Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: How should we use computers to train?

Author: Lanny DiBartolomeo

Date: 14:30:55 12/18/98

Go up one level in this thread


On December 18, 1998 at 17:24:43, Komputer Korner wrote:

>On December 18, 1998 at 16:48:26, Lanny DiBartolomeo wrote:
>
>>On December 18, 1998 at 16:24:12, Komputer Korner wrote:
>>
>>>On December 18, 1998 at 15:56:22, Lanny DiBartolomeo wrote:
>>>
>>>>On December 18, 1998 at 12:57:08, Christopher R. Dorr wrote:
>>>>
>>>>>On December 18, 1998 at 03:28:13, Komputer Korner wrote:
>>>>>
>>>>>>On December 18, 1998 at 03:24:21, blass uri wrote:
>>>>>>
>>>>>>>
>>>>>>>On December 18, 1998 at 02:54:53, Komputer Korner wrote:
>>>>>>>
>>>>>>>>On December 17, 1998 at 09:50:25, Christopher R. Dorr wrote:
>>>>>>>>
>>>>>>>>>On December 16, 1998 at 17:11:20, Komputer Korner wrote:
>>>>>>>>>
>>>>>>>>>>This is good news about the CM7 opening book editor, but if they don't also add
>>>>>>>>>>the capability of analyzing in player player mode with the engine showing on
>>>>>>>>>>screen analysis and score eval while taking back and moving forward moves, then
>>>>>>>>>>ChessMaster will still remain a toy program. This is the single most important
>>>>>>>>>>feature in chess programs which all the high end programs have.
>>>>>>>>>>--
>>>>>>>>>>Komputer Korner
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>KK, exactly why do *you* get to decide what's the "single most important
>>>>>>>>>feature"? I've said it before, and I'm saying it again: I'm a USCF Master
>>>>>>>>>(pretty strong and serious), and I haven't used this feature in at least a year.
>>>>>>>>>My best friend is a USCF Master...he doesn't use it either. Our City Champion is
>>>>>>>>>a USCF 2300, and I asked him what he uses his computer for...he uses it to play
>>>>>>>>>games, and to analyse games from his students.
>>>>>>>>>
>>>>>>>>>So tell me....since we three serious, master-level players don't consider this
>>>>>>>>>feature very important *at all*, exactly how is it the 'single most important'
>>>>>>>>>feature?
>>>>>>>>>
>>>>>>>>>Yes, I know that you know chess software, but you simply don't speak for all (or
>>>>>>>>>probably evenb a majority) of computer-using chess players?
>>>>>>>>>
>>>>>>>>>Your obsessing on features that many consider trivial (opening book editor, this
>>>>>>>>>analysis mode) detracts from your toherwise informative and even-handed analysis
>>>>>>>>>of these programs.
>>>>>>>>>
>>>>>>>>>Chris Dorr
>>>>>>>>>USCF Life Master
>>>>>>>>
>>>>>>>>Well then you simply aren't using computers to their fullest extent in helping
>>>>>>>>you study chess. I don't care what your rating is. Even Kasparov uses player
>>>>>>>>player mode with on screen PV analysis to look at positions. He isn't looking at
>>>>>>>>the score evaluation of course but he is certainly looking at the PVs. Buy a top
>>>>>>>>rated prgram like Rebel 10, M-Chess Pro 8 , Junior 5 or the upcoming Hiarcs 7
>>>>>>>>and use them properly to analyze positions and you will begin to understand how
>>>>>>>>to study chess with chess engines. Perhaps an opening book editor isn't that
>>>>>>>>important for some even though it has helped me win more than 1 game, but
>>>>>>>>analysis in player player mode with on screen analysis of PVs IS the most
>>>>>>>>important feature of a chess program. CM 6000 doesn't have that. It could be
>>>>>>>>possible to set up a macro with an add on utility as Richard Fowell says to work
>>>>>>>>around this but since all top chess programs have this feature, then I have the
>>>>>>>>right to call CM6000 a toy program without that feature.
>>>>>>>>--
>>>>>>>>Komputer Korner
>>>>>>>
>>>>>>>I do not think it is a big problem because I can use take back and switch sides
>>>>>>>if I want to see the PV
>>>>>>>
>>>>>>>Uri
>>>>>>
>>>>>>Yes but that slows you down considerably. Since this should be your number 1
>>>>>>activity in studying chess, a lot of time will be wasted.
>>>>>>--
>>>>>>Komputer Korner
>>>>>
>>>>>KK, I don't mean this offensively at all.
>>>>>
>>>>>But what are your qualifications to tell others exactly what their 'number 1
>>>>>activity in studying chess' should be? Specifically, I would like to know
>>>>>
>>>>>1. Are you recognized as a Master by any organization (Candadian, USCF, FIDE,
>>>>>etc.) ? What is your current rating? You don't have to be a Master to be a good
>>>>>teacher, but you do have to have a relatively high rating to impart chess
>>>>>wisdom, and have it taked seriously.
>>>>>
>>>>>2. How many years experience do you have as a trainer or coach (so that we have
>>>>>a basis for evaluating how seriously to take you training recommendtions) ?
>>>>>
>>>>>3. How have your students done in tournament play? How much has their
>>>>>strength/rating increased? In order that we can tell whether or not your
>>>>>training recommendations work.
>>>>>
>>>>>I'm not trying to be sarcastic, or attacking at all. As a reviewer, you are
>>>>>extremely well qualified, but before we give credence to your training advice,
>>>>>I'd like to know where it's coming from.
>>>>>
>>>>>Fair is fair, so I'll answer these questions about me first, since I disagree
>>>>>with your training recommendations, and would advise otherwise.
>>>>>
>>>>>
>>>>>1. I am recognized by the USCF as a Life Master.  My current rating has dropped
>>>>>below USCF 2200, I am currently around USCF 2150 or so. I need 2 more games to
>>>>>have my FIDE rating published. Currently, it would be about 2125.
>>>>>
>>>>>2. I have been professionally teaching chess since 1990. I have worked
>>>>>extensively with both individual students, and with classes and chess camps.
>>>>>
>>>>>3. My students have won state grade level championships (different students) for
>>>>>fifth, sixth, seventh, eight, and ninth grades. One has finished second in the
>>>>>National Junior H.S. championship. One has won the state Junior H.S.
>>>>>Championship. I have worked with students from 5 years old to 80 years old. The
>>>>>average improvement for my students in their first year of lessons with me has
>>>>>been about 300 points USCF, compared to an average of their non-lesson-taking
>>>>>peers of less than 100 points.
>>>>>
>>>>>I say these things not to attack you, or to make others think differently about
>>>>>either you or me, but because you are telling the readers that they shoudl be
>>>>>focused on using the computer to go back and forth through their games as their
>>>>>primary training focus. I don't believe that they should.
>>>>>
>>>>>For the vast majority of USCF 900-1500 students, I believe that a complete
>>>>>evaluation of all tournment games should be their first priority. CM6K does this
>>>>>very well, with it's annotation features. Their second priority should be
>>>>>developing an understanding of how a game 'flows', i.e. how it evolves from an
>>>>>opening to a middlegame, to an ending. This they can do by playing through games
>>>>>from a database in a particulr opening. By playing through many of them quickly,
>>>>>they can start to see where this piece goes, or how this pawn structure mutates
>>>>>into this kind of attack. Simply having the program evaluate positions while
>>>>>going back and forth (what you seem to be advocating) does *nothing* to enhance
>>>>>this goal. These two goals go well together. When they have a decent
>>>>>understanding of these game flows, then they should move on to playing many
>>>>>games against a variety of players rated within 200 points of them (both better
>>>>>and worse) to get a feel for making decisions in the kinds of positions likely
>>>>>to arise from their openings; and then having these games analysed by the
>>>>>program too. Again, CM6K's handicapping and customization features are truly
>>>>>excellent for this. Finally, I believe that graphing their games (ala Lev
>>>>>Alburt) is vital. Compare CM6K's numeric evaluations to those done by the
>>>>>student himself (self-annotation is a requisite), and see what kinds of errors
>>>>>are happening (Is your king getting attacked early in the middlegame frequently,
>>>>>because you leave it in the center too long? Are you getting beaten because you
>>>>>allow bad pawn structures? Are you sacrificing material too readily? Are you
>>>>>missing simple tactics?), and attacking your errors in order of importance.
>>>>>
>>>>>If the average USCF player (current average rating is about 1300-1400), wants to
>>>>>improve quickly, I would advise them to do what I have suggested above, and work
>>>>>through the CM6K tutorials. For most players, I believe this will halp them much
>>>>>more than going back and forth through a single game or a few games, and seeing
>>>>>how a program evaulates the position.
>>>>>
>>>>>I'm interested to hear how you think a developing player should use the computer
>>>>>to help their progress.
>>>>>
>>>>>Thanks,
>>>>>
>>>>>Chris Dorr
>>>>>USCF Life Master
>>>>
>>>>Very Excellent advice! I have always believed in this type of training and even
>>>>train this way still and forever. I do alot of chess puzzles
>>>>(positional,tactical,endgame)I play solitair chess and I am at master strength
>>>>and feel my improvement is still growing! I play cm6000 standard time controls
>>>>after, have it analyse my games and I go over these games and search for a
>>>>related weakness be it trouble against the bishop pair or underestimating things
>>>>in situations then I fix these things.  I  write down my thought process and see
>>>>if there is a "wrong pattern" in my thinking. And once I attain IM I will then
>>>>go into serious book-up or if I feel by not booking up is slowing down my chess
>>>>growth. I told this as a reply to K K in his posts and he called me  a lazy
>>>>chessplayer and said I didnt really want to improve(im more words than not)
>>>>of course I failed to mention to him I was Master But that shouldnt Matter !
>>>>I spend alot of time, too much to be booking up when I feel I should be putting
>>>>my energy into learning and UNDERSTANDING the hidden details of the game.
>>>
>>>You are not using a chess program to it's fullest extent. After using player
>>>player mode with on screen engine analysis of a strong positional engine to go
>>>over your games, you will soon realize that you have missed out on an incredible
>>>resource that computer chess programs can provide. Your way of training has some
>>>merit but it can't approach the "Truth to the position approach" that player
>>>player mode with taking back and moving forward with on screen engine analysis
>>>provides. All GMs will tell you this. You must study your games intensively
>>>using this approach. In fact all chess teachers use this approach when looking
>>>at positions with their students. There is no substitute for it and I am
>>>astonished that you would argue otherwise.
>>>Komputer Korner
>>
>>I am not arguing the point if it is good or not but the way Im doing it will be
>>just as good what did these GMs do before the chess programs were out? how did a
>>master increase to GM before the chess programs were out? The same way I am,
>>going over my games finding weaknesses, here and there getting advice from
>>stronger players and when I play solitair chess I can see how far off I am but
>>other times I can see how close I am this shows me what positions Im good in or
>>Bad in and can fix if you still find faults in your play and can fix it then
>>your still growing its if you find faults and cant fix or the faults are so
>>suttle you cant put your finger on it but at this point you need a GM or
>>stronger player to show you this and I dont believe a chess engine would be able
>>to show you this it cant explain to you the reason why.
>
>Of course if you are a lower rated player then you should study all the books
>and tutorials you can and if you have a human teacher then use him but what if
>you don't have a teacher? And if you have read almost all the chess books out
>there as I have, what do you do next? Go over your tournament games of course.
>My method uses all your methods plus it takes advantage of the new power of
>chess engine strength. Your attitude is that chess engines have nothing to teach
>you. If you are fighting a war and somebody offered a way to bombard the other
>side (not a perfect bombardment but one that is effective nevertheless) then
>wouldn't you use the new weapon. You sound like a luddite who is against
>technology. All I am saying is use  the tools available and CM6000 does not
>provide the best tool short of getting a human master's advice. It's tutorials
>are great but if your rating is beyond the tutorials then CM6000 doesn't do the
>job that other higher end programs do. Use one and actually go over your
>tournament games using the back and forward method and you will realize what I
>mean.
>--
>Komputer Korner
whats a luddite? :))HAHAHA!! not being sarcastic it just sounds funny!



This page took 0 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.