Question to the members of the ranking lists..

Discussion of computer chess matches and engine tournaments.

Moderators: hgm, Harvey Williamson, bob

Forum rules
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
Steve B
Posts: 3697
Joined: Tue Jul 31, 2007 2:26 pm
Contact:

Re: Question to the members of the ranking lists..

Post by Steve B » Fri Jul 02, 2010 11:26 pm

lkaufman wrote:
Steve B wrote:
lkaufman wrote: If you ever have any desire to sponsor any other computer-human match of any type, just message me.
how about this for an idea??
we have played several matches here pitting dedicated computers vs the forum
these games can be found in the "Tournament Forum"

1 game match
my Conny 3.6 v a real GM
QR odds..GM has white
TC 3 minutes per move
moves to be posted here
GM..YOU
:P

we would have to be on the honor system of course
i certainly trust you to not use any engine help and to time yourself
i could send you the computer after the game ends and you could check each move that was played
with pondering OFF i doubt very highly that the computer would make different moves (even though there might be some built in random factor the computer uses)
i would pay round trip postage for shipping the computer back and forth

appearance fee for you..win lose or draw .. to be discussed by email
game to begin once you receive the appearance fee

moves to be posted here (i will set up a thread for it)at least once per day but i would not claim a win by time forfeit so not strictly enforced
more of a guideline really

if you are game ..please let me know
Best Regards
Steve
Yes, I am agreeable in principle to this. But I am only about a 2400 (FIDE) GM, not a 2700 as you were talking about. I recall that when SuperConstellation came out I was able to do well giving it knight odds, though I don't recall the exact level I tested. Since your machine is somewhat weaker, maybe I have some chance at rook odds though I'm probably an underdog. There would be no need to send me the machine, you clearly have no motive to cheat in this experiment. Anyway if you still want to go ahead with this, message me about the details.
As a Training match for the game against you i pit the Conny 3.6 against that 2700 PC Engine
same game conditions as stated above except that pondering was ON for Conny because i cant turn it off and didn't want to turn the computer off each move
come to think of it..as you only have to post one move per day i guess you have to not think about the position in between moves
pondering off.. on your part so to speak
:P

the Engine had white and played without its QR
TC -3Min Avg.


while i cant show the entire game as it would give away too much i can show the final position which resulted in a win for Conny

[d] 3N4/8/2p5/3k4/6pp/2r5/7P/6K1 b - - 0 48

when your game has concluded i will post the entire PGN
just a bit more proof that a R can overcome a 1000+ Elo disadvantage

The Pressure is On Regards
Steve

lkaufman
Posts: 3647
Joined: Sun Jan 10, 2010 5:15 am
Location: Maryland USA
Contact:

Re: Question to the members of the ranking lists..

Post by lkaufman » Fri Jul 02, 2010 11:55 pm

I'll just say that playing the objectively best move at every turn is not the best strategy when giving rook odds. That's my one advantage in this context compared to a 2700 engine.

Steve B
Posts: 3697
Joined: Tue Jul 31, 2007 2:26 pm
Contact:

Re: Question to the members of the ranking lists..

Post by Steve B » Sat Jul 03, 2010 11:30 am

lkaufman wrote:I'll just say that playing the objectively best move at every turn is not the best strategy when giving rook odds. That's my one advantage in this context compared to a 2700 engine.
actually i think it might be harder for the Conny to draw or win against you then a 2700 Engine ..not only are you a GM but it is not lost on me that at one time you were probably the worlds foremost authority on dedicated computers and knew the nuances of almost every machine on the market

hopefully now that your main focus has been PC Engines for many years this information has all but been forgotten by you
in fact.. im banking on it

Rex Chess Regards
Steve

lkaufman
Posts: 3647
Joined: Sun Jan 10, 2010 5:15 am
Location: Maryland USA
Contact:

Re: Question to the members of the ranking lists..

Post by lkaufman » Sat Jul 03, 2010 1:50 pm

Well, I remember enough to say that Connie is basically a full-width searcher doing 4-5 plies in the middle game in 3 minutes, so if the oopportunity arises for a tactic that would require a six ply search by Connie to see, I can expect Connie to miss it. But making such opportunites arise won't be easy at rook odds.

Steve B
Posts: 3697
Joined: Tue Jul 31, 2007 2:26 pm
Contact:

Re: Question to the members of the ranking lists..

Post by Steve B » Sat Jul 03, 2010 2:06 pm

lkaufman wrote:Well, I remember enough to say that Connie is basically a full-width searcher doing 4-5 plies in the middle game in 3 minutes, so if the oopportunity arises for a tactic that would require a six ply search by Connie to see, I can expect Connie to miss it. But making such opportunites arise won't be easy at rook odds.
:shock:

any chance we can play a game of GO instead?

Sigh Regards
Steve

IWB
Posts: 1539
Joined: Thu Mar 09, 2006 1:02 pm

Re: Question to the members of the ranking lists..

Post by IWB » Sat Jul 03, 2010 4:52 pm

Hi

The only reason to argue for you is to argue.
Rainer Marian wrote: There is a lack of an explanatory statement.
Wrong. the term 'correleation' is not (only) a mathematical one! If someone takes it only as such it is a missuse of the word.
Rainer Marian wrote: Correlation is a certain method of measuring the associations between ....
Whatever yo try to make out of it, 'correlation' is just a relation between things. The word as such has nothing to do with mathematics.) Especially not, if you read the questionable sentence as a whole and you are willing to understand it!)
Rainer Marian wrote: ... BTW, your quoted wiktionary explanation is so general, abstract and without any obligation...
Rainer Marian wrote:
My quoted explenation is as the word is - correct: general, abstract and without any obligation!
Rainer Marian wrote: ... that you can correlate by this definition a wedding with a funeral or a chess game with a striptease.
Of course you can compare a wedding with a funaral (and thats something I mean without double sense! :-D ) even if you just compare to say that it does'nt correlate.
Rainer Marian wrote: Solution
What you really could have meant, but you haven't verbalized it ...
Please leave it on my what I meant to verbalize - and I did exactly that!

How is the german saying: Ich bin nur verantwortlich für das was ich sage, nicht für das was du verstehst! (I am only repsonsible for what I am saying, not for what you understand!) And in this case please see my first sentence in this post.

Bye
Ingo

Steve B
Posts: 3697
Joined: Tue Jul 31, 2007 2:26 pm
Contact:

Re: Question to the members of the ranking lists..

Post by Steve B » Sat Jul 03, 2010 8:51 pm

Played a second training game(40/2) against the PC Engine

this time the engine(White) managed a draw

FINAL POSITION
[D] 8/8/8/6K1/2Q5/6B1/2k5/q4r2 b - - 0 76

PGN will follow upon completion of GM Kaufman v Novag Constellation 3.6 Match Regards
Steve

rainhaus
Posts: 143
Joined: Sun Feb 01, 2009 6:26 pm

Re: Question to the members of the ranking lists..

Post by rainhaus » Wed Jul 07, 2010 4:42 pm

The controversial intro of the Ipon list :
The Elo numbers shown here do not correlate to human ratings
________________________________________________
IWB wrote:Hi
The only reason to argue for you is to argue.
A lame excuse, when someone runs out of arguments. I know this kind of discussion from my mother in-law :) No, quite the contrary. I try to remain constructively and I want to spread my humble knowledge. Perhaps you are reading too selective and superficial. See the next point.
IWB wrote:... the term 'correleation' is not (only) a mathematical one! If someone takes it only as such it is a missuse of the word.
Who reads carefully has a clear advantage! Of course I pointed out the common and the mathematical/statistical use of the term "correlation". There are a lot of terms which are defined specifically and for every day use. However, a ranking list is a statistical object and therefore it's obvious to use the terms in the statistical sense. Help me, I don't understand why this is not evident for the author of a rating list!
See the answers of Larry, Bob and Dann, they also were puzzling about what you could have meant To avoid such errors you could attach a text like: please do not use the sentence in the statistical sense, because I haven't any idea of it.., or, I'm an advocate of the everyday speech.., or, whatever is the reason why. But cancelling would be the best, because the information is wrong in any case. See the next point.
IWB wrote:Whatever yo try to make out of it, 'correlation' is just a relation between things. The word as such has nothing to do with mathematics.) Especially not, if you read the questionable sentence as a whole and you are willing to understand it!)
Oh yeah, good old Aristoteles: "The whole is more than the sum of its parts". I like it and this principle almost always fits. Let's look to the sense of your key sentence, without to be bound to the words. Let's take "relation " instead of "correlation". Then your key sentence reads as follows: The Elo numbers shown here have no relation to human ratings It doesn't change anything. You can read it as a whole or by analyzing the terms. You can call it relation, correlation, association, but it is simply not the truth. Because of the simple reason that there IS a strong relation between the Elo values of your list and human ratings. I've told you this very simple relation. Once more: the higher the Elos the higher the playing strength. Is this a relation, yes or no? Does it concern humans and engines, yes or no? Is this a correlation, yes or no?
I've told you too, what probably is your misunderstanding. If you let my proposals come to nothing, it's not my pigeon (for a German: it's not my beer)
IWB wrote: Of course you can compare a wedding with a funaral (and thats something I mean without double sense! :-D ) even if you just compare to say that it does'nt correlate.
Well, all right, you really can correlate all with everything, because one of the three possibilities "+" ,"-" or "0" is always true. Comparing weddings and funerals you'll probably get a strong positive correlation with the variables tears, handkerchiefs, friends, speeches, cravats. With the variables laughing, dancing and black and white colours you'll get a very negative correlation. But attention please, in some continents and countries people is also dancing an wearing white at a funeral! With Elos, zombies and cucumbers you'll get probably 0 correlation. Do you know the refrain of the correlation song? If there is no relation, you'll find 0 correlation and a lack of association...
These smiling reflections bring us back to the fact we should constantly remind, there is strong positive linear correlation between human ratings and engine ratings. Don't disclaim the linear correlation, it's a pleasure and a continuous wellspring of surprising connections :)
IWB wrote: Please leave it on my what I meant to verbalize - and I did exactly that!
Well, I think, "please leave it to me what I wanted to verbalize..", sounds a little bit better. But heaven forbid, it's only another well-meant correction of verbalisation ;)
IWB wrote:How is the german saying: Ich bin nur verantwortlich für das was ich sage, nicht für das was du verstehst! (I am only repsonsible for what I am saying, not for what you understand!) And in this case please see my first sentence in this post.
Bye
Ingo
I don't mind. But often I prefer my variation: I'm only responsible for that what I write but not for that what you don't know :mrgreen:

Best
Rainer
Thread viewImage
flat view
is a bad view
without thread view

Post Reply