I had a programm (FTP-trigger) that could send a file every X seconds to an FTP server. For whatever reason that program stopped working for me (varous computers, various Windows OS, no luck). The programm was quite old and the sourcecode is lost)
Does anyone know another program that can send a specific file every X second to a FTP server?
Because of this problem I cant transmit the current tourney
As of this writing it looks like the battle with Stockfish for first place on your list is incredibly close and may come down to one or two elo. An exciting finish is in the works. Komodo 10.4 took top place on both CCRL 40/40 lists (4 cpu and 1 cpu) by a single elo point; if we did so on IPON too someone will call it a conspiracy .
lkaufman wrote:
As of this writing it looks like the battle with Stockfish for first place on your list is incredibly close and may come down to one or two elo. An exciting finish is in the works. Komodo 10.4 took top place on both CCRL 40/40 lists (4 cpu and 1 cpu) by a single elo point; if we did so on IPON too someone will call it a conspiracy .
Right now it is a plus of 1.1% (2180 games). SF8 had exactly the same percentage. That is unbelivable close (pun intended )!
Regarding the "conspiracy, fraud, whatever", someone will do that anyhow .. it is as it is, I am quite relaxed with that after so many years.
Thanks for the new Komodo, the battle up there (top 3, hopefully 4) is much more exciting than the years of boring gridlock we had!
Lyudmil Tsvetkov wrote:Komodo is currently beating Houdini.
but of course, as Jean will say, Houdart already has a +200 elo engine up his sleeve, that he will publish in 3 years' time.
Jean-Baptiste won't say that.
Komodo might reach the #1 spot this time on all list, even by not beating Stockfish 8 directly thanks to contempt. That's still something nice.
The challenge is still to catch up Stockfish dev for the next TCEC. In the meantime, good work by still be able to improving so fast this great engine.
JJJ wrote:
Komodo might reach the #1 spot this time on all list, even by not beating Stockfish 8 directly thanks to contempt. ..
I have no idea were this is comming from. Olease check the "news" on inwoba.de from the 2016.06.20. That was Komodo 10 with default contempt and without. The difference to the second engine (SF7 at that time) changes by 2 elo after 3300. That is totaly necglecable! In contrary, without contempt K10 was able to be above 50% in direct comparision against SF7 )only 220 games!) while it lost with default contempt. Unfortunately for SF7 with the loss against 'K10 contempt 0' is got less points as well which nearly equalized the result (again only 2Elo difference, all within any error bar).
In short: With or without contempt for Komodo doesn't change enough that any human can feel it and rating lists have a hard time proving a difference beyond any doubt because of error bar.
Btw, at 16.6.18 I made a similar test with SF7, just in case someone is interested.
If contempt doesn't change that much the Elo in rating list, maybe the default contempt should be set to 0 instead of 10, the rating list would be more accurate to me. I d prefer to see the real score of Komodo against Stockfish and Houdini instead a lower one due to contempt.
JJJ wrote:If contempt doesn't change that much the Elo in rating list, maybe the default contempt should be set to 0 instead of 10, the rating list would be more accurate to me. I d prefer to see the real score of Komodo against Stockfish and Houdini instead a lower one due to contempt.
I am with you here, but the developers decided different ... and as it really doesn't matter that much ... \_(ツ)_/
.