Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: The law of diminishing returns

Author: José Carlos

Date: 16:05:35 07/13/02

Go up one level in this thread


On July 13, 2002 at 17:16:05, Rolf Tueschen wrote:

>On July 13, 2002 at 16:57:51, José Carlos wrote:
>
>>On July 13, 2002 at 15:09:18, Rolf Tueschen wrote:
>>
>>>On July 13, 2002 at 08:02:09, José Carlos wrote:
>>>
>>>>On July 13, 2002 at 07:15:53, Rolf Tueschen wrote:
>>>>
>>>>>On July 13, 2002 at 07:09:02, José Carlos wrote:
>>>>>
>>>>>>On July 13, 2002 at 05:35:24, Uri Blass wrote:
>>>>>>
>>>>>>>On July 12, 2002 at 19:16:31, José Carlos wrote:
>>>>>>>
>>>>>>>>On July 12, 2002 at 14:56:11, Ed Schröder wrote:
>>>>>>>>
>>>>>>>>>Hi CCC,
>>>>>>>>>
>>>>>>>>>In Rebel I maintain a statistic file, on every iteration a counter is
>>>>>>>>>incremented with 1 (see column 2) representing the iteration depths Rebel has
>>>>>>>>>searched. When a new best move is found a second counter is incremented with 1
>>>>>>>>>(see column 3) representing how many times a new best move has been found on the
>>>>>>>>>given iteration depth, between brackets the percentage is calculated.
>>>>>>>>>
>>>>>>>>>As you can see the very first plies Rebel often changes to new best moves,
>>>>>>>>>however when the depth increases and increases the chance Rebel will change its
>>>>>>>>>mind drops and drops. From 16 plies on the chance a new better move is found is
>>>>>>>>>below 2%.
>>>>>>>>>
>>>>>>>>>I wonder what this all means, it is still said (and believed by many) that a
>>>>>>>>>doubling in computer speed gives 30-50-70 elo. That could be very well true for
>>>>>>>>>lower depths but the below statistic seem to imply something totally different,
>>>>>>>>>a sharp diminishing return on deeper depths.
>>>>>>>>>
>>>>>>>>>Interesting also is colum 4 (Big Score Changes), whenever a big score difference
>>>>>>>>>is measured (0.50 up or down) the percentage is calculated. This item seems to
>>>>>>>>>be less sensitive than the change in best move. However the maintained "Big
>>>>>>>>>Score Changes" statistic is not fully reliable as it also counts situations like
>>>>>>>>>being a rook or queen up (or down) in positions and naturally you get (too) many
>>>>>>>>>big score fluctuations. I have changed that and have limit the system to scores
>>>>>>>>>in the range of -2.50 / +2.50 but for the moment have too few games played to
>>>>>>>>>show the new statistic.
>>>>>>>>>
>>>>>>>>>Anyway the number of positions calculated seem to be more than sufficient (over
>>>>>>>>>100,000) to be reliable. The origin came from extensive testing the latest Rebel
>>>>>>>>>via self-play at various time controls.
>>>>>>>>
>>>>>>>>  Hi Ed, if I get this right, the second column (moves searched) is the number
>>>>>>>>of positions in which the program has reached the depth given by column 1. If it
>>>>>>>>was really "moves", there would be about 3x in depth 2 than in depth 1.
>>>>>>>>  Then the idea is that many more changes happen in low depths because the
>>>>>>>>program is there many more times, so I (ignoring "Big Changes") calculated a
>>>>>>>>couple of other numbers:
>>>>>>>>  The ratio moves changes / moves searched and the relative % of changes from
>>>>>>>>ply to ply:
>>>>>>>>
>>>>>>>>                 SEARCH OVERVIEW
>>>>>>>>                 ===============
>>>>>>>>
>>>>>>>>  (A)     (B)            (C)           (D)             (E)
>>>>>>>>Depth    Moves          Moves     Moves Changed /   rel % of changes from
>>>>>>>>       Searched        Changed    Moves Searched    ply n-1 to n
>>>>>>>>
>>>>>>>> 1     113768         0 =  0.0%        0
>>>>>>>> 2     113768     44241 = 38.9%    0.388870333
>>>>>>>> 3     113768     34262 = 30.1%    0.30115674        77.44%
>>>>>>>> 4     113194     32619 = 28.8%    0.288168984       95.69%
>>>>>>>> 5     113191     30697 = 27.1%    0.271196473       94.11%
>>>>>>>> 6     108633     28516 = 26.2%    0.262498504       96.79%
>>>>>>>> 7     108180     25437 = 23.5%    0.235135885       89.58%
>>>>>>>> 8     102782     22417 = 21.8%    0.218102391       92.76%
>>>>>>>> 9      82629     15400 = 18.6%    0.186375244       85.45%
>>>>>>>>10      59032      9144 = 15.5%    0.154899038       83.11%
>>>>>>>>11      39340      5183 = 13.2%    0.131748856       85.05%
>>>>>>>>12      23496      2350 = 10.0%    0.100017024       75.91%
>>>>>>>>13      12692       957 =  7.5%    0.075401828       75.39%
>>>>>>>>14       6911       396 =  5.7%    0.057299957       75.99%
>>>>>>>>15       4032       193 =  4.8%    0.047867063       83.54%
>>>>>>>>16       2471        72 =  2.9%    0.029138001       60.87%
>>>>>>>>17       1608        26 =  1.6%    0.016169154       55.49%
>>>>>>>>18       1138        17 =  1.5%    0.014938489       92.39%
>>>>>>>>19        921         6 =  0.7%    0.006514658       43.61%
>>>>>>>>20        795         7 =  0.9%    0.008805031      135.16%
>>>>>>>>21        711         1 =  0.1%    0.00140647        15.97%
>>>>>>>>22        636         2 =  0.3%    0.003144654      223.58%
>>>>>>>>23        574         5 =  0.9%    0.008710801      277.00%
>>>>>>>>24        507         1 =  0.2%    0.001972387       22.64%
>>>>>>>>25        451         3 =  0.7%    0.006651885      337.25%
>>>>>>>>26        394         1 =  0.3%    0.002538071       38.16%
>>>>>>>>27        343         2 =  0.6%    0.005830904      229.74%
>>>>>>>>28        296         2 =  0.7%    0.006756757      115.88%
>>>>>>>>29        269         0 =  0.0%    0                  0.00%
>>>>>>>>
>>>>>>>>  Column (D) means the probability at a certain position at a certain depth to
>>>>>>>>get a change, according to your data, for a random position (I assume you chose
>>>>>>>>random positions, because this data comes from real games).
>>>>>>>
>>>>>>>No
>>>>>>>
>>>>>>>I assume that the positions that was searched to big depthes like 16 are only
>>>>>>>positions that the program had enough time to search in the game to depth 16.
>>>>>>>
>>>>>>>These positions are not random positions from games.
>>>>>>>I expect in random positions from games to see at least 10% changes at depth 16.
>>>>>>>
>>>>>>>Uri
>>>>>>
>>>>>>  It's interesting that Ed, who has been doing chess programming for a lot of
>>>>>>years rely on statistical data, and you, absolute newbie to chess programming
>>>>>>can 'expect'. Quite amazing.
>>>>>>
>>>>>>  José C.
>>>>>
>>>>>Very telling about your lack of knowledge about interdisciplinary thinking.
>>>>>
>>>>>Rolf Tueschen
>>>>
>>>>  Well, you needed several hundred posts from Dann to understand the simple
>>>>concept of elo ratings. Lack of knowledge is easy to solve, while lack of
>>>>intelligence is a real problem.
>>>>  BTW, interdisciplinary thinking has nothing to do with validating intuitions
>>>>through experiments.
>>>>
>>>>  José C.
>>>
>>>Your habits are a bit strange for CCC. You want to insult people for their
>>>intelligence? Didn't you know that this is out of fashion?
>>
>>  Did you feel insulted? Oh, sorry, I didn't insult you, really.
>>
>>>Also you cannot prove
>>>your visions.
>>
>>  Visions? I don't have visions. Maybe you take me for someone else ?!
>>
>>>But I can prove where you lack of knowledge. Look at this:
>>>
>>>How do you know if or when I understood Elo system? Dann didn't
>>>explain anything to _me_,
>>
>>  Don't feel bad because Dann had to explain that to you. It can happen to
>>everybody.
>>
>>>He was the only one having the courage to give his verdict about SSDF
>>>Elo system - _with_ me! We two the only ones. And you were dreaming of his role
>>>as _my_ teacher? That's funny.
>>
>>  I'm glad you enjoied Dann's lessons. Dann is very good at that. I also always
>>enjoy his posts.
>>
>>>You do not  understand what validity means... ;-)
>>
>>  Good argument!
>>
>>>You have no idea of what interdisciplinary means too.
>>
>>  Damn, you leave me without words!
>>
>>>You are the typical expert
>>>with narrow views.
>>
>>  Thanks for calling me expert... bah, just a little degree in computer science
>>and a few publications don't make me an expert...
>>
>>>Do not insult Uri.
>>
>>  I didn't. He knows it.
>>  BTW, do you feel the need to defend him? Don't you think he is capable to
>>defend himself? I think it's you who is insulting Uri.
>>
>>>Because he knows a lot about chess.
>>
>>  The first thing where we agree! Cheers!
>>
>>>Know
>>>what I mean? Chess is the basis for computerchess. :)
>>
>>  Words of wisdom...
>>
>>>Only interdisciplinary help could enlighten you. If you have questions, please
>>>tell me, I'll try to do my best for you.
>>>
>>>Rolf Tueschen
>>
>>  Thank you very much. I'll ask you anything I don't understand.
>>
>>  José C.
>
>No reason to become so upset only because I told you not to insult Uri.
>You have insulted him on his lack of intelligence

  Uri knows I didn't. It seems _you_ are not capable to understand. I'm sorry,
I'm not gonna explain _you_ what I said to Uri. He understood. That's enough.
Please, stop defending him from nothing.

>and your argument was a very weak one.
>Uri must be wrong because he told you something different to Ed Schröder.

  He understood me. Everybody did except you.

>As I
>told you then - you should take some lessons in interdisciplinary interactions.

  I read it the first time. No need to repeat it over and over. You seem like a
parrot.

>That would also cure you from the hybris of seeing yourself better (or perhaps
>on the top) than Uri only because you have "programmed" a few things.

  Oh, that's it. You see yourself worse (or maybe on the bottom). That's your
problem then, don't blame me on it.

>BTW since when you believe in the magic of getting rid of some false guesses,

  Now you tell me what I believe in? You read my mind? Don't bother, there's
nothing interesting inside :)

>you wrote, just with a little irony? Are you no longer responsible then
>for what
>you've said? Who taught you such strange nonsense? ;-)

  Interesting fantasy.

>Rolf Tueschen

  As this has become completely off topic, if you want to continue the
discussion email me. Otherwise, feel free to post whatever you want here. I
won't answer. So, you have the chance to say the last word. Use it.

  José C.



This page took 0.01 seconds to execute

Last modified: Thu, 15 Apr 21 08:11:13 -0700

Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.