open source battle IvanHoe v Stockfish at long time-control
Moderators: hgm, Rebel, chrisw
-
- Posts: 3309
- Joined: Wed Mar 08, 2006 8:15 pm
Re: open source battle IvanHoe v Stockfish at long time-cont
I have feeling 200 games was overkill. In first 100 games SF won 52-48 and last 100 it won 50,5-49,5. Did it give any new information to us, I doubt..
Jouni
-
- Posts: 10121
- Joined: Thu Mar 09, 2006 12:57 am
- Location: van buren,missouri
Re: open source battle IvanHoe v Stockfish at long time-cont
Thanks for your time and work Pal. Some nice games to look at.PawnStormZ wrote:
Standings after game 200
Stockfish 2.2.2 JA SSE42 102.5 + 34 = 137 - 29
IvanHoe999946h 97.5 + 29 = 137 - 34
My 500 games at 5/3 with the latest 32 bit versions came out
about the same.
These two engines must be twins. .
What have you got lined up for the next round of games.
Best to you,
Gerold.
-
- Posts: 880
- Joined: Mon Feb 15, 2010 6:43 am
open source battle IvanHoe v Stockfish at long time-cont
Hi George
You have mentioned several times about this not being the "best" or "strongest " IvanHoe version. Please post the link here (no e-mail, no pms) to the "strongest" Ivan engine version, and I will be sure to use it in my next match.
This match is now finished, so you will not need to look at it on page 1 for too much longer.
Take care
You have mentioned several times about this not being the "best" or "strongest " IvanHoe version. Please post the link here (no e-mail, no pms) to the "strongest" Ivan engine version, and I will be sure to use it in my next match.
This match is now finished, so you will not need to look at it on page 1 for too much longer.
Take care
-
- Posts: 880
- Joined: Mon Feb 15, 2010 6:43 am
open source battle IvanHoe v Stockfish at long time-cont
Hi Jouni
I thought that the 200 games was just a better "sample size". At 1 point SF was up by 9 points so the longer time allowed them to even out a bit. With only 5 points separating them after 200 games, I wonder what the rating difference is?
Take care
I thought that the 200 games was just a better "sample size". At 1 point SF was up by 9 points so the longer time allowed them to even out a bit. With only 5 points separating them after 200 games, I wonder what the rating difference is?
Take care
-
- Posts: 880
- Joined: Mon Feb 15, 2010 6:43 am
open source battle IvanHoe v Stockfish at long time-cont
Hi Gerold
Glad that you enjoyed the match. Along the lines of what you said about your similar result between these 2 at shorter time, there is much talk about "scaling" and how different time controls favor different engines.
I was thinking about playing a RR of 6 or so of the top free engines at maybe game in 2 min plus 1 sec per move, followed by the same group, with the same openings, at the long control. Then we could see if there are any big differences on the same pc, under the exact same conditions, except for the time.
Does anyone have an idea of how many games need to be played with 6 engines to give a "decent" sample size? Of course at the short control 1,000 games is not a problem, but George (and others?) would have my head if I tried to post 1K long games!
Take care
Glad that you enjoyed the match. Along the lines of what you said about your similar result between these 2 at shorter time, there is much talk about "scaling" and how different time controls favor different engines.
I was thinking about playing a RR of 6 or so of the top free engines at maybe game in 2 min plus 1 sec per move, followed by the same group, with the same openings, at the long control. Then we could see if there are any big differences on the same pc, under the exact same conditions, except for the time.
Does anyone have an idea of how many games need to be played with 6 engines to give a "decent" sample size? Of course at the short control 1,000 games is not a problem, but George (and others?) would have my head if I tried to post 1K long games!
Take care
-
- Posts: 880
- Joined: Mon Feb 15, 2010 6:43 am
open source battle IvanHoe v Stockfish at long time-control
The 200 games are available here...
http://www.datafilehost.com/download-9efabfcb.html
-
- Posts: 3245
- Joined: Thu Mar 09, 2006 9:10 am
Re: open source battle IvanHoe v Stockfish at long time-cont
Thanks !PawnStormZ wrote:
The 200 games are available here...
http://www.datafilehost.com/download-9efabfcb.html
My engine was quite strong till I added knowledge to it.
http://www.chess.hylogic.de
http://www.chess.hylogic.de
-
- Posts: 1033
- Joined: Sat Feb 04, 2012 10:03 pm
-
- Posts: 10121
- Joined: Thu Mar 09, 2006 12:57 am
- Location: van buren,missouri
Re: open source battle IvanHoe v Stockfish at long time-cont
Thanks for the reply Pal.PawnStormZ wrote: Hi Gerold
Glad that you enjoyed the match. Along the lines of what you said about your similar result between these 2 at shorter time, there is much talk about "scaling" and how different time controls favor different engines.
I was thinking about playing a RR of 6 or so of the top free engines at maybe game in 2 min plus 1 sec per move, followed by the same group, with the same openings, at the long control. Then we could see if there are any big differences on the same pc, under the exact same conditions, except for the time.
Does anyone have an idea of how many games need to be played with 6 engines to give a "decent" sample size? Of course at the short control 1,000 games is not a problem, but George (and others?) would have my head if I tried to post 1K long games!
Take care
Be looking forward to the next games you play.
At the short control would be nice to have 2k-3k games.
Best,
Gerold.
P.S. Note the number of hits you had on these games. Nice.
-
- Posts: 4790
- Joined: Sat Mar 11, 2006 12:42 am
Re: open source battle IvanHoe v Stockfish at long time-cont
PawnStormZ wrote: Hi George
You have mentioned several times about this not being the "best" or "strongest " IvanHoe version. Please post the link here (no e-mail, no pms) to the "strongest" Ivan engine version, and I will be sure to use it in my next match.
This match is now finished, so you will not need to look at it on page 1 for too much longer.
Take care
Look, pal what I am saying is no surprise on Immortal- rather just common knowledge. I will just mention a few that can demolish competition that 9.46h cannot.
Ivanhoe B47f0.2
T52E
B46e
9.47cWHB
B47d
B47e
46h
pal- I could name 4 or 5 more, but what's the point. I tested them all ag. same opponents (Rybka, Stockfish, Komodo, Critter, Fritz 13) more than once. 9.46h compared to the ones above and 4 or 5 more- is relatively not very strong. What can I say.
I don't mean at all to be ugly when I say this: it's your tournament. If I was running a version and 10 more were stronger- I would hope someone was nice enough to point that out to me- for sure. But that does not include me to then expect the person who told me that to go running around finding all the versions for me.
Best,
george