Ng8-f6 Ne4xf6 Be7xf6 Bf1-d3 Ke8-g8 Ke1-g1 Bf6xd4 Nf3xd4 Qd8xd4
16 53 107 2633540 16 2452085 0 e2-e4 e7-e6 d2-d4 d7-d5 Nb1-c3 d5xe4 Nc3xe4 Bf8-e7 Ng1-f3 Ng8-f6 Ne4xf6 Be7xf6 Bf1-d3 Nb8-c6 c2-c3 Ke8-g8 Ke1-g1
17 30 153 4554049 17 2968741 0 e2-e4 e7-e5 Ng1-f3 Ng8-f6 Nf3xe5 Nf6xe4 Nb1-c3 Ne4xc3 d2xc3 Bf8-d6 Ne5-f3 Nb8-c6 Qd1-e2 Bd6-e7 Bc1-e3 Ke8-g8 Ke1-c1 d7-d5
17 30 163 5464681 17 3352565 0 e2-e4 e7-e5 Ng1-f3 Ng8-f6 Nf3xe5 Nf6xe4 Nb1-c3 Ne4xc3 d2xc3 Bf8-d6 Ne5-f3 Nb8-c6 Qd1-e2 Bd6-e7 Bc1-e3 Ke8-g8 Ke1-c1 d7-d5
# nodes 5464681 <13% qnodes> time 1631ms nps 3350509 eps 1180407 nneps 0 tbhits 0 splits 0 badsplits 0
48 27 109 226 48 206 0 e2-e4 e7-e5 Ng1-f3 Nb8-c6 Bf1-b5 a7-a6 Bb5-a4 Ng8-f6 Ke1-g1 Bf8-e7 Rf1-e1
49 34 209 522 49 249 0 e2-e4 e7-e5 Ng1-f3 Nb8-c6 Bf1-b5 a7-a6 Bb5-a4 Ng8-f6 Ke1-g1 Bf8-e7 Rf1-e1 b7-b5 Ba4-b3 d7-d6 c2-c3 Ke8-g8
50 33 311 841 50 269 0 e2-e4 e7-e5 Ng1-f3 Nb8-c6 Bf1-b5 Ng8-f6 Ke1-g1 Nf6xe4 Rf1-e1 Ne4-d6 Nf3xe5 Bf8-e7 Bb5-f1 Nc6xe5 Re1xe5
51 32 408 1134 51 277 0 e2-e4 e7-e5 Ng1-f3 Nb8-c6 Bf1-b5 a7-a6 Bb5-a4 Ng8-f6 Ke1-g1 Bf8-e7 Rf1-e1 b7-b5 Ba4-b3 d7-d6 c2-c3 Ke8-g8 h2-h3 Bc8-b7 d2-d4 Rf8-e8 Nb1-d2 Be7-f8 a2-a4
52 31 506 1444 52 284 0 e2-e4 e7-e5 Ng1-f3 Nb8-c6 Bf1-b5 Ng8-f6 Ke1-g1 Nf6xe4 Rf1-e1 Ne4-d6 Nf3xe5 Bf8-e7 Bb5-f1 Nc6xe5 Re1xe5 Ke8-g8 d2-d4 Be7-f6 Re5-e1 Rf8-e8 c2-c3 Re8xe1
53 34 615 1768 53 287 0 e2-e4 e7-e5 Ng1-f3 Nb8-c6 Bf1-b5 Ng8-f6 Ke1-g1 Nf6xe4 Rf1-e1 Ne4-d6 Nf3xe5 Bf8-e7 Bb5-f1 Nc6xe5 Re1xe5 Ke8-g8 d2-d4 Be7-f6 Re5-e1 Rf8-e8 c2-c3 Re8xe1 Qd1xe1 Nd6-e8 Bc1-f4
54 33 728 2083 54 285 0 e2-e4 e7-e5 Ng1-f3 Nb8-c6 Bf1-b5 Ng8-f6 Ke1-g1 Nf6xe4 Rf1-e1 Ne4-d6 Nf3xe5 Bf8-e7 Bb5-f1 Nc6xe5 Re1xe5 Ke8-g8 d2-d4 Be7-f6 Re5-e1 Rf8-e8 c2-c3 Re8xe1 Qd1xe1 Nd6-e8 Bc1-f4 d7-d5 Nb1-d2 c7-c6 Bf1-d3
55 34 840 2388 55 284 0 e2-e4 e7-e5 Ng1-f3 Nb8-c6 Bf1-b5 Ng8-f6 Ke1-g1 Nf6xe4 Rf1-e1 Ne4-d6 Nf3xe5 Bf8-e7 Bb5-f1 Nc6xe5 Re1xe5 Ke8-g8 d2-d4 Be7-f6 Re5-e1 Rf8-e8 c2-c3 Re8xe1 Qd1xe1 Nd6-e8 Bc1-f4 d7-d5 Nb1-d2 c7-c6 Bf1-d3
56 34 840 2398 56 285 0 e2-e4 e7-e5 Ng1-f3 Nb8-c6 Bf1-b5 Ng8-f6 Ke1-g1 Nf6xe4 Rf1-e1 Ne4-d6 Nf3xe5 Bf8-e7 Bb5-f1 Nc6xe5 Re1xe5 Ke8-g8 d2-d4 Be7-f6 Re5-e1 Rf8-e8 c2-c3 Re8xe1 Qd1xe1 Nd6-e8 Bc1-f4 d7-d5 Nb1-d2 c7-c6 Bf1-d3

[Dann Corbit]

I see that Scorpio is now using 100% of my CPU.
Still nothing is happening with my GPU.
That seems backwards.

Sorry Dann, I was asleep. Can you post the result of "scorpio.bat go quit" from the bin/Windows directory ?
That will tell me what settings it intalled scorpio with and how many pps it is able to achieve.

You are already contributing games as shown here http://scorpiozero.ddns.net/contrib.php !

Scorpio is rather hungry for CPU cores, so depends on how many you have whether to run two separate clients or just one.
I have a relatively new CPU (ryzen 9) with a modest rtx 2070 super, and i have not found that beats this combo on any of my vast.ai instances interms
of game generation rate -- even with an RTX 2080 ti. Most have very few few cores per GPU there.

Every few runs, I get output like this:

Code: Select all

C:\source\rtSafe\safeContext.cpp (133) - Cudnn Error in nvinfer1::rt::CommonContext::configure: 7 (CUDNN_STATUS_MAPPING_ERROR)
FAILED_EXECUTION: Unknown exception
C:\source\rtSafe\safeContext.cpp (133) - Cudnn Error in nvinfer1::rt::CommonContext::configure: 7 (CUDNN_STATUS_MAPPING_ERROR)
FAILED_EXECUTION: Unknown exception
C:\source\rtSafe\safeContext.cpp (133) - Cudnn Error in nvinfer1::rt::CommonContext::configure: 7 (CUDNN_STATUS_MAPPING_ERROR)
FAILED_EXECUTION: Unknown exception
C:\source\rtSafe\safeContext.cpp (133) - Cudnn Error in nvinfer1::rt::CommonContext::configure: 7 (CUDNN_STATUS_MAPPING_ERROR)
FAILED_EXECUTION: Unknown exception
C:\source\rtSafe\safeContext.cpp (133) - Cudnn Error in nvinfer1::rt::CommonContext::configure: 7 (CUDNN_STATUS_MAPPING_ERROR)
FAILED_EXECUTION: Unknown exception
C:\source\rtExt\engine.cpp (845) - Cuda Error in nvinfer1::rt::ExecutionContext::executeInternal: 77 (an illegal memory access was encountered)
C:/source/rtSafe/cuda/genericIMMAPooling.cu (FAILED_EXECUTION: Unknown exception

The calibration run was still going when I ran this, so if I should stop it, let me know and I will re-run it:

Code: Select all

C:\Users\dannc\chess\nn-dist-master\Scorpio\bin\Windows>scorpio.bat go quit
feature done=0
Number of cores 64 of 64
treeht 83886080 X 320 = 25600.0 MB
processors [192]
ht 67108864 X 16 = 1024.0 MB
eht 262144 X 8 X 192 = 768.0 MB
pht 32768 X 24 X 192 = 144.0 MB
EgbbProbe 4.3 by Daniel Shawul
egbb_cache 4084 X 8216 = 32.0 MB
0 egbbs loaded !
nn_cache 131072 X 1568 = 196.0 MB
Loading neural network : C:\Users\dannc\chess\nn-dist-master\Scorpio\nets-scorpio/ens-net-20x256.uff
With "INT8" precision
Loading graph on /gpu:0
Loading graph on /gpu:1
0. main_input 2048 = 32 8 8
1. policy/Reshape 1024 = 1024 1 1
2. value/BiasAdd 3 = 3 1 1
Neural network loaded !
loading_time = 7s
# rnbqkbnr/pppppppp/8/8/8/8/PPPPPPPP/RNBQKBNR w KQkq - 0 1
# [st = 12502ms, mt = 29250ms , hply = 0 , moves_left 10]
32 29 87 22810 32 25979 0 e2-e4 e7-e5 Ng1-f3 Nb8-c6 Bf1-b5 a7-a6 Bb5-a4 Ng8-f6 Ke1-g1 Bf8-e7 Rf1-e1 b7-b5 Ba4-b3 d7-d6 c2-c3 Ke8-g8 h2-h3 Bc8-b7 d2-d4 Rf8-e8 Nb1-d2 Be7-f8 a2-a4 h7-h6 Bb3-c2 Ra8-b8
33 28 176 45563 33 25814 0 e2-e4 e7-e5 Ng1-f3 Nb8-c6 Bf1-b5 a7-a6 Bb5-a4 Ng8-f6 Ke1-g1 Nf6xe4 d2-d4 b7-b5 Ba4-b3 d7-d5 d4xe5 Bc8-e6 c2-c3 Bf8-e7 Nb1-d2 Ne4-c5 Bb3-c2 d5-d4 Nd2-b3 d4xc3 Nb3xc5 Be7xc5 Bc2-e4 Qd8-d7 b2xc3 Ra8-d8 Qd1xd7
34 28 263 68816 34 26126 0 e2-e4 e7-e5 Ng1-f3 Nb8-c6 Bf1-b5 a7-a6 Bb5-a4 Ng8-f6 Ke1-g1 Nf6xe4 d2-d4 b7-b5 Ba4-b3 d7-d5 d4xe5 Bc8-e6 c2-c3 Bf8-e7 Nb1-d2 Ne4-c5 Bb3-c2 d5-d4 Nd2-b3 d4xc3 Nb3xc5 Be7xc5 Bc2-e4 Qd8-d7 b2xc3 Ra8-d8 Qd1xd7 Be6xd7
35 30 351 87781 35 24987 0 e2-e4 e7-e5 Ng1-f3 Nb8-c6 Bf1-b5 a7-a6 Bb5-a4 Ng8-f6 Ke1-g1 Bf8-e7 Rf1-e1 b7-b5 Ba4-b3 d7-d6 c2-c3 Ke8-g8 h2-h3 Bc8-b7 d2-d4 Rf8-e8 Nb1-d2 Be7-f8 a2-a4 h7-h6 Bb3-c2 Ra8-b8 Bc2-b1 b5-b4 a4-a5 b4xc3
36 30 440 111028 36 25182 0 e2-e4 e7-e5 Ng1-f3 Nb8-c6 Bf1-b5 a7-a6 Bb5-a4 Ng8-f6 Ke1-g1 Bf8-e7 Rf1-e1 b7-b5 Ba4-b3 d7-d6 c2-c3 Ke8-g8 h2-h3 Bc8-b7 d2-d4 Rf8-e8 Nb1-d2 Be7-f8 a2-a4 h7-h6 Bb3-c2 Ra8-b8 Bc2-b1 b5-b4 a4-a5 b4xc3
37 30 530 134485 37 25374 0 e2-e4 e7-e5 Ng1-f3 Nb8-c6 Bf1-b5 a7-a6 Bb5-a4 Ng8-f6 Ke1-g1 Bf8-e7 Rf1-e1 b7-b5 Ba4-b3 d7-d6 c2-c3 Ke8-g8 h2-h3 Bc8-b7 d2-d4 Rf8-e8 Nb1-d2 Be7-f8 a2-a4 h7-h6 Bb3-c2 Ra8-b8 Bc2-b1 b5-b4 a4-a5 b4xc3
38 30 619 156336 38 25243 0 e2-e4 e7-e5 Ng1-f3 Nb8-c6 Bf1-b5 Ng8-f6 Ke1-g1 Nf6xe4 Rf1-e1 Ne4-d6 Nf3xe5 Bf8-e7 Bb5-f1 Nc6xe5 Re1xe5 Ke8-g8 d2-d4 Be7-f6 Re5-e1 Rf8-e8 c2-c3 Re8xe1 Qd1xe1 Nd6-e8 Bc1-f4 d7-d5 Nb1-d2 c7-c6 Bf1-d3 g7-g6 Nd2-f3 Ne8-g7
39 30 708 178118 39 25143 0 e2-e4 e7-e5 Ng1-f3 Nb8-c6 Bf1-b5 Ng8-f6 Ke1-g1 Nf6xe4 Rf1-e1 Ne4-d6 Nf3xe5 Bf8-e7 Bb5-f1 Nc6xe5 Re1xe5 Ke8-g8 d2-d4 Be7-f6 Re5-e1 Rf8-e8 c2-c3 Re8xe1 Qd1xe1 Nd6-e8 Bc1-f4 d7-d5 Nb1-d2 c7-c6 Bf1-d3 g7-g6 Nd2-f3 Ne8-g7
40 30 796 197223 40 24758 0 e2-e4 e7-e5 Ng1-f3 Nb8-c6 Bf1-b5 Ng8-f6 Ke1-g1 Nf6xe4 Rf1-e1 Ne4-d6 Nf3xe5 Bf8-e7 Bb5-f1 Nc6xe5 Re1xe5 Ke8-g8 d2-d4 Be7-f6 Re5-e1 Rf8-e8 c2-c3 Re8xe1 Qd1xe1 Nd6-e8 Bc1-f4 d7-d5 Nb1-d2 c7-c6 Bf1-d3 g7-g6 Nd2-f3 Ne8-g7
41 30 871 213803 41 24541 0 e2-e4 e7-e5 Ng1-f3 Nb8-c6 Bf1-b5 Ng8-f6 Ke1-g1 Nf6xe4 Rf1-e1 Ne4-d6 Nf3xe5 Bf8-e7 Bb5-f1 Nc6xe5 Re1xe5 Ke8-g8 d2-d4 Be7-f6 Re5-e1 Rf8-e8 c2-c3 Re8xe1 Qd1xe1 Nd6-e8 Bc1-f4 d7-d5 Nb1-d2 c7-c6 Bf1-d3 g7-g6 Nd2-f3 Ne8-g7
# nodes 2884141 <0% qnodes> time 8806ms nps 327519 eps 0 nneps 24397 tbhits 0
# Tree: nodes 276306 depth 31 pps 24455 visits 213818 qcalls 0 scalls 0
move e2e4
Bye Bye