Life In 19x19 http://lifein19x19.com/ |
|
LZ's progression http://lifein19x19.com/viewtopic.php?f=18&t=15718 |
Page 8 of 21 |
Author: | nbc44 [ Wed Sep 19, 2018 11:18 pm ] | ||
Post subject: | Re: LZ's progression | ||
Last attachment. P.S. to be continued
|
Author: | nbc44 [ Fri Sep 21, 2018 3:06 am ] | ||
Post subject: | Re: LZ's progression | ||
Another test (l0 v15, validation.exe .... --gpu 0 --gpu 1 --noponder -t 12 -q -d -r 10 --timemanage off) #157 vs #178 : a). by visits 8001:3201 (157-178.zip) Code: The first net is worse than the second #157 v #178 ( 110 games) wins black white #157 43 39.09% 21 38.18% 22 40.00% #178 67 60.91% 34 61.82% 33 60.00% 55 50.00% 55 50.00% 110 games played. Status: 1 LLR -3.00809 Lower Bound -2.94444 Upper Bound 2.94444 P.S. more better than last test :^)
|
Author: | nbc44 [ Sun Sep 23, 2018 8:52 pm ] | ||
Post subject: | Re: LZ's progression | ||
Yet another test (l0 v15, validation.exe .... --gpu 0 --gpu 1 --noponder -t 12 -q -d -r 5 --timemanage off) #157 vs #178 : a). by visits 9601:3201 (157-178-5.zip) Code: The first net is worse than the second
#157 v #178 ( 241 games) wins black white #157 112 46.47% 57 46.34% 55 46.61% #178 129 53.53% 66 53.66% 63 53.39% 123 51.04% 118 48.96% 241 games played. Status: 1 LLR -2.95061 Lower Bound -2.94444 Upper Bound 2.94444
|
Author: | nbc44 [ Mon Sep 24, 2018 11:51 pm ] | ||
Post subject: | Re: LZ's progression | ||
So-called "time parity" test - 10 min per game. Code: gogui-twogtp -black "C:\APPS\l0gpu\leelaz.exe --gtp --weights=C:\APPS\net\d351f06e.gz -t 12 --gpu 0 --gpu 1 --noponder" -white "C:\APPS\l0gpu\leelaz.exe --gtp --weights=C:\APPS\net\58d1222a.gz -t 12 --gpu 0 --gpu 1 --noponder" -games 100 -sgffile 157-178-10m -auto -time 10 -komi 7.5 -verbose Part1. #157 (black) vs #178 : +48-52=0
|
Author: | nbc44 [ Wed Sep 26, 2018 12:42 am ] | ||
Post subject: | Re: LZ's progression | ||
Part2. #157 (white) vs #178 : +60-40=0 (???) Finally: #157 vs #178 : +108-92=0
|
Author: | moha [ Wed Sep 26, 2018 5:50 am ] |
Post subject: | Re: LZ's progression |
nbc44 wrote: So-called "time parity" test - 10 min per game. This says almost nothing without mentioning your gpu or average nps.
|
Author: | nbc44 [ Wed Sep 26, 2018 5:55 am ] |
Post subject: | Re: LZ's progression |
moha wrote: nbc44 wrote: So-called "time parity" test - 10 min per game. This says almost nothing without mentioning your gpu or average nps. 2 х 1080ti with Temp. limit 80 C and Power limit 90%. First two moves (part 1 & 2): |
Author: | moha [ Wed Sep 26, 2018 6:43 am ] |
Post subject: | Re: LZ's progression |
Interesting, there seems quite a visit fluctuation between positions (even for the 1st move on empty board, and even outside tree reuse?). Edit: Never mind, I figured it out. Btw, what is the actual visit fluctuation, for example between early and late positions? Does it remain within these 25-30%? |
Author: | nbc44 [ Thu Sep 27, 2018 3:11 am ] | |||
Post subject: | Re: LZ's progression | |||
moha wrote: Btw, what is the actual visit fluctuation, for example between early and late positions? Does it remain within these 25-30%? FYI, "big data" in action . Format - "black~white". 1). by visits
|
Author: | nbc44 [ Thu Sep 27, 2018 3:13 am ] | |||
Post subject: | Re: LZ's progression | |||
2). by node/s
|
Author: | moha [ Thu Sep 27, 2018 7:51 am ] |
Post subject: | Re: LZ's progression |
Thanks. Looking at LZ's reported "visits" some huge swings are expected since that includes tree reuse. But it seems there actually are large fluctuations in n/s as well, which is not subject to tree reuse (it's taken as playouts/time IIRC). Bottom line is, it's not surprising if results with fixed (asymmetrical) visits differ somewhat from fixed per-game time. |
Author: | nbc44 [ Tue Oct 02, 2018 1:19 am ] | |||
Post subject: | Re: LZ's progression | |||
Yet another so-called "time parity" test - 10 min per game. Code: --precision half -t 12 --gpu 0 --gpu 1 --noponder Part1: #157 (black) vs #181 : +32-68=0 Part2: #157 (white) vs #181 : +35-65=0 Finally: #157 vs #181 : +67-133=0 Is it a miracle or just an ordinary bug of "precision half" option? A farewell to #157? P.S. L0-next 29/09/2018
|
Author: | Vargo [ Tue Oct 02, 2018 7:54 am ] |
Post subject: | Re: LZ's progression |
Another brick in the (same) wall #181 v. #157 (visits=12801 for #157 and visits=3201 for 181) #181 wins 7-3 (twogtp v1.4.10, noponder, average length : 237 moves, #157 takes 10% more time than #181) Attachment: Attachment:
|
Author: | nbc44 [ Tue Oct 02, 2018 3:13 pm ] | |||
Post subject: | Re: LZ's progression | |||
Quick answer: Code: C:\APPS\l0gpu\validation.exe -k 157-181-10 -b C:\APPS\l0gpu\leelaz -n C:\APPS\net\d351f06e.gz -o "-g -v 12801 --gpu 0 --gpu 1 --noponder -t 12 -q -d --timemanage off -w" -b C:\APPS\l0gpu\leelaz -n C:\APPS\net\68824bbc.gz -o "-g -v 3201 --gpu 0 --gpu 1 --noponder -t 12 -q -d --timemanage off -w" Code: 6 wins, 4 losses 10 games played. #157 wins 6 - 4 edit. Code: gogui-twogtp -black "C:\APPS\l0gpu\leelaz.exe --gtp --weights=C:\APPS\net\d351f06e.gz -t 12 --gpu 0 --gpu 1 -v 12801 --noponder --timemanage off" -white "C:\APPS\l0gpu\leelaz.exe --gtp --weights=C:\APPS\net\68824bbc.gz -t 12 --gpu 0 --gpu 1 -v 3201 --noponder --timemanage off" -games 5 -sgffile 157-181-visits -auto -komi 7.5 -verbose & gogui-twogtp -black "C:\APPS\l0gpu\leelaz.exe --gtp --weights=C:\APPS\net\68824bbc.gz -t 12 --gpu 0 --gpu 1 -v 12801 --noponder --timemanage off" -white "C:\APPS\l0gpu\leelaz.exe --gtp --weights=C:\APPS\net\d351f06e.gz -t 12 --gpu 0 --gpu 1 -v 3201 --noponder --timemanage off" -games 5 -sgffile 157-181-visits -auto -komi 7.5 -verbose #157 wins 8 - 2 x 2 Any idea?
|
Author: | Vargo [ Tue Oct 02, 2018 9:18 pm ] |
Post subject: | Re: LZ's progression |
We must have the same kind of PC (I have 2x1080Ti, i9-7920X), and except for your -t 12 -q -d , I have the same commands. On mine, a 10min (per side) game lasts in fact around 820 sec, as LZ doesn't use up all its time, and a 12801v3201 visits game lasts about the same time, but slightly favours #157, as it takes 10% more time than 181. 181 won 66% of your 200 games at 10 min. 181 won 7 of its 10 games 181 won 4 of its 10 games 181 won 2 of its 10 games that's 43% of the 30 games, but with ~10% less time than 157 I'd trust your 200 games result. Probably 3x10 games is not enough, or ... maybe it's 181's way of asking for 10% more time |
Author: | nbc44 [ Tue Oct 02, 2018 10:38 pm ] |
Post subject: | Re: LZ's progression |
Due to the fact that your processor is faster, I suggest repeating your last test (10 games) with my settings: Code: validation -k 157-181 -b leelaz -n #157net -o "-g -v 12801 --gpu 0 --gpu 1 --noponder -t 12 -q -d --timemanage off -w" -b leelaz -n #181net -o "-g -v 3201 --gpu 0 --gpu 1 --noponder -t 12 -q -d --timemanage off -w" P.S. I have 2 x 1080ti too and Xeon E5-1650 v4, win10. P.P.S. Benchmark: |
Author: | Vargo [ Wed Oct 03, 2018 8:10 am ] |
Post subject: | Re: LZ's progression |
I don't think that the CPU is important, we have the same GPUs, and so, very similar benchmarks |
Author: | splee99 [ Thu Oct 04, 2018 6:23 pm ] |
Post subject: | Re: LZ's progression |
If you have two GPU's, why don't you try assign GPU0 to 181 and GPU1 to 157? I know this would make the speed slower, but it will make a fair match because some data maybe cached in a GPU during the game. |
Author: | nbc44 [ Fri Oct 05, 2018 6:22 pm ] | |||
Post subject: | Re: LZ's progression | |||
Last so-called "time parity" test - 10 min per side. Code: --precision single -t 12 --gpu 0 --gpu 1 --noponder Part1: #157 (black) vs #181 : +31-69=0 Part2: #157 (white) vs #181 : +53-47=0 Finally: #157 vs #181 : +84-116=0 A farewell to #157!!! P.S. L0-next 29/09/2018
|
Author: | Vargo [ Sat Oct 20, 2018 9:39 pm ] |
Post subject: | Re: LZ's progression |
New network (183) Two 20 game matches #183 v #157 at time parity : Match 1 : LZ 015, GPU : 1x1080, no pondering, 5 min. per side , corresponding to ~800 visits for #183 and to ~2800 visits for #157 #183 wins 11-9 (all games by resignation, about 240 moves per game, and 230" used per side and per game) Match 2 : GPU : 2x1080Ti, no pondering, 7 min. per side , corresponding to ~3400 visits for #183 and to ~12800 visits for #157 #183 wins 12-8 (all games by resignation, about 240 moves per game, and 320" used per side) Finally, 183 is better than 157, even at low time settings (less than 2" per move with a 1080) If someone wants the games... |
Page 8 of 21 | All times are UTC - 8 hours [ DST ] |
Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group http://www.phpbb.com/ |