Author: Albert Silver
Date: 11:20:39 01/06/04
Go up one level in this thread
>>This suggests you thought it ONLY ran in automatic pilot, else why the
>>complaint?
>
>The _only_ "complaint" (I would call it a comment) I have with CA in _this_ area
>is regarding limited number of engines support. You can sit and manually control
>analysis in almost any GUI, but in some you have more engines available.
Ok, so the suggestion for future builds or versions would be to include the
possibility to work with more engines at the same time on this. Although, I'm
certain Victor or someone else on the CA development team is reading this, It's
a good suggestion, and I'll forward it.
>>>salvage draws. Blass also did not just set up his favourite program and let it
>>>make all his moves. He used multiple engines and multiple analysis methods. He
>>>used engines as analysis tools, not all knowing oracles.
>>
>>Yes, he guided the engines, helping them search forward towards interesting
>>lines. I'm aware he didn't just let them run, but the fact is that the engines
>>did the calculating and evaluating, subject to his supervision.
>
>OK, so we both seem to agree Blass did what a lot of other top corr players
>these days do, use chess engines creatively. And by doing this he was able to
>beat some strong players. I don't see the relevance. It does not surprise me
>that Blass, who is knowlegable about both chess and chess programs, would do
>well. That said, Blass only has 8 ICCF rated games at the ICCF ratings web-site,
>so he has a relatively small number of rated games from which to make a
>judgement about the extent of engine use in correspondence chess.
I never said he is the one who made comments on the amount of engine use in
correspondence chess. The opinion was my own, no matter how many or few may
share it. As to proof, clearly that is anothers tory, so it stands as an opinion
alone.
>
><snip>
>
>>>>>>>>Actually, CA has a LOT more to it than that over CB with all due respect.
>>>>>>>
>>>>>>>I agree. And with all due respect, CB also has a lot more over CA.
>>>>>>
>>>>>>Such as?
>>>>>
>>>>>I have already mentioned the two biggest ones: many strong engines supported and
>>>>>the ability (in Fritz) to run tournaments.
>>>>
>>>>You can run tournaments in CA, though the support is certainly not as extensive
>>>>as in Fritz. However you claimed this severely hampered its use as a top-level
>>>>analysis tool, and with all due respect, I think that's bunk.
>>>
>>>And I know it isn't bunk. I didn't win the US corr championship, twice, by using
>>>bunk methods. I didn't win a world championship semi-finals by the highest score
>>>in any of the sections, 9/10, by using bunk methods.
>>
>>I don't know why you keep throwing titles in my face, nor do I see the
>>relevance. I'm not claiming you don't know how to analyse, I'm saying you
>>underestimate CA. The question isn't whether CB or Fritz are top-level tools, as
>>I never questioned this, and still do not. You questioned CA as a top-level
>>tool, and that has been the core of the argument.
>
>You, in effect, call one of my analysis methods "bunk", and then wonder why I
>defend my qualifications? I should think it would be obvious.
Now, who's putting words into whose mouth? :-) I'll re-iterate it, even though
it's only a few lines above:
"You can run tournaments in CA, though the support is certainly not as extensive
as in Fritz. However you claimed this severely hampered its use as a top-level
analysis tool, and with all due respect, I think that's bunk."
It's not your method that I opined is bunk, but that your claim that without
identical support to that of Fritz, it is 'severely hampered as a top-level
analysis tool'.
You said that two items severely hampered its use as a top-level analysis tool:
the number of top engines, which I'll gladly concede with a reservation (see
below), and its lack of Fritz-like engine tournament support. This last item is
bunk. IMO of course.
I'll also note that Chessbase 8 comes with Fritz 5, so if you really wish to
analyze with Fritz 8, Hiarcs 9, Shredder 7/8, and Junior 8, you're going to have
to pay an *extra* US$212!! In fact, to even have ONE modern engine, you must pay
an extra US$53. Chess Assistant 7 comes with Chess Tiger 14/15 already.
>
><snip>
>
>>>I don't say CA is not a top level tool. I say it is missing a top level feature.
>>
>>You said "These limitations of Chess Assistant are all severe for top level
>>analysis."
>
>Yes. Just like I presume you would say CB/Fritz's relatively poor tree support
>is a severe limitation for top level analysis. But perhaps we have a different
>meaning of severe and you would not say this about CB. I think no chess program
>has everything, at least not yet. I believe CA, Fritz and CB all have important
>and missing features (severe limitations) for analysis. I am not trying to
>single out CA.
I'd like to know, other than the engine support mentioned above, which you would
like to see. This is obviously not an attack but a question.
>>Yes. Without meaning to be rude, I'd point out that you also listed a number of
>>features, that supposedly were also absent in CA and which are not.
>
>How does one do it (backsolve)? I have changed evaluations in CA trees and the
>roots don't change, and when I go to the "tree", "tree operations", "minimax"
>menu I get a confusing (to me) dialog box. I just want to take the _existing_
>tree and run a minimax on it so that it has all the evals & assessments in
>that same tree are updated.
It does indeed request you to choose the tree you want to minimax, but as it
only has two items requested, it doesn't seem terribly confusing. Still, your
request is quite logical.
>>Truly, it can do all this in the tree?
>
>Fritz can't do this with trees, but I wasn't talking about in trees. Fritz/CB
>can import games from a database into a single game, with variations, and then
>do the above on the game and all the variations. More or less the same thing,
>although CA's being able to do this in a tree is much nicer.
This really is quite different. I know it may not seem, so, but in my example,
we're talking about constantly working with the entire tree for the entire
database. Why is this different you might ask? To begin with, it means that if I
make a correction on a leaf evaluation that could come from different openings,
ALL of those openings will see revisions in the evaluations. For example,
suppose in that example on the d-pawn isolani, I revise the evaluation. It would
correct the earlier evaluations of all the openings that lead to it. If I had
grouped up a bunch of games in CB only from the Panov and the QGA, I'd still not
find my corrected evaluation in the Nimzo-Indian, and others. This is merely one
example, and I'm sure I can find others, but time is pressing once more.
Albert
>>I have underestimated it too then and
>>will check this out. Though building large trees in CB or Fritz is pretty much
>>impossible in a practical manner.
>
>Yes. If the game/variations is too big then analysing it all, and sorting
>through it all, is cumbersome in Fritz. I like CA trees, even though as I stated
>before I still need to learn more about all the capabilities. (I would be more
>motivated to do so if CA supported more engines :-)
>
>>>Albert, I can see you love CA. I do too.
>>
>>Yes, I do. Peace.
>
>Peace.
>
>Robin
>
>> Albert
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.