It is time to revolutionize chess GUIs

[Albert Silver]

?? I don't agree with this advice at all. You really think the key to chess progress is to not study openings and only study blunders?

yes, tactics not openings....at most opening PRINCIPLES.

You are misinformed. Pattern recognition is only one part of the equation.

What are the other parts of the equation....Dont make me laugh by saying opening study ...

Until you are a pretty good player, it's all about tactics. Most 1600 and below players blunder in just about every game, either failing to spot a blunder or making a blunder. Of course failing to spot a blunder is a blunder too. When I say blunder I mean simple blunders, outright hanging a piece or missing something as simple as a knight fork. Things that a 2 ply search with quies would pick up in most programs.

At low ELO levels, raising your blunder threshold by 1 ply is a major ELO boost for humans (and computers too.)

Yes, visualization (seeing ahead) is a big part of the equation too. The pattern recognition helps filter the moves, but if you can't see more than two moves ahead, no amount of chess knowledge is going to save you. Thankfully, this is something that can be developed too, but it is one of the most taxing.

If you are less than 1800 ELO the easiest way to improve your chess is to get a book on tactics with puzzles and solve the puzzles as quickly as you can in your head. If the book is too hard get a simpler book but I think it's better for the problems to be on the easy side than on the hard side although it should be somewhat challenging. It's not good to have to ponder 20 minutes on each problem because it is too advanced when you could be building patterns in your brain with easier problems. Use a stopwatch - for learning to work best there must be pressure.

I think for most players this will produce the biggest improvement in the shortest amount of time. If you are already a master it probably won't make as much difference, but I think even strong players have to stay sharp.

Assuredly, and I couldn't agree more, however you seem to think my comment on visualization was a comment for beginners. The discussion with Charles was what a master has that a beginner does not. I disagreed that the difference between a beginner and a master is merely one of basic principles combines with superior pattern recognition.

Among other things, a master has far better developed visualization skills, which is neither the domain of pattern recognition nor chess principles. As you said: adding a ply to one's calculations can make a huge difference in skill.

[yanquis1972]

albert, you are technically titled, correct (or could be?). at least CM, maybe FM?

[tomgdrums]

Think of it... top GMs can't win a match against top chess engines, period.
For us mere mortals, amateur players, there is no chance, engines are merciless tactical monsters.

Why not focus development on better chess GUIs ? New ways of imputing moves, better 3D boards, better spoken analysis, better and faster search functions, better tools for opening analysis, better and more detailed statistics of your own games?

For OTB players, it doesn`t really matter if we're losing to a 3200 ou a 3250-rated engine.

Engine development should concentrate in strategical concepts or different styles of play.

It was a pleasure to play against CS Tal, Tiger, Junior and Rebel because of their aproach to the "styles" concept.

I hate to keep beating a dead horse, but the proliferation of Ippo clones is sharply reducing the diversity of chess programs, and I know there are computer chess enthusiasts who are real connoisseurs of playing style differences and enjoy seeing programs with different styles clashing - and also understand what they are seeing. Are those days going away? When the only difference between programs is ELO, it starts to get boring doesn't it?

Once there was the Power Chess program with a nice female voice to talk with us. I`m really tired of the sarcasm of Fritz, and the ridiculous window that opens when I make a blunder. Why not a GUI that talk gently to the player?

It does not make any sense to facelift a GUI and not add new features.

CA has the same program since version 8.
CB improvement? I see no great difference between CB 8 and CB 11. In fact, CB 8 had a more "intuitive" interface. The pull down menus in new CB interface are really ugly to work, just as bad as the new MS Word.

Today`s processors and graphics capabilities are great. Please improve your GUIs.

I agree with you Don! I like the engines with styles and personality!

[Albert Silver]

albert, you are technically titled, correct (or could be?). at least CM, maybe FM?

I never quite qualified for FM (50 Elo short), but could be CM if I paid the FIDE charges they ask.

[Ponti]

I guess WinBoard doesn't have a 3D board because I don't like 3D boards much.

- Perhaps it`s time to think different. There are users that like it. My coach is a GM and he likes it.

Of course I cannot really imagine exactly how the speech that you describe would actually sound (I suppose the dots indicate long pauses), but what I imagine does not really sound at all like the speech produced by WinBoard for JAWS. So I guess it is mainly a matter of what speech interface you use.

- I think it doesn`t matter the interface, am I wrong? Modern sound interfaces inside PCs and Macs are powerfull, often 5 or 7 channel devices...

Why Chessmaster sells a lot? Marketing strategy and a friendly interface.
They are selling products for entertainment. No edition of program configuration .ini files (gee, that reminds me of Xboard - green/yellow board by default, tried to change that, I could't). That has changed in new versions of Winboard, thanks.

I think Winboard has progressed _a_lot, but it is still far from CB or CA's interfaces when it comes to usability or beauty.

Chess Academy used to be a great program, but somehow I think they stopped developping it. I have a version of Chess Academy, I installed a virtual machine running Win2000 for my WinXP-64 just to play it. It has so many features for training, and you can play against engines too. It`s old, and installing a new engine is not so easy as within Fritz GUI or Arena, but it works.

I liked Chess Genius interface too. It doesn't make any sense that they don`t want to produce a 64-bit version, so I think CG will be portable only. That is as stupid as not making a SMP version of Tiger, IMHO, but the authors do want they want, those are their creations...

I use lokasoft's interface when I play Rebel, an engine I like a lot. The interface is good, but 3D board is awful, as awful is the sound too; but it is very customizable.

Shredder`s interface is marvelous, I like it a lot. It is simple and it just works. 3D boards/pieces could be better. Sound could be better. If there were more database/ training features it would be my main interface to play and train chess.

There are many interfaces that are engine-oriented, i.e., engine matches are priority. Those I really don`t care about, they are not for me.

[hgm]

- I think it doesn`t matter the interface, am I wrong? Modern sound interfaces inside PCs and Macs are powerfull, often 5 or 7 channel devices...

It depends on how it is used. Sound is not necessarily the same as speech. If the application stores recordings of the various words as WAV files, it counts as any sound, and I agree it should be fast. But if it uses a text-to-speech interface, I could imagine it will be slow, because the interface wil have to calculate the wave form before sending it to the sound driver. Developing a good text-to-speech converter is a science in itself, again orders of magnitude more effort than writing a GUI. So I can imagine that what is available for free is pretty low quality. WinBoard for JAWS uses the text-to-speech interface of the JAWS screen reader,which is pretty good. (As can be expected, as this is actually the main task of a screen reader.)

As to 3D in WinBoard:

WinBoard is open-source software. The fact that I am not very interested in using 3D boards, and have no experience in image rendering myself, does not have to stop others from programming such thing. From what I can see it should be a quite independent task, that hardly has to interact with the rest of the GUI. It is just given a position, and should then display it, while the user can perform all kinds of manipulations on it of which WinBoard could stay completely unaware. Only when the user somehow moves a piece to another square this should be reported to WinBoard. So the information flow is just positions in one direction, and moves in the other, and that's it.

If someone would write something like that and wants to make it available, I could easily incorporate it in WinBoard.

[fluviumque]

I am not completely sure what you mean here, but it seems to me that WinBoard does it in the way you want. Wth Analyze File the engine PVs are added to the game moves as PGN variations, and can be accessed by normal variation-tree walking. Currently stillonly for thewhole file, but I am planning to change the menu item for it to "Analyze Game", where it would act on the already loaded game, starting at the currently displayed move.

Ah, that's great, thanks. I downloaded Winboard now because also I wanted to try Crazyhouse offline against an engine. To my knowledge this should be possible with free Sjeng. Unfortunately I failed in installing the engine properly. I think I have to add a line to "Engine - Install 1st" text field; but didn't get it to work. Sjeng does not appear in the WinBoard startup dialog. I'll try it next weekend.

[fluviumque]

Unfortunately I failed in installing the engine properly. I think I have to add a line to "Engine - Install 1st" text field; but didn't get it to work. Sjeng does not appear in the WinBoard startup dialog. I'll try it next weekend.

Got it. Altered winboard.ini directly. Looking forward playing crazyhouse

[hgm]

The "Install 1st" should in priciple allow you to edit the list of engines nside WinBoard itself. But I noticed that when you add a line at the bottom of the edit dalog, but forget to terminate it with a linefeed, the line disappears again. This might be what happened. I still have to look into why it behaves in this strange way.

[fluviumque]

The "Install 1st" should in priciple allow you to edit the list of engines nside WinBoard itself. But I noticed that when you add a line at the bottom of the edit dalog, but forget to terminate it with a linefeed, the line disappears again. This might be what happened. I still have to look into why it behaves in this strange way.

Well, Winboard is kind of nice, but I found myself spending a lot of time solving problems. Currently, with Houdini and the fUci switch (EDIT: now also with ""polyglot.exe ..\Houdini\houdini.ini" /fd=".""), it prefers to have Houdini allocate all 8 GB RAM of my system, making it unusable and really hard to kill Winboard.
Also, I haven't found out to make it use a westernized font for Shogi.
I found out how to use "Chess Merida" font by changing a font command line in the winboard.ini. For Shogi, I added /renderPiecesWithFont *blah to the respective line in the startup screen. The result is, it uses a default font without the additional Shogi symbols (e.g. the lance is invisible), but also in winboard.ini the chessmerida line has disappeared...
I think I'll set up Winboard shortcuts for variations only.