Scrabble freewares (e.g. Quackle) can dump their evaluations into log
files. Would that be an interesting facility for the kind of work you do ?
On Wed, Jun 25, 2008 at 3:47 PM, Roy Brunjes <[EMAIL PROTECTED]> wrote:
> Here is an example of how I use this feature:
>
> 1. Set up a position.
>
> 2. Start infinite analysis and let it run for a few minutes.
>
> 3. Study the history to see what move the engine thought was best at ply
> 10, 11, 12, 13, etc. See how it refines its "thinking" as the number of ply
> increases. This helps the user learn more about the engine in use and where
> its strengths may lie (it's ability to spot a particular move in 30 secs vs
> 190 secs is sometimes useful info. With 3.6.24 today, if I leave the machine
> and return after 4 minutes, I don't know if the current "best move"
> displayed was the same move it thought was best at earlier ply or not. The
> 3.6.23 way allows you to see how quickly engine X can spot move Y without
> having to sit there and watch the output -- just scroll back on the history
> to see the earliest that engine found the key move.
>
> To me, this is very useful info when trying to learn about different
> engines, and it is not the same as setting the multivariation setting to
> display the top 4 moves (which also slows the time to reach a specific
> depth). Chessbase offers this feature for sure in every engine it has
> shipped since Fritz 5.16 (if not sooner -- that's as far back as I go with
> them). Lokasoft offers it -- in fact, every single engine/GUI combination I
> have EVER used has this feature. It is their default way of displaying
> engine thinking output.
>
> I hope that helps clarify why I think it is a useful feature.
>
>
>
>
> ----- Original Message ----
> From: "[EMAIL PROTECTED]" <
> [EMAIL PROTECTED]>
> To: [email protected]
> Sent: Wednesday, June 25, 2008 12:58:21 PM
> Subject: Scid-users Digest, Vol 23, Issue 32
>
> Send Scid-users mailing list submissions to
> [email protected]
>
> To subscribe or unsubscribe via the World Wide Web, visit
> https://lists.sourceforge.net/lists/listinfo/scid-users
> or, via email, send a message with subject or body 'help' to
> [EMAIL PROTECTED]
>
> You can reach the person managing the list at
> [EMAIL PROTECTED]
>
> When replying, please edit your Subject line so it is more specific
> than "Re: Contents of Scid-users digest..."
> Today's Topics:
>
> 1. Re: Scid-users Digest, Vol 23, Issue 29 (Alexander Wagner)
> 2. R: Re: Scid-users Digest, Vol 23, Issue 29 (Giorgio Bellegotti)
> 3. Database for Scid (Pascal Georges)
> 4. Re: Database for Scid: CentriScid (Alexander Wagner)
>
>
> -----Inline Message Follows-----
>
> Hi!
>
> If I remember correctly we had this discussion already once the new engine
> dialog was introduced in cvs. If Roy did not use cvs (and I guess he did
> not) then indeed the behaviour changed in 3.6.24 for the first time. It just
> took a bit from 3.6.23 to 3.6.24. It was thought of as an improvement,
> therefore, Roy, this could happen on upgrading. ;)
>
> To the behaviour: my guess is that Roy has some usage for this history
> output, otherwise he'd not have the wish to get it back. Obviously, he even
> considers to exchange all the new features of 3.6.24 just to get it back and
> there was something done between .23 and .24... (Acutally, I'd also guess
> that there must be some usage for this output as otherwise no engine would
> waste the cpu-cycles to display it...) Its not to argue about whether this
> is good wish or not, its just a statement. Probably, Roy could give an
> example where he uses this output. Just for us to better understand.
> Personally, I think it could be valuable in case there are two lines that
> get pretty much the same evaluation, ie. the two or three best lines
> alternate depending on the search depth eg. but that's just a guess.
>
> A way to solve this problem IMHO could come with Giorgios enhancements to
> the analysis dialogue, and maybe there could be a way to show the evaluation
> history upon request. Acutally, there's plenty of space if only one line of
> analysis is displayed. one could imagine to show the mainline as it is done
> right now, add a free line (ie. some space) to visually separate it clearly
> and scroll the history below that. IMHO would give the advantages of both
> worlds. A click on the history could make it invisible like the current
> analysis is hidden by clicking within the window.
>
> Concerning whether there is another application that offers this feature,
> hm well, I see no reason to limit oneself to the abilities of other apps ;)
>
> Just some thoughts, though.
>
> ----- original Nachricht --------
>
> *Betreff:* Re: [Scid-users] Scid-users Digest, Vol 23, Issue 29
> *Gesendet:* Mi, 25. Jun 2008
> *Von:* Pascal Georges
>
> Sorry, I made a confusion between various versions of Scid, and I thought
> the engine output history was removed since a long time for UCI engine. And
> indeed it is, as I find this kind of output (history) is confusing and of
> little use (I am not aware of any chess app that displays full history).
>
> Pascal
>
> 2008/6/25 Roy Brunjes <[EMAIL PROTECTED]>:
>
>> Pascal,
>>
>> I disagree that this is a mis-configuration. Showing multiple PVs is one
>> thing, what I am asking for is entirely different. Please look closely at
>> the example again and I think you should see what I am asking for is NOT
>> just setting multiPV mode to 4 or something similar. MultiPV set to 4 slows
>> down the search as well which a single PV analysis with history enabled (to
>> show the results of each previous ply) does not.
>>
>> Also, how can it be a misconiguration if it worked differently in 3.6.23
>> and no changes were made to the configuration for 3.6.24? If there is no
>> configuration change, the behavior of the software should be the same -- at
>> least that is what most people expect with software upgrades in my
>> experience.
>>
>> Respectfully,
>>
>> Roy
>>
>
>
>
> --- original Nachricht Ende ----
> Hi!
>
> If I remember correctly we had this discussion already once the new engine
> dialog was introduced in cvs. If Roy did not use cvs (and I guess he did
> not) then indeed the behaviour changed in 3.6.24 for the first time. It just
> took a bit from 3.6.23 to 3.6.24. It was thought of as an improvement,
> therefore, Roy, this could happen on upgrading. ;)
>
> To the behaviour: my guess is that Roy has some usage for this history
> output, otherwise he'd not have the wish to get it back. Obviously, he even
> considers to exchange all the new features of 3.6.24 just to get it back and
> there was something done between .23 and .24... (Acutally, I'd also guess
> that there must be some usage for this output as otherwise no engine would
> waste the cpu-cycles to display it...) Its not to argue about whether this
> is good wish or not, its just a statement. Probably, Roy could give an
> example where he uses this output. Just for us to better understand.
> Personally, I think it could be valuable in case there are two lines that
> get pretty much the same evaluation, ie. the two or three best lines
> alternate depending on the search depth eg. but that's just a guess.
>
> A way to solve this problem IMHO could come with Giorgios enhancements to
> the analysis dialogue, and maybe there could be a way to show the evaluation
> history upon request. Acutally, there's plenty of space if only one line of
> analysis is displayed. one could imagine to show the mainline as it is done
> right now, add a free line (ie. some space) to visually separate it clearly
> and scroll the history below that. IMHO would give the advantages of both
> worlds. A click on the history could make it invisible like the current
> analysis is hidden by clicking within the window.
>
> Concerning whether there is another application that offers this feature,
> hm well, I see no reason to limit oneself to the abilities of other apps ;)
>
> Just some thoughts, though.
>
> ----- original Nachricht --------
>
> *Betreff:* Re: [Scid-users] Scid-users Digest, Vol 23, Issue 29
> *Gesendet:* Mi, 25. Jun 2008
> *Von:* Pascal Georges
>
> Sorry, I made a confusion between various versions of Scid, and I thought
> the engine output history was removed since a long time for UCI engine. And
> indeed it is, as I find this kind of output (history) is confusing and of
> little use (I am not aware of any chess app that displays full history).
>
> Pascal
>
> 2008/6/25 Roy Brunjes <[EMAIL PROTECTED]>:
>
>> Pascal,
>>
>> I disagree that this is a mis-configuration. Showing multiple PVs is one
>> thing, what I am asking for is entirely different. Please look closely at
>> the example again and I think you should see what I am asking for is NOT
>> just setting multiPV mode to 4 or something similar. MultiPV set to 4 slows
>> down the search as well which a single PV analysis with history enabled (to
>> show the results of each previous ply) does not.
>>
>> Also, how can it be a misconiguration if it worked differently in 3.6.23
>> and no changes were made to the configuration for 3.6.24? If there is no
>> configuration change, the behavior of the software should be the same -- at
>> least that is what most people expect with software upgrades in my
>> experience.
>>
>> Respectfully,
>>
>> Roy
>>
>
>
>
> --- original Nachricht Ende ----
>
>
> -----Inline Attachment Follows-----
>
> Hi,
>
> To get a base for Scid that can be used for training (for example) and to
> keep track of extra info, I think there is a workaround by mixing indexed
> flags and PGN tags.
> Each game can get one or several flags that are :
>
> IDX_FLAG_START // Game has own start position.
> IDX_FLAG_PROMO // Game contains promotion(s).
> IDX_FLAG_UPROMO // Game contains promotion(s).
> IDX_FLAG_DELETE // Game marked for deletion.
> IDX_FLAG_WHITE_OP // White openings flag.
> IDX_FLAG_BLACK_OP // Black openings flag.
> IDX_FLAG_MIDDLEGAME // Middlegames flag.
> IDX_FLAG_ENDGAME // Endgames flag.
> IDX_FLAG_NOVELTY // Novelty flag.
> IDX_FLAG_PAWN // Pawn structure flag.
> IDX_FLAG_TACTICS // Tactics flag.
> IDX_FLAG_KSIDE // Kingside play flag.
> IDX_FLAG_QSIDE // Queenside play flag.
> IDX_FLAG_BRILLIANCY // Brilliancy or good play.
> IDX_FLAG_BLUNDER // Blunder or bad play.
> IDX_FLAG_USER // User-defined flag.
>
> So imagine you have a big (or small ... it also works) DB and want to keep
> track of tactics. So each relevant game gets the flag
> IDX_FLAG_TACTICS
> and for example the PGN tag is appended :
> FLAG_TACTICS_data "Removal of the guard/23/black/easy"
>
> that is in order : type/move/side/difficulty
>
> Searches in base are fast because the prefilter of indexed flags.
>
> The data added to "tactics" should be defined once for all and part of
> Scid's UI. That way when saving a game, we can add to the dialog box the
> options needed to set a flag and its data.
>
> The most interesting flags are (I think)
> Middlegame
> Endgame
> Tactics
>
> So for each category could someone list the necessary fields like, for the
> tactics example :
>
> Category Tactics :
> type : pin, overburden, ... , undefined
> move : the move number
> side : white or black
> difficulty : very easy, easy, medium, difficult, very difficult,
> undefined
> solved : solved or unsolved
> comment : free text
>
> Pascal
>
> _______________________________________________
> Scid-users mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/scid-users
>
>
> -------------------------------------------------------------------------
> Check out the new SourceForge.net Marketplace.
> It's the best place to buy or sell services for
> just about anything Open Source.
> http://sourceforge.net/services/buy/index.php
> _______________________________________________
> Scid-users mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/scid-users
>
>
-------------------------------------------------------------------------
Check out the new SourceForge.net Marketplace.
It's the best place to buy or sell services for
just about anything Open Source.
http://sourceforge.net/services/buy/index.php
_______________________________________________
Scid-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/scid-users
