World Draughts Forum

BertTuyt wrote: ↑

Sun Mar 22, 2020 16:28

Sidiki, the databases (.cprx) are from 2015, only the DB64Header.cprx (17-8-2019) has a new format.

Bert

Thank Bert,
I downloaded all the files, i will try a dxp games.

Thank again.

Sidiki

Yves, then you might have an issue with the preload, as you do not have sufficient memory. Can you try to change the damage.ini file (directory Engines) into: variant = normal book = 0 threads = 6 tt-shared = 1 tt-size = 25 bb-cache = 1 bb-size = 7 bb-preload = -1 or try, variant = normal book = 0 ...

I have released a new version of kingsrow, as well as a version that uses the hub interface, and a hub tournament manager. The hub version of kingsrow is the identical search engine to the full UI version, but without a UI. "Tournament manager'" is kind of misnomer right now, as it only controls ma...

Herewith the last update of my test. Also here saturation mode, although I expect that with an even higher win/lose rate, the curve will go upwards again. But will not further test this. Games W D L U T E 10000 79 76 0 3 158 195 10000 31 127 0 0 158 69 10000 22 135 0 1 158 49 10000 25 132 0 1 158 5...

Sidiki, to quantify ELO differences like 10, you need indeed far more games. However when multiple matches show the same pattern, I'm confident that Damage 15.3 is still weaker as Scan 3.1, ballpark figure 10 ELO. I will further try to optimize the evaluation function, and will certainly before end...

Krzysztof, re reading my older emails. I will issue a version which works on your 16 Core Threadripper. As you can read in other posts, Im now working on getting closer to Kingsrow/Scan. Based on the simulations (DXP Matches), I guess Damage 15.3 is around 10 ELO points weaker compared with these (...

Why ? We are speaking of neural Network many computer or super computers working together. Do you know what's a petaflop, Krzysztof Grzelak ? You saw the link of the proof of what Rein said and he confirmed what Bert said about the 1700 years computing of " a standard computer" as i don't know, you...

Good reply Rein, All this prove again that supercalculators are very very powerfull. I remember a post, sorry i haven't the link that spoken of a french supercalculator, not the best, and they said that the data that this machine calculate in Day, will be reached by all the humans with a calculator...

You do understand that Google/Deepmind used 1700 years worth of computing *in parallel* in order to achieve all that in 4 hours? You trust too much Google. The 1700 years of computing comes from Gian-Carlo Pascutto, author of Leela-Zero, the open source program that actually reproduced the Alphazer...

Krzysztof Grzelak wrote: ↑

Sat Feb 22, 2020 11:21

Rein Halbersma wrote: ↑

Fri Feb 21, 2020 18:30

You do understand that Google/Deepmind used 1700 years worth of computing *in parallel* in order to achieve all that in 4 hours?

You trust too much Google.

I am asking myself if A0 can beat RedFish that crashed LC0

I still do not believe that there is a huge ELO gain possible beyond Scan / Kingsrow with normal time settings. But advanced approaches like A0, can learn us much about the relevant features (like material value man/king temp, left right balance, game phase) in draughts, which are now (more or less...

As stated in a previous post, I will do now some tests to measure ELO as a function of games with increased win/loss ratio. I used the same data set, but only used 1 out of X of the draw games. The test now running is with X = 3, so only on out of 3 draws are used, effectively increasing the decisi...

Cool Rein,
It's so the big difference between A0 and our NN patterns.
If we own the same multi computers than Google we must reach this kind of super draughts program also by self learning renforcement.!?
Thank again..
We are waiting for this evening Damage results.

Sidiki

OK, i thought that it's was the same things with the weights.
Thank

I wanted to know how the strength of the Damage engine scales with the number of learning games used to calculate the weights for the evaluation function. For this purpose I started several 158 game DXP matches, 2 Min/Game. Further settings as mentioned in the first post. For the training set I onl...