![[nz]](/mwf/v12/flags/nz.png "New Zealand")
Hello, the author of the Tarrasch Chess GUI here.
I would just like to chip in here and point out that I do take on board feedback from users and implement their suggestions. I've put an email exchange from 16th February below where user "Jorge" requested a simple feature and I implemented it for him within 24 hours. I could provide other evidence of similar events in the past. I will also note that the Tarrasch Chess GUI is also open source, so any sufficiently skilled and motivated user can implement anything they want.
Obviously I don't implement everything anybody asks for. Sometimes I don't agree that their suggestion would improve the program. Sometimes I would love to implement the suggestion but immediate implementation doesn't fit schedule or my current priorities, so I have to put the suggestion on a wish-list for later. For example Engine v Engine has been on a wish list ever since I started. If I choose to continue to develop the program, eventually I will get to it. Developing anything involves compromises and tradeoffs I am sure you agree.
I can demonstrate a history of Tarrasch improving steadily. I am the first to admit it is not suitable for everyone, but I have many happy users who tell me they like the program a lot.
*** email excerpt from 16th February
Hello Jorge,
Thank you for your kind evaluation of my program.
As it happens I was in the process of creating a maintenance release when I received your email yesterday. Because others have asked for the same thing, and because you asked so nicely, and not least because it is easy to do, I decided to add your requested option to this release.
I uploaded the new release to triplehappy.com 5 minutes ago. If you go there now and download Tarrasch again, you will get the new version, including the "No Auto Flip" option on the Options->General menu.
Kind regards,
Bill.
Jorge Hernandez wrote:
I would just like to chip in here and point out that I do take on board feedback from users and implement their suggestions. I've put an email exchange from 16th February below where user "Jorge" requested a simple feature and I implemented it for him within 24 hours. I could provide other evidence of similar events in the past. I will also note that the Tarrasch Chess GUI is also open source, so any sufficiently skilled and motivated user can implement anything they want.
Obviously I don't implement everything anybody asks for. Sometimes I don't agree that their suggestion would improve the program. Sometimes I would love to implement the suggestion but immediate implementation doesn't fit schedule or my current priorities, so I have to put the suggestion on a wish-list for later. For example Engine v Engine has been on a wish list ever since I started. If I choose to continue to develop the program, eventually I will get to it. Developing anything involves compromises and tradeoffs I am sure you agree.
I can demonstrate a history of Tarrasch improving steadily. I am the first to admit it is not suitable for everyone, but I have many happy users who tell me they like the program a lot.
*** email excerpt from 16th February
Hello Jorge,
Thank you for your kind evaluation of my program.
As it happens I was in the process of creating a maintenance release when I received your email yesterday. Because others have asked for the same thing, and because you asked so nicely, and not least because it is easy to do, I decided to add your requested option to this release.
I uploaded the new release to triplehappy.com 5 minutes ago. If you go there now and download Tarrasch again, you will get the new version, including the "No Auto Flip" option on the Options->General menu.
Kind regards,
Bill.
Jorge Hernandez wrote:
> Bill :
>
> This is an excellent program.
> I have some Mate in 2 problems related to empassant pawn case
> and this was the only program that could solve them.
> ( Stockfish engine ).
>
>rkB5/ppp4R/3B1P2/3PpK2/8/5Q2/8/8 w - e6 0 1
>
>5Nk1/5pPp/4PB1p/1KpP2p1/6R1/1B5p/P7/8 w - C6 0 1
>
> However I would like an option
> to switch off the automatic flipboard when playing
> Blacks ( with white always down ).
>
> Thanks for all,
>
> Jorge
>
>
![[mx]](/mwf/v12/flags/mx.png "Mexico")
Thanks for your efforts Bill, I remember liking the GUI and hope you choose to keep developing the program.
As I've said somewhere else, there's still no GUI that is capable of doing it all, or does it on a user-friendly manner as of yet, so unlike chess engines that have already an amazing strength and adding 20 or 100 elo wouldn't be very noticeable (unless one plays engines against one another) in the terrain of Chess GUIs there's still a lot of room to cover.
As I've said somewhere else, there's still no GUI that is capable of doing it all, or does it on a user-friendly manner as of yet, so unlike chess engines that have already an amazing strength and adding 20 or 100 elo wouldn't be very noticeable (unless one plays engines against one another) in the terrain of Chess GUIs there's still a lot of room to cover.
Thanks Uly. If you haven't already, please try V2 of Tarrasch (it's been out for a few months now). It is a vast step up from the original V1 version.
Okay, I did, here's my feedback:
1 - It seems it's impossible to install engines, instead the user has to specify the engine's path every time he wants to load.
2 - No easy way to set up engine parameters, as one would need to type them out every time; one can only set 4 parameters (Rybka 3 Persistent Hash requires at least 6.)
3 - I tried loading Critter 1.4 64bit, after giving it 64MB RAM, 4 Max CPU cores,and clicking the "Kibitzer Start/stop button" (this starts Infinite Analysis?), the engine never showed anything.
4 - After trying that with Rybka 2.3.2a, it did run, but seemed to be stuck at MultiPV=3 without a way to switch her to Single PV.
5 - The Kibitz panel was restricted to 3 lines of output, compare what Shredder Classic shows in the same screen space:
Tarrasch
Shredder
It seems a lot of space is wasted by the Notation Window, I'd suggest switching them around, so that the notation appears below the board and the engine kitbitzing appears at its right (Heh, it seems it's my main problem with the GUI as I also suggested it for previous version); it's more important for the Analysis user to see more analysis lines than being able to click on a full game, as it's shown in the screenshot, with a 29 moves game a lot of space is wasted (Or you could sort it like this for a compromise between notation space and engine output space.)
It would be great to take a look at your prioritized list of future enhancements.
1 - It seems it's impossible to install engines, instead the user has to specify the engine's path every time he wants to load.
2 - No easy way to set up engine parameters, as one would need to type them out every time; one can only set 4 parameters (Rybka 3 Persistent Hash requires at least 6.)
3 - I tried loading Critter 1.4 64bit, after giving it 64MB RAM, 4 Max CPU cores,and clicking the "Kibitzer Start/stop button" (this starts Infinite Analysis?), the engine never showed anything.
4 - After trying that with Rybka 2.3.2a, it did run, but seemed to be stuck at MultiPV=3 without a way to switch her to Single PV.
5 - The Kibitz panel was restricted to 3 lines of output, compare what Shredder Classic shows in the same screen space:
Tarrasch
Shredder
It seems a lot of space is wasted by the Notation Window, I'd suggest switching them around, so that the notation appears below the board and the engine kitbitzing appears at its right (Heh, it seems it's my main problem with the GUI as I also suggested it for previous version); it's more important for the Analysis user to see more analysis lines than being able to click on a full game, as it's shown in the screenshot, with a 29 moves game a lot of space is wasted (Or you could sort it like this for a compromise between notation space and engine output space.)
It would be great to take a look at your prioritized list of future enhancements.
> Okay, I did, here's my feedback:
>
> 1 - It seems it's impossible to install engines, instead the user has to specify the engine's path every time he wants to load.
>
> 2 - No easy way to set up engine parameters, as one would need to type them out every time; one can only set 4 parameters (Rybka 3 Persistent Hash requires at least 6.)
>
> 3 - I tried loading Critter 1.4 64bit, after giving it 64MB RAM, 4 Max CPU cores,and clicking the "Kibitzer Start/stop button" (this starts Infinite Analysis?), the engine never showed anything.
>
> 4 - After trying that with Rybka 2.3.2a, it did run, but seemed to be stuck at MultiPV=3 without a way to switch her to Single PV.
>
> 5 - The Kibitz panel was restricted to 3 lines of output, compare what Shredder Classic shows in the same screen space:
>
> It seems a lot of space is wasted by the Notation Window, I'd suggest switching them around, so that the notation appears below the board and the engine kitbitzing appears at its right (Heh, it seems it's my main problem with the GUI as I also suggested it for previous version); it's more important for the Analysis user to see more analysis lines than being able to click on a full game, as it's shown in the screenshot, with a 29 moves game a lot of space is wasted (Or you could sort it like this for a compromise between notation space and engine output space.)
>
> It would be great to take a look at your prioritized list of future enhancements.
Hey Uly,
I agree that Tarrasch needs to deal with engine parameters better than it does. However one of the reasons I like Tarrasch GUI is that the notation window is so large and SO easy to work with. It is the best GUI for entering verbal annotations. In fact no other GUI is close when it comes to entering verbal annotations. Which is important to me when I am annotating games and positions. I haven't had a problem with loading engines or getting any of them to work.
Having said that, Shredder is still the easiest for getting engines into the analysis fray!!
I also like Tarrasch for quickie analaysis on the fly. When I am studying a position (always without an engine first) but want to create verbal annotations and/or actual variations on my own, Tarrasch is hands down the easiest GUI for that kind of study.....When I just need to verify a move choice with an engine.
Like you I haven't seen a GUI that is a one stop shop for all needs. (although I am getting very curious about the upcoming Hiarcs GUI)
> It is the best GUI for entering verbal annotations. In fact no other GUI is close when it comes to entering verbal annotations. Which is important to me when I am annotating games and positions. I haven't had a problem with loading engines or getting any of them to work.
I think annotations should have their own window, seeing a game with many comments is just otherwise a mess.
Try pasting this PGN into Tarrasch and it won't let you navigate it (in Shredder Classic it's fine):
[Event "24 hours per move"]
[Site "Rybka forum"]
[Date "2009.11.04"]
[Round "?"]
[White "Tzafrir, Shahar"]
[Black "Rybka Warriors, Forum"]
[Result "1-0"]
[ECO "B20"]
[Annotator "Tzafrir,Shahar"]
[PlyCount "135"]
[EventDate "2009.11.04"]
[EventType "corr"]
{A 24 hour/move game: HEM vs. the Rybka forum team; part of two game
simultaneous white/black match that started on November 4th 2009 and concluded
on December 31st 2009. The other game (Rybka-Warriors had White) ended in a
draw. State of the art chess-engines / computers were used by both sides
throughout the game. The Rybka team played in consultation using a team-vote
approach to decide their moves. According to my post mortem analysis, Black
is losing by force after its 23rd move.} 1. e4 c5 2. b4 cxb4 3. Nf3 Nf6 $6 {3.
..d5 and 3. ..d6 are more common. Judging by this game, it's clear why.
White's aim in the Wing gambit is to gain time, and Nf6 plays to White's hands.
} 4. e5 Nd5 5. a3 e6 (5... bxa3 6. Bxa3 $32) 6. axb4 Bxb4 7. Na3 {With a
c4->Nc2->Nxb4 plan.} Nc6 ({Only one previous rated game on record with 7.Na3:
[Event "Moscow RGSU-B"] [Date 2006.11. 16"] [Round "11"] [White "Kislov, Oleg"]
[Black "Mozharov, Mikhail"] [Result "1-0"] [WhiteElo "2267"] [BlackElo "2418"]
} 7... O-O 8. c4 Nf4 9. Nc2 Ba5 10. Ba3 Re8 11. h4 f6 12. exf6 Qxf6 13. g3 Ng6
14. Bd6 Nc6 15. Bd3 Bc7 $4 16. Bxc7 {1:0}) 8. c4 Nde7 9. Nc2 $146 O-O $6 {An
unwarranted concession of Black's bishop} (9... Ba5 $142 $13) 10. Nxb4 {I
spent 4 tempi to exchange the bishop realizing I have sufficient development
advantage (from the early b4 sac and e5 push) to not incur any penalty doing
so. Black has anyway spent and will spend equal time on his knights jumping
around. More important, White's subsequent plan and moves all follow naturally
after this exchange.} Nxb4 11. Ba3 Nec6 {This is Black's *sixth* Knight jump
in the first eleven moves of the game - 54.5% of Black's opening moves
dedicated to a knight tour.} 12. Qb1 a5 13. Bd3 {Provoking f5 or even better
(for White) g6} f5 (13... h6 $142 $11 {I would have forced a draw had Black
played h6.} 14. Bh7+ Kh8 15. Be4 Rb8 16. O-O d5 17. exd6 Qxd6 18. h3 Rd8 19.
Rd1 b5 20. Bxc6 Qxc6 21. Bxb4 axb4 22. Qxb4 Qxc4 23. Qxc4 bxc4 24. Ne5 Rd5 25.
Nxc4 $11 {While 13. ..h6 was safer, Black can't be blamed, being a pawn up,
for not sensing the danger the position holds and avoiding the drawing line.})
(13... g6 $143 14. h4 $40) 14. Be2 {Lost a tempo but achieved provoking the
weakening f5 that allows the subsequent h3->g4 break ideas for White. An extra
benefit is that f5 also slows down the development of Black's c8 bishop and
cramps Black's position.} d5 15. exd6 Qxd6 16. d4 Bd7 {Start of a slow Black
plan to place the bishop at h5. I'm not sure this was the best plan for Black
as it allows White to peacefully finish development. Then again, it's already
hard to come with a viable other plan.} ({Possibly a better/faster plan was
b6->Bb7->Rad8->Rfe8, or b6->Ba6} 16... b6 $142 17. O-O Bb7 $13 (17... Ba6 $144
$13)) 17. Qb2 Be8 18. O-O Bh5 19. Rfd1 Rae8 $6 {Likely the culprit of Black's
subsequent trouble.} (19... Rac8 $142 20. Rd2 $13 (20. h3 $144 $13) (20. Ng5
$144 Bxe2 21. Qxe2 $13)) 20. Rd2 b6 {Black has a hard time finding a plan as
evident by this last 'waiting' move.} 21. Rad1 $14 Re7 {By now White has more
than enough compensation for the saced pawn - White position is harmonious and
Black needs to struggle to find an equalizing plan.} 22. h3 $1 {Preparing g4!
I played this now before Qb3 and Bb2 to 'hide' the g4 plan and try and make
this look like an inconspicuous 'luft' or waiting move by White. I expected
another waiting move from Black in return per the recent two non-moves Black
just played - most likely h6 that was indeed played. Objectively it's quite
hard to find a plan or a move here for Black that doesn't hurt its position
further.} h6 23. Qb3 Rc8 {According to my post-mortem analysis, White is
winning by force after this move.} 24. Bb2 Nd8 $2 {A 'computer like' move that
is not only 'not pretty' but worse it loses control of the key e5 square and
probably underestimated (or missed?) the strength of White's subsequent two
shots. However it's not easy to come up with a move that offers harder
resistance, as Black is already in a predicament. My analysis post mortem
shows Black is losing by force after White's 24th Bb2 move.} (24... Bxf3 $144 {
An attempt that doesn't seem to work.} 25. Bxf3 Kh7 26. Bc3 $14 (26. Be2 $13))
(24... Rce8 $144 {An attempt that doesn't seem to work.} 25. Nh4 Bxe2 $8 26.
Rxe2 Qd8 (26... g5 $143 27. Ng6 Rg7 28. Ne5 Nxe5 29. Rxe5 Rge7 30. Bc3 Qc7 31.
Bxb4 axb4 32. Qxb4 Rd8 33. Re3 $16) 27. Qf3 (27. Qg3 Rd7 28. Rde1 Qg5) (27. d5
Rd7 28. Nf3 exd5 29. Red2 Kh7 30. cxd5 Ne7 31. d6 Nec6 $13) 27... Rf7 28. Qh5
Qg5 29. Qxg5 hxg5 30. Nf3 Nd8 31. d5 g4 32. hxg4 fxg4 33. Nd4 $16) (24... f4
$142 $5 {Appears to have been Black's best attempt; however it's refuted as
well.} 25. Nh4 $142 (25. Bc3 $143 Ree8 (25... Rb8 26. Qb1 Rd8 27. Ng5 hxg5 28.
Bxh5 $14) (25... Rd8 26. Qb1 b5 27. d5 exd5 28. cxd5 Qc5 29. dxc6 Rxd2 30. Rxd2
Qxc3 31. Qf5 $14) 26. d5 exd5 27. cxd5 Nd8 28. Qb2 Nf7 29. Bxg7 Qg6 30. Nh4
Qxg7 31. Qxg7+ Kxg7 32. Bxh5 Re5 33. Bxf7 Kxf7 34. d6 $13) 25... a4 $142 (25...
Bxe2 $143 26. Rxe2 Rd7 27. d5 Rcd8 28. Bxg7 Kxg7 29. Qf3 Qe7 30. Nf5+ exf5 31.
Rxe7+ Nxe7 32. Re1 $16) 26. Qxa4 Ra7 (26... Bxe2 $143 27. Rxe2 Ra7 28. Qb3 $14)
27. Qb3 Na5 28. Qa3 Bxe2 (28... Nac6 $143 29. Qc3 Bxe2 30. Rxe2 Na5 31. Ba3 $18
) 29. Rxe2 Rcc7 (29... Rxc4 $143 30. d5 $16) 30. Rde1 Nxc4 31. Qb3 Na5 32. Qf3
Nc2 33. Rd1 $16) 25. g4 $1 {White's pieces are all optimally placed. It's time
to convert White's positional advantage to material with the g4 and d5 break.
Any delay in the break could have allowed Black to equalize. It's like knowing
to not fall in love in a stock you own and when to sell - it was hard for me
to 'ruin' White's beautiful position with 25. g4, but the time has come to
break and convert.} fxg4 26. Ne5 Be8 (26... Ndc6 $143 27. hxg4 Be8 28. g5 hxg5
29. Qg3 Nxe5 30. dxe5 Qc7 31. Rd6 Na6 (31... Ba4 32. R1d4 $18) 32. Qxg5 $16)
27. d5 $3 {Thematic break. Chess engines, even after a long analysis time,
don't realize this is the strongest forcing move here.} (27. Bxg4 h5 $14) 27...
exd5 28. Bxg4 Ndc6 {Black is forced to give up an exchange looking for an
endgame he hopes is drawish. Black's alternatives per the following analysis
were worse.} (28... Rcc7 $143 $4 29. Rxd5 Nxd5 30. Rxd5 Qf6 31. Rxd8 $18) (
28... Rc5 $143 29. cxd5 h5 (29... Bb5 30. f4 Rcc7 (30... Kf8 31. Bd4 Rxd5 32.
Bf3 Ndc6 33. Bxd5 $18) 31. Bf5 Be8 32. Rg2 $18) 30. Bf5 h4 31. Nc6 Ndxc6 32.
dxc6+ Bf7 33. Qxb4 axb4 34. Rxd6 $18) (28... Ne6 $142 {28. ..Ne6 offered an
interesting defence although White still reaches a very promising position
through 33. Qe3!} 29. cxd5 Nc5 30. Qc3 Ra8 (30... Rb8 $143 31. Nc4 Qf4 32. Rd4
Ne4 33. Qe3 Qxe3 34. fxe3 Ba4 35. Rc1 Nf6 36. d6 $18) 31. Nc4 Qg6 $8 32. d6 Ne4
33. Qe3 $1 (33. Qc1 $143 Re6 (33... Nxd2 34. dxe7 $18) 34. Kh2 (34. f4 Nxd2 35.
Rxd2 Re4 36. d7 Rd8 37. Nd6 Rxd7 38. Nxe4 Rxd2 39. f5 Rc2 40. fxg6 Rxc1+ 41.
Bxc1 Bxg6 $13) 34... Na2 35. Qa1 Nxd2 36. Rxd2 Bf7 37. Be5 Nb4 38. Nxb6 $14)
33... Nxd2 34. dxe7 Nxc4 35. Qe6+ Kh7 (35... Qxe6 $143 36. Bxe6+ Kh7 37. Rd8
Nxb2 38. Rxa8 $18) 36. Rd8 $16) 29. Bxc8 Nxe5 30. Bxe5 Rxe5 (30... Qxe5 $143
31. cxd5 Bf7 32. Bg4 $14) 31. cxd5 {Black now faces a major decision - 31. ..
Bf7 vs. 31. ..Rg5+. According to the game and the analysis herein, both lose
by force, although 31. ..Rg5+ offered tougher resistance via some very
difficult lines.} Bf7 $4 {Black expects 31. ..Bf7 to lead to a drawish endgame.
However this is refuted in the game and further in the following analysis.
The endgame reached after 31. ..Bf7 while very much outside the horizon of
current chess engines, is losing by force with a simple plan. The game was
played at a 24 hours per move time-control. However the Black team took a
'time out' and spent 4 days on this difficult move. I dare guess the Black
team ended up playing the more dubious 31. ..Bf7?? due to having over-analyzed
31. ..Rg5+ and not being able to find a forcing drawing line there, opted for
the seemingly more drawish endgame after 31. ..Bf7. I further believe that
with only 24 hours for the move, the Black team was more likely to have played
the more natural looking 31. ..Rg5+.} (31... Rg5+ $142 {Leads to some very
complicated lines, with 3 main tries for Black, and a very narrow path White
must walk through, with a series of 'only moves' White must find to maintain
the win / avoid a draw.} 32. Kh1 $8 {Black now has 3 main attempts: 32. ..Qf6,
32. ..Bh5, 32. ..a4. And further 2 easier to refute attempts: 32.. Bg6, 32. ..
Bf7. I believe only 32. ..Qf6 challenges White (and I present its refutation
too), but I'll quickly address all five attempts with their respective
refutations.} {First attempt: 32. ..Qf6} Qf6 ({Second attempt: ..Bh5} 32... Bh5
$143 33. Bg4 Bxg4 (33... Bg6 $143 34. Re2 Kh7 35. Re6 Qf4 36. d6) (33... Na6
$143 34. Bxh5 Rxh5 35. Qf3 Rg5 36. Rc2 $16) (33... a4 $143 34. Qg3 Bxg4 35.
hxg4 Qd7 36. f3 Nxd5 $16) 34. hxg4 Rxg4 35. Qh3 Re4 (35... Qd7 $143 36. d6 Nc6
37. f3 Rg1+ 38. Kh2 Qxh3+ 39. Kxh3 Rxd1 40. Rxd1 Kf7 41. Rc1 Nb8 $16) (35... h5
$143 36. f3 Rg6 37. Qc8+ Kh7 38. Qf5 $18) (35... Rf4 $143 36. Qe6+ Qxe6 37.
dxe6 Re4 38. Rd6 $18) (35... Rg6 $143 36. Qc8+ Kh7 37. Qf5 Kg8 38. Rc1 Rf6 39.
Rc8+ $18) 36. Qc8+ Qf8 37. Qxf8+ Kxf8 38. d6 Re8 39. d7 Rd8 $16) ({Third
attempt: ..a4} 32... a4 $143 33. Qf3 Bh5 34. Bg4 Bxg4 35. hxg4 Na6 36. Rd4 Nc5
37. Kg2 Kh7 38. Qf4 Rg6 39. Qxd6 Rxd6 40. Rb4 $18) ({Fourth attempt: ..Bg6:}
32... Bg6 $143 33. Be6+ Kh7 (33... Kh8 $143 34. Qe3 $18) 34. Qf3 Bh5 (34... Nc6
$143 35. h4 Rh5 36. Kg2 Ne5 37. Qf4 $18) 35. Bf5+ Kh8 36. Bg4 Bxg4 37. hxg4 $18
) ({Fifth attempt: ..Bf7:} 32... Bf7 $143 33. Be6 Qf4 (33... Na6 $143 34. Re2
$18) 34. Qe3 Qxe3 (34... Qf6 $143 35. f4 $16) 35. fxe3 Bxe6 (35... Re5 $143 36.
Bxf7+ Kxf7 37. d6 $18) 36. dxe6 Re5 37. Rd6 $18) 33. Re1 $3 {All other
attempts e.g. 33. d6+, 33. h4 only draw.} Ba4 (33... Bf7 $143 34. Be6 a4 35.
Qd1 a3 36. Bxf7+ Qxf7 37. d6 a2 38. d7 a1=Q 39. d8=Q+ Kh7 40. Qe8 Qc3 41. Qe4+
Qf5 42. f3 Nc6 43. Qdc2 Qxc2 44. Rxc2 Nd4 45. Rc3 Rg6 46. Qxf5 Nxf5 $18 {
Reaches the same type of won end-game the game has reached. White will win the
b pawn, force exchanging rooks etc as in the game.}) 34. Qe3 Kh8 35. Be6 Nc2 $8
36. Rxc2 Bxc2 37. f4 Rg6 $8 38. Qe5 (38. d6 $143 Ba4 39. Qf3 Qd8 40. f5 Rf6 41.
Qd5) 38... Kh7 (38... Qxe5 $143 39. fxe5 $18) 39. f5 Rg5 $8 40. Qxf6 gxf6 41.
d6 Ba4 (41... Rg3 $143 42. d7 Rxh3+ 43. Kg2 Rd3 44. Rc1 Ba4 45. Rc7 $18) 42.
Rc1 Rg7 43. h4 $18) 32. Be6 $18 {White plan is simple: win Blacks a & b pawns
without making any concessions. Activate the rooks, force swapping rooks and
reach a won K+R+PP vs. K+N+PP 8 pieces end-game. Chess engines have a hard
time evaluating this endgame, forming and executing this plan as it's outside
the horizon and 6-pieces tablebase offer surprisingly little help. Instead the
chess engines shuffle pieces without their eval making much progress. This is
a nice example where a long term human plan sees, well, longer.} Bxe6 (32...
Qe7 33. Bxf7+ Qxf7 34. f4 Rf5 35. Qc4 Rxf4 36. Qc8+ Kh7 37. d6 $18) 33. dxe6
Qxe6 34. Rd8+ Kh7 $8 35. Qxe6 Rxe6 36. R8d6 Re4 37. Rxb6 a4 38. Ra1 Nd5 (38...
Nd3 $144 39. Ra3 Rd4 (39... Nf4 40. Ra6) 40. Rb7) 39. Rb7 h5 $2 {Unnecessary
weakening of the pawn structure. Black's plan was to play h4 later and try to
create a fortress. However it does not work.} 40. Kg2 $1 {Essential part of
the winning plan. Not allow the king to get caged.} Rd4 (40... h4 {[%emt 0:00:
00] Post mortem one of Black's comments was they believed 40. ..h4 could have
created a fortress. This segment refutes that.} 41. Rg1 Ne7 (41... Re6 42. Kf3
Ne7 43. Rg4 Rf6+ 44. Kg2 Nf5 (44... Ng6 45. Ra7 a3 (45... Nf4+ 46. Kf1 Ng6 47.
Raxa4 Rc6 48. Kg2 Rf6 49. Rge4 Rb6 50. f4 Rb7 51. f5 Ne7 52. Rxh4+ Kg8 $18) 46.
Rxa3 Nf4+ 47. Kf1 Ng6 48. Ke2 Nf4+ 49. Kd2 Rd6+ (49... Ng6 50. Ke3 Rb6 51. f4
Kg8 52. Rg5 Ne7 53. Kf3 Rb7 54. Kg4 Rc7 55. Kxh4 $18) 50. Ke3 Nd5+ 51. Kf3 Nf6
52. Rxh4+ $18) 45. Rxa4 Kg6 (45... Rg6+ 46. Rg4 Ra6 47. Rgb4) 46. Rba7 Kh5 47.
R4a6 Rxa6 (47... Rf8 48. f4 Rd8 49. Kf3) 48. Rxa6 Nd4 49. Ra5+ Kg6 50. Ra4 Nf5
51. Kf3 Nh6 52. Rxh4) (41... Rc4 42. Kh2 Nc7 43. Rg4 Rc2 44. Rxh4+ $18) 42. Kf3
Re6 43. Rg4 Rf6+ 44. Rf4 Nf5 45. Ra7 a3 46. Rxa3 Kg6 47. Ra5 Kg5 48. Rg4+ Kh6
49. Rga4 Nd4+ 50. Ke3 Nf5+ 51. Ke4 Nd6+ 52. Kd5 Ne8 53. Rxh4+ Kg6 $18) 41. Ra7
Nc3 $6 {Losing another tempo. However there are no better alternatives by now
anyway.} 42. Rc1 Nd5 43. Rc6 $1 {Much better than 43. Rg1 - a plan that some
chess engines opt for but doesn't seem to work. 43. Rc6 activates the rook.}
Nf4+ 44. Kg3 {White's plan is to accept a few Checks from Black, and in
between play f3 and h4 to get control of g5 to place there a rook later -
threatening to win either g or h pawn thus forcing exchanging rooks.} a3 45.
Rcc7 Ne6 $8 46. Re7 Rd6 47. Rxa3 {First part of the plan complete.} Kh6 48.
Rea7 {White now needs to force exchanging a pair of rooks to reach a won
end-game. Still most chess engines are unable to determine a winning plan here.
} Nc5 49. f3 $1 Rg6+ 50. Kf2 Rd6 51. h4 $1 {Second part of the plan complete.
Getting control of the key g5 square to allow placing a rook there in some
possible lines.} Nd3+ 52. Kg2 Rg6+ 53. Kh2 Nc5 54. Rc3 Ne6 55. Ra5 Nf4 56. Rf5
Ne2 {Ends the game at once by giving up Blacks h5 pawn, however all other
options lose fast as well.} (56... Rf6 $144 57. Rcc5 Kg6 (57... Rxf5 58. Rxf5
Ne2 59. Re5 Nd4 60. Kg3 g6 61. Kf4 Nc6 62. Rd5 Nb4 63. Rd6 Na2 64. Ke5 Nc3 65.
Kf6 Ne2 66. Rd8 Kh7 67. Kg5 Kg7 68. Rd7+ Kf8 69. Kf6 $18 {With mate shortly})
58. Rg5+ Kh6 59. Rcf5 Rxf5 (59... Ne2 60. Rxh5+ Kg6 61. Rhg5+ Kh6 62. Rxf6+
gxf6 63. Rc5 $18 {With mate shortly.}) 60. Rxf5 Ne6 61. Re5 $18 {With mate
shortly. This position highlights White's end-game plan after Black's 31. ..
Bf7 - exchanging rooks and reaching a won position, even had Black not
weakened the h6 pawn.}) 57. Rcc5 Ra6 58. Rxh5+ Kg6 59. Rhg5+ Kh6 60. Rc8 g6 (
60... Kh7 61. Rc7 Rg6 62. Rxg6 Kxg6 63. h5+ $18 {Reaching a won 6 pieces table
base positions in a few moves.}) 61. Rgc5 (61. Rg4 {Most chess engines prefer
Rg4, however as played 61. Rgc5 ends the game equally fast.}) 61... Ra4 (61...
Ra7 $144 62. R5c7 Rxc7 $8 63. Rxc7 $18 {Reaches a won 6 pieces table-base
position in a few moves}) 62. Rc4 Ra3 (62... Ra2 $144 63. Rc2 Rxc2 64. Rxc2 $18
{Reaches a won 6 pieces table-base position in a few moves.}) 63. R4c7 g5 $8
64. Rc6+ Kg7 65. R8c7+ Kf8 66. Rf6+ Kg8 67. Re6 Kf8 68. Rxe2 {Followed by mate.
The game ran from November 4th until December 31st 2009.} 1-0
Yeah Tarrasch wouldn't read that PGN.
But for the way I like to study I do like verbal annotations in the same window. I don't find too many comments to be messy. Not when I am writing them for myself and my own study purposes.
Tarrasch isn't perfect but for my study methods it is definitely the next best GUI to Shredder Classic. (Shredder Classic still being the best at this point for my study purposes)
It is funny how many different things people want from a GUI. It would be hard to make one that works for everybody.
But for the way I like to study I do like verbal annotations in the same window. I don't find too many comments to be messy. Not when I am writing them for myself and my own study purposes.
Tarrasch isn't perfect but for my study methods it is definitely the next best GUI to Shredder Classic. (Shredder Classic still being the best at this point for my study purposes)
It is funny how many different things people want from a GUI. It would be hard to make one that works for everybody.
See my reply to Uly below.
Actually if you paste the moves and annotations only (i.e. everything beyond the intro [tags]) Tarrasch reads this perfectly. Sorry I only just noticed thesew updates to the thread. I have no time now, but I'll take a look and get back to you when I get a chance.
Indeed Tarrasch does read it correctly if you paste it just paste the intro tags!!
Just to reinforce the point that Tarrasch handles this material really well. Again, if I haven't made it clear already, don't attempt to put [PGN Tags] in the moves window, they don't belong there. This is obvious to regular Tarrasch users, because the normal workflow of entering moves, saving the game; or loading a game from a .pgn file never puts the pgn tags in the moves window.
![[nl]](/mwf/v12/flags/nl.png "Netherlands")
> don't attempt to put [PGN Tags] in the moves window
It might be a nice improvement to just let Tarrasch ignore any tags, rather than crash or refuse it when it encounters them. Often users might try to paste games exported from other Chess software, which exports them as PGN to the clipboard. It is a bit of a pain to require them to first paste it into NotePad (or a similar editor), then select the 'body of the game, before they can paste it into Tarrasch.
You are absolutely right of course. It had never occurred to me to paste the tags as well as the moves, but it should have occurred to me that someone might try that and a simple improvement (now on my todo list) will be to interpret the tags in this situation.
> 1 - It seems it's impossible to install engines, instead the user has to specify the engine's path every time he wants to load.
You understand now why I don't take it very seriously when somene says it is much easier to install engines in Tarrasch then in WinBoard?
Don't get me wrong: it is not my intention to pick on Tarrasch GUI. I think anyone working on a free Chess GUI should be applauded, as it is between 10 and 100 times more work than writing an engine. I even downloaded Tarrasch GUI myself. (Too bad it doesn't run; it only pops up an error box that it is "not a valid Win32 application.) It is further also a good thing that GUIs are doing things in a different way, because that offers the user a wider variety of choice.
But it is also clear that the develpment of Tarrasch GUI has just started, and there still is a very long way to go before all conceivable or even essential features will have been implemented. And it is pretty obvious that any piece of equipment that can do very much will be more difficult to control than a piece of equipment that can do only a few things. Having only 3 buttons always in view is superb. As soon as the number of buttons that is needed grows to a few hundred, having them all in one-click range would be a disaster.
> You understand now why I don't take it very seriously when somene says it is much easier to install engines in Tarrasch then in WinBoard?
I still find the Tarrasch GUI easier to use for what I want and need.
Which according to you is a good thing:
> It is further also a good thing that GUIs are doing things in a different way, because that offers the user a wider variety of choice.
Yes, indeed, a very good thing. Remember the bicycle vs. spaceship?
![[us]](/mwf/v12/flags/us.png "United States")
Would be great exercise to pedal to the moon and back..
I have replied to this with a new topic.
> Yes, indeed, a very good thing. Remember the bicycle vs. spaceship?
Sure I do!
The bicycle=practical, outlasted many modes of transportation, easily accessible, a solid piece of craftsmanship that also helps people maintain a healthy lifestyle.
Spaceship=Inaccessible, usually discarded quickly, forces users to sit in uncomfortable positions for a long period of time while they wait for the spaceship to get somewhere, and can often cause nausea as well...even in people trained for space flight.
Exactly. Hence I said: don't use a space ship whan a bicycle can do the job.
But for interplanetary travel a bicycle is no good, and you will have no choice as to put up with the complexity, fuel cost and nausea. Attempts to make the bridge of a star ship look like a bicycle will be futile, and for a space-ship designer advice from bike riders is pretty useless.
But for interplanetary travel a bicycle is no good, and you will have no choice as to put up with the complexity, fuel cost and nausea. Attempts to make the bridge of a star ship look like a bicycle will be futile, and for a space-ship designer advice from bike riders is pretty useless.
I still hope for Bicycles that can travel in space. In fact, I think that if the Aquarium GUI didn't have the Ribbon System, separation of modes, and unneeded options complexity, it would have been close to that.
![[is]](/mwf/v12/flags/is.png "Iceland")
> separation of modes
This is actually one of the strengths of Aquarium. This kind of separation is used in programs which offer a great number of different features. One example is Outlook which uses the same system (one of Microsoft's most complex products). In my experience this is absolutely mandatory for new users. For experienced users like myself it's a godsend and in practice I very seldom switch modes. Based on what you say,
> I use 99% of my time analyzing my chess games, and only play engine matches rarely
you would also almost always stay in Database mode. You would have access to all the analysis tools there and none of the "fluff" that is for unrelated tasks such as engine matches.
This is just good design as experienced UI designers will tell you. Otherwise we would have the situation described by h.g.muller:
> As soon as the number of buttons that is needed grows to a few hundred, having them all in one-click range would be a disaster.
> You would have access to all the analysis tools there and none of the "fluff" that is for unrelated tasks such as engine matches.
We already talked about the "fluff", I showed up screenshots that showed that the "fluff" used by the separation of modes is higher than the "fluff" used by those modes (if they weren't separated) as the "fluff" is usually compacted in menu items that use less space than the Mode Switching Bar.
I also showed that this mode separation leads to more clicking and mouse navigation.
You claimed all I did was "pixel counting", but then, if screen usage, how much the user has to move the mouse, and how much the user has to click, to do something, is worse, it begs the question of what is being improved?
>One example is Outlook which uses the same system (one of Microsoft's most complex products).
And a reason I don't use Outlook. I'd rather use Gmail or Yahoo's mail interface, though I have to admit they've gotten worse with each update.
>In my experience this is absolutely mandatory for new users. For experienced users like myself
It seems you're talking about specific experience with Aquarium. I'd like to challenge you to a correspondence chess match in where I'd be, I believe, according to you, outmatched because I wouldn't use Aquarium nor IDEA, software and tools that I've read somewhere are now "essential" for analysis, so I'd expect I wouldn't be a problem for you.
The match would not have a time limit (but with expectation of average 48 hours/move) so you could play it in your free time.
>> One example is Outlook which uses the same system (one of Microsoft's most complex products).
> And a reason I don't use Outlook. I'd rather use Gmail or Yahoo's mail interface, though I have to admit they've gotten worse with each update.
I agree! I don't use outlook either and yes Gmail's interface has gotten worse with each update. I will say that Microsoft's use of the ribbon has gotten a little better in office but i do agree with you that it does lead to more clicking. (which is why I don't use Aquarium...although Aquarium does have some great things about it)
> which is why I don't use Aquarium...although Aquarium does have some great things about it
Yeah, if Winboard is a Spaceship, Aquarium is a Time Machine
> I also showed that this mode separation leads to more clicking and mouse navigation.
The things you mention are of little, if any, practical value because they are not a good representation of the usage profiles of actual users. I can confidently state that based on my own experience, other users I've been in contact with, and training sessions that I have given. I'll repeat what I've pointed out before. No disrespect, but the words of actual users weigh more regarding pros and cons than those who don't use Aquarium at all (like yourself):
"I've been in contact with more Aquarium users than most. That experience has shown that the cleanly separated modes are actually a great help for users. You only see the things that are relevant to what you are doing (e.g. playing engine matches) instead of having the GUI cluttered with a huge number of buttons and options that have nothing to do with the task that you are working on."
Since Aquarium offers so many features that go above and beyond what Shredder GUI can do, it's not a good comparison regarding UI design decisions. A good solution for Shredder isn't automatically a good solution for Aquarium. A bicycle bell is a good solution for bicycles, but not for a motorcycle.
> I'd like to challenge you to a correspondence chess match
I have set myself certain goals in correspondence chess and I have already allocated the time I have for this hobby until next year.
If the purpose of your challenge was to understand if Aquarium is severely hampered by the faults that you see with the program, I think that question was convincingly answered by the results of WBCCC 2011 where some of the strongest players on the forum participated. It was won by ppipper who used Aquarium for his analysis.
A game between the two of us wouldn't add much to that result. My guess is that you would still doubt the value of Aquarium if it was the preferred analysis tool of the correspondence world champion.
> No disrespect, but the words of actual users weigh more regarding pros and cons than those who don't use Aquarium at all (like yourself):
On the contrary, if you base your opinion only on people that use and love Aquarium, ignoring the opinions of the people that have used and hated it, you're doing nothing more than survivor bias.
If 80% of the users didn't like it and left, and 20% stayed, you'd only be able to communicate with that 20%, and see that 100% of that 20% matches your experience, gives good reviews, can't live without Aquarium, etc. That all people that use Aquarium loves it doesn't show anything, though.
>A bicycle bell is a good solution for bicycles, but not for a motorcycle.
And on this, if you find the motorcycle your main vehicle of transportation, and only talk with people that do it, you're leaving out all the people that don't like that motorcycle, because they're not here anymore. My claim is that a motorcycle that is well designed, and does what your one does without being hard to use and confusing, could be built, and optimizing screen usage, and mouse navigation and clicking would be a good start.
You keep making general comments, when one goes into specifics (the "fluff"of the GUI and wasted user time clicking and navigating) you back out to the general "in my experience and the experience of people that are in love with Aquarium..."
>A game between the two of us wouldn't add much to that result.
Unless I defeated you or drew you without problems without Aquarium. I'm sure that's a result you wouldn't expect.
> if you base your opinion only on people that use and love Aquarium.
Where did I say that I only base my opinion on people that use and love Aquarium?
You start from a false premise so what you are saying makes no sense as a reply to my post.
> if you find the motorcycle your main vehicle of transportation, and only talk with people that do it
You make the same error here.
> Unless I defeated you or drew you without problems without Aquarium. I'm sure that's a result you wouldn't expect.
Wrong assumption again! Anything can happen in a single game of chess as most players realize. You are a good player. Drawing or losing isn't unexpected and neither is beating you.
I don't really understand where you are trying to take this discussion.
> I don't really understand where you are trying to take this discussion.
Huh? You started the discussion, I just posted this:
>>>>>>In fact, I think that if the Aquarium GUI didn't have the Ribbon System, separation of modes, and unneeded options complexity, it would have been close to that.
And didn't mean to discuss it further. Then you claimed that people that use Aquarium are happy with the separation of modes, Ribbon system, and unneeded options complexity. I do not dispute that. What I dispute is that you're focusing on the wrong population of users.
Yes, I think more users would have remained using Aquarium without such problems, including Kapaun, me, other beta testers that no longer used the program and other Aquarium users that complained on the forum but are no longer around.
The users that still use Aquarium would have used it anyway even if it had a Menu system instead of Ribbons.
Anyway, I don't expect this discussion to change anything in Aquarium, but I still hope that interface developers reading this can take Aquarium as an example of how NOT to do things. If your GUI is powerful, power users are going to stick with it, but you don't have to make design decisions that only keep power users happy, but make others run away.
> Huh?
Yes, instead of focusing on the topic, you:
1) Support your views by making false assumptions as I have pointed out several times.
2) Challenge me to a game of chess which also has nothing to do with what we were discussing
> Then you claimed that people that use Aquarium are happy with the separation of modes, Ribbon system, and unneeded options complexity.
I have to correct you once more. The only point I raised in the reply to your post was about different views or modes provided by Aquarium. You, however, took the discussion all over the place, challenging me to a game of chess, making false assumptions and now this... a statement which simply isn't true.
> What I dispute is that you're focusing on the wrong population of users.
Wrong assumption again and repeating it won't make it true. Obviously you know very little about me.
> Yes, I think more users would have remained using Aquarium without such problems, including Kapaun, me, other beta testers that no longer used the program and other Aquarium users that complained on the forum but are no longer around.
Churn is a well-known business issue and it hits most companies one way or another. Instead of your normal microscopic view try to look at the whole picture. Think about the thousands of users who have taken up Aquarium in recent years. Many of them leaving Fritz, Shredder, etc. for Aquarium. Why did they do that? Compare that to the few users on the list you have been collecting for so long (How kinky is that? Maintaining such a list for software that you don't even use
).Clearly there are a lot of people who don't support your hostile view towards Aquarium. You don't see those people writing hundreds of posts over a period of several years criticizing the software they used before switching to Aquarium. Normal people are just happy with their current choice and that's what matters to them. Maybe you can learn something from them?
This thread is not about Aquarium, but about Winboard, and also Tarrasch GUI, and their programmers (and perhaps other programmers interested reading this), they can see Aquarium, and decide to go for its design choices or not. If they only hear your side of the story of "thousands of users who have taken up Aquarium in recent years. Many of them leaving Fritz, Shredder, etc. for Aquarium", they may decide for going to a Ribbon System and mode separation, while I claim such things aren't necessary for powerful implementation of features like IDEA and Aquarium Trees (which I believe is the real reason of users switching to Aquarium, they can automate position analysis in a way not available on other GUIs, and store their analysis effectively.)
When the Fritz GUI moved to the Ribbon Interface I didn't see a single user jumping in joy saying something like "oh yeah! That's superior to the Menu system, I'm glad they're making the move", but I saw several critiquing the move as GUI developers not knowing what to implement and just doing that to look current, just like with the Mate-o-meter, the spy eye, and whatnot. Switching to the Ribbon System was a blunder, and I hold a Menu System is superior (in navigation and number of clicking, something you've been unable to refute).
So my attacks are more against those "features" that Aquarium has, and not specifically about Aquarium.
Now, if I misread your posts and built a straw-man to attack, I apologize, but I still think that points I've made are valid, and think that there are people that share my hostile point of view of Aquarium, and people that love Aquarium that agree it's un-user friendly and overly complex.
Because there's not much to critique about them. The equivalent you'd see would be like "I can't let the GUI analyzing important positions by itself while it allows me to interact and guide the analysis", and "the GUI doesn't allow me to store my analysis in a tree view to check later, instead having me rely on PGNs that suck at transpositions and that become an unreadable mess as more positions are added", "that's why I switched to Aquarium". So, Aquarium could have a lot more troubles but by being the only GUI with some power features, users would be willing to put up with any problems just to use those features. It doesn't mean the flaws aren't there.
Nothing is being achieved with these discussions because they're not for making a change in Aquarium, they're for enticing GUI developers to try to implement the power features but not the bad design decisions.
Those sound like conformists and I'm not one. I miss some features that Aquarium that would have worked for me if it didn't take so many time to use them, or learn to use them. So, yes, I'm looking forward to having different Chess GUIs having such power features implemented, and I am hopeful, as it's exactly what happened with engine learning. I've been trying to make engine developers aware of "Position Learning" and getting them to implement the feature, and more and more engines have it, I see no reason for Chess GUIs to lack features like Trees to store analysis or some Automated Interactive Analysis feature like IDEA, Winboard has been improving in the Analysis feature department with each new version, and it could become a fad.
So, people were happy with Rybka 4.1, and with Critter 1.2, that does not mean one should be a conformist and call it a day. Critter 1.4 came with an excellent Session File feature so it could store analysis (superior to other methods because the engine can use the analysis), Rybka 5 could have Persistent Hash back. I claim this progress is more meaningful than just adding elo points to engines, and that the same could happen to Chess GUIs.
By following your advice I'd have just been happy that Shredder and Rybka 3 have learning, and be done with it.
When the Fritz GUI moved to the Ribbon Interface I didn't see a single user jumping in joy saying something like "oh yeah! That's superior to the Menu system, I'm glad they're making the move", but I saw several critiquing the move as GUI developers not knowing what to implement and just doing that to look current, just like with the Mate-o-meter, the spy eye, and whatnot. Switching to the Ribbon System was a blunder, and I hold a Menu System is superior (in navigation and number of clicking, something you've been unable to refute).
So my attacks are more against those "features" that Aquarium has, and not specifically about Aquarium.
Now, if I misread your posts and built a straw-man to attack, I apologize, but I still think that points I've made are valid, and think that there are people that share my hostile point of view of Aquarium, and people that love Aquarium that agree it's un-user friendly and overly complex.
>You don't see those people writing hundreds of posts over a period of several years criticizing the software they used before switching to Aquarium
Because there's not much to critique about them. The equivalent you'd see would be like "I can't let the GUI analyzing important positions by itself while it allows me to interact and guide the analysis", and "the GUI doesn't allow me to store my analysis in a tree view to check later, instead having me rely on PGNs that suck at transpositions and that become an unreadable mess as more positions are added", "that's why I switched to Aquarium". So, Aquarium could have a lot more troubles but by being the only GUI with some power features, users would be willing to put up with any problems just to use those features. It doesn't mean the flaws aren't there.
Nothing is being achieved with these discussions because they're not for making a change in Aquarium, they're for enticing GUI developers to try to implement the power features but not the bad design decisions.
>Normal people are just happy with their current choice and that's what matters to them. Maybe you can learn something from them?
Those sound like conformists and I'm not one. I miss some features that Aquarium that would have worked for me if it didn't take so many time to use them, or learn to use them. So, yes, I'm looking forward to having different Chess GUIs having such power features implemented, and I am hopeful, as it's exactly what happened with engine learning. I've been trying to make engine developers aware of "Position Learning" and getting them to implement the feature, and more and more engines have it, I see no reason for Chess GUIs to lack features like Trees to store analysis or some Automated Interactive Analysis feature like IDEA, Winboard has been improving in the Analysis feature department with each new version, and it could become a fad.
So, people were happy with Rybka 4.1, and with Critter 1.2, that does not mean one should be a conformist and call it a day. Critter 1.4 came with an excellent Session File feature so it could store analysis (superior to other methods because the engine can use the analysis), Rybka 5 could have Persistent Hash back. I claim this progress is more meaningful than just adding elo points to engines, and that the same could happen to Chess GUIs.
By following your advice I'd have just been happy that Shredder and Rybka 3 have learning, and be done with it.
> If they only hear your side of the story
You can hear many different views about Aquarium here on the forum. However, there is only one person who neither uses or understands Aquarium who continues writing hundreds of posts about it. That person is you.
> I claim such things aren't necessary for powerful implementation of features like IDEA and Aquarium Trees
This sentence pretty much summarizes your ignorance of Aquarium. Of course they aren't necessary just for IDeA and Aquarium trees. You can work perfectly well with both features without switching modes.
> Now, if I misread your posts and built a straw-man to attack, I apologize, but I still think that points I've made are valid
Actually, most of what you have written in reply to my posts is total nonsense as I have clearly pointed out.
> think that there are people that share my hostile point of view of Aquarium
Such hostility is quite rare. I've never seen anything like your relentless hostility before.
> people that love Aquarium that agree it's un-user friendly and overly complex
If you single out the features that are common with Aquarium and other similar software, I think the complexity is approximately the same. The hard part for some people are the novel features not found in any other software, such as IDeA.
>> Normal people are just happy with their current choice and that's what matters to them. Maybe you can learn something from them?
> Those sound like conformists and I'm not one.
Don't flatter yourself! The thousands of people who have switched to Aquarium are conformists?! Well, they switched to Aquarium because they like it more than the software they used before. Hardly the trademark of a conformist. Are they conformists because they are happy with the switch to Aquarium? Not at all! Aquarium users are very vocal and there are literally hundreds of improvements in Aquarium which were based on user feedback. They criticize and suggest improvements to the software that they are using because they like it and want it to become even better.Let's take a look at yourself for comparison. You like Shredder. I hardly ever see you criticize Shredder or suggest improvements. Instead, when Shredder doesn't meet your requirements, you complement it with another piece of software. You transfer lines and positions between the the two and spend time manually entering information, evaluations, coloring moves etc. (Now, how many clicks and pixels is that
).For a self-appointed non-conformist it's surprising see you present this view about the software previously used by Aquarium users:
> there's not much to critique about them.
You seem to be the ultimate conformist
> I've been trying to make engine developers aware of "Position Learning" and getting them to implement the feature
The difference is that position learning is a simple concept which you are actually familiar with and use and understand. That's in stark contrast with Aquarium which you neither use nor understand.
Therefore I fully support your efforts in pushing engine authors towards better analysis features. If I felt that you were talking nonsense (as you do with Aquarium), then I wouldn't do that.
Personally, I don't understand how you can have such a deep repulsion for a product that motivated you into lending your creative ideas and applied skills to.
It wasn't as if you were hired to do the job and had to do it or get fired.
I consider Aquarium for better or worse a collaboration between artisans both amateur and professional responding to a community of like minded individuals seeking a product that demonstrates and meets that communities particular needs. Some how it failed to meet your expectations and you have worked hard to garner like minded individuals into your corner and explode it. But for what purpose?
If Fritz does it for you or Shredder! That's a good thing isn't it?
You have a great deal of computer chess knowledge under belt - and I learn something from you every day when you post the info freely. You are a great asset to this forum for your chess knowledge and skills.
But when it comes to Aquarium you seem to pull a Larry David.
It wasn't as if you were hired to do the job and had to do it or get fired.
I consider Aquarium for better or worse a collaboration between artisans both amateur and professional responding to a community of like minded individuals seeking a product that demonstrates and meets that communities particular needs. Some how it failed to meet your expectations and you have worked hard to garner like minded individuals into your corner and explode it. But for what purpose?
If Fritz does it for you or Shredder! That's a good thing isn't it?
You have a great deal of computer chess knowledge under belt - and I learn something from you every day when you post the info freely. You are a great asset to this forum for your chess knowledge and skills.
But when it comes to Aquarium you seem to pull a Larry David.
You misunderstand the three button concept. They don't represent all the functionality, I have complex .pgn editing functionality for example. There is no need for a button for every possible feature. Some buttons are provided to make very important and common functions simple and obvious. Tarrasch doesn't run under Windows 2000, which is why it doesn't work for you (addressed on another thread). I admit this is suboptimal.
> You misunderstand the three button concept. They don't represent all the functionality, I have complex .pgn editing functionality for example. There is no need for a button for every possible feature
+1
Hi Uli,
1 - There is no doubt that engine management in Tarrasch is rudimentary. I keep all engines in the same directory, I use Rybka, Houdini, Stockfish and Komodo (as well as my own Tarrasch Toy engine for blindfold training for example) without problems in this fashion. I am happy enough to run them with default parameters, so changing engines is just a few mouse clicks. Directory navigation as such is never necessary with this setup. I completely understand that sophisticated users have extra needs. In truth Tarrasch is probably not the gui for such people. Although of course a proper engine management system is on the wishlist.
2- Yes another reflection of the rudimentary engine management. I added the custom UCI parameters as a workaround for keen, sophisticated users. It's a shame the feature is insufficient for your needs, I'll certainly bear this in mind for the next round of improvements. Incidentally, yesterday I added a "no-install" Tarrasch .zip download. Tarrasch is lightweight enough (just a couple of megabytes for the .exe) that you could easily unzip multiple copies of this to multiple locations, then customise each of the .ini files for particular purposes. Admittedly a workaround rather than a good solution.
3- I downloaded Critter for the first time and tried it out without problems (it looks to be a nice engine, kudos to its author who has posted on this thread and could conceivably read this). The problem you had is that by default Critter uses its own book (OwnBook=true) and you asked it to analyse the opening position which (of course) is in its book. It refuses to do this and just supplies the "Best Move" instead. Arguably this is an engine bug, I have noticed Stockfish does exactly the same thing. Certainly I should detect and report the situation properly. Apart from this quirk, Critter has no problems kibitzing (yes, infinite analysis, I use the same terminology as ChessBase). I show a screenshot below
4- I wonder if you are using Tarrasch V2.00a, it had a rather nasty bug that in some circumstances eliminated the 4th kibitz line. That bug, plus a decoupling of engine and game directory navigation that greatly helps with problem 1 above, was fixed for V2.01a. It is true that Tarrasch is always in 4 line multi-PV mode, this is a compromise solution that is good in most circumstances, a reflection of Tarrasch's philosophy.
5- I think your objection here reflects the reality that Tarrasch is designed for chess players rather than computer chess addicts. It just might not be the GUI for you. I much prefer a large area for moves, commentary and analysis. Tarrasch is now a very good .pgn editor. Just click anywhere and start typing to enter comments or variations. I've got so used to this that I expect it other programs and part of me is always surprised that such an obvious thing is only implemented in Tarrasch (I could be wrong about that - I haven't used every GUI by any means).
I should collate all my scraps of information and put together a coherent road map for the future. I am not going to do it right now, I will put it on my blog in due course. At the moment I suspect the basic plan will be V2.1=revamp screen (customisable, resizable graphics, more flexible layout, panel sizes). V2.1=proper engine management system. V3.0=Engine v Engine. It is hard to find time to work on Tarrasch, I try nbt to put myself under too much pressure to move the program forward, after all it is free, and I am quite proud of what I have already achieved.
1 - There is no doubt that engine management in Tarrasch is rudimentary. I keep all engines in the same directory, I use Rybka, Houdini, Stockfish and Komodo (as well as my own Tarrasch Toy engine for blindfold training for example) without problems in this fashion. I am happy enough to run them with default parameters, so changing engines is just a few mouse clicks. Directory navigation as such is never necessary with this setup. I completely understand that sophisticated users have extra needs. In truth Tarrasch is probably not the gui for such people. Although of course a proper engine management system is on the wishlist.
2- Yes another reflection of the rudimentary engine management. I added the custom UCI parameters as a workaround for keen, sophisticated users. It's a shame the feature is insufficient for your needs, I'll certainly bear this in mind for the next round of improvements. Incidentally, yesterday I added a "no-install" Tarrasch .zip download. Tarrasch is lightweight enough (just a couple of megabytes for the .exe) that you could easily unzip multiple copies of this to multiple locations, then customise each of the .ini files for particular purposes. Admittedly a workaround rather than a good solution.
3- I downloaded Critter for the first time and tried it out without problems (it looks to be a nice engine, kudos to its author who has posted on this thread and could conceivably read this). The problem you had is that by default Critter uses its own book (OwnBook=true) and you asked it to analyse the opening position which (of course) is in its book. It refuses to do this and just supplies the "Best Move" instead. Arguably this is an engine bug, I have noticed Stockfish does exactly the same thing. Certainly I should detect and report the situation properly. Apart from this quirk, Critter has no problems kibitzing (yes, infinite analysis, I use the same terminology as ChessBase). I show a screenshot below
4- I wonder if you are using Tarrasch V2.00a, it had a rather nasty bug that in some circumstances eliminated the 4th kibitz line. That bug, plus a decoupling of engine and game directory navigation that greatly helps with problem 1 above, was fixed for V2.01a. It is true that Tarrasch is always in 4 line multi-PV mode, this is a compromise solution that is good in most circumstances, a reflection of Tarrasch's philosophy.
5- I think your objection here reflects the reality that Tarrasch is designed for chess players rather than computer chess addicts. It just might not be the GUI for you. I much prefer a large area for moves, commentary and analysis. Tarrasch is now a very good .pgn editor. Just click anywhere and start typing to enter comments or variations. I've got so used to this that I expect it other programs and part of me is always surprised that such an obvious thing is only implemented in Tarrasch (I could be wrong about that - I haven't used every GUI by any means).
I should collate all my scraps of information and put together a coherent road map for the future. I am not going to do it right now, I will put it on my blog in due course. At the moment I suspect the basic plan will be V2.1=revamp screen (customisable, resizable graphics, more flexible layout, panel sizes). V2.1=proper engine management system. V3.0=Engine v Engine. It is hard to find time to work on Tarrasch, I try nbt to put myself under too much pressure to move the program forward, after all it is free, and I am quite proud of what I have already achieved.
> The problem you had is that by default Critter uses its own book (OwnBook=true) and you asked it to analyse the opening position which (of course) is in its book. It refuses to do this and just supplies the "Best Move" instead. Arguably this is an engine bug, I have noticed Stockfish does exactly the same thing
I guess all engines with own book do that. Yet, all such engines work fine in infinite analysis in other GUIs, so I guess it's a problem with missing commands not being sent by the GUI.
Anyway, looking forward to future improvements.
Actually it happens in most GUI that I know of. The solution is to check off "own book" and it doesn't have this problem anymore.
>> The problem you had is that by default Critter uses its own book (OwnBook=true) and you asked it to analyse the opening position which (of course) is in its book. It refuses to do this and just supplies the "Best Move" instead. Arguably this is an engine bug, I have noticed Stockfish does exactly the same thing
> I guess all engines with own book do that. Yet, all such engines work fine in infinite analysis in other GUIs, so I guess it's a problem with missing commands not being sent by the GUI.
>
> Anyway, looking forward to future improvements.
I had the same problem with Critter in the Shredder GUI. (the "own book" logjam)
added to my TODO list
> added to my TODO list
Thanks Richard, although I don't think it is that big a deal. I was just confirming to Uly that it also occurs in other GUIs.
It only takes me two seconds to click off "own book". Critter is a great engine!
Thanks for this. I've heard good things about your engine in the past but never tried it out before. Congratulations on producing such an accomplished piece of work.
>>and I am quite proud of what I have already achieved.
@Bill Forster; rightfully so. It's a good gui.
<3 Tarrasch :-)
> If you BUY Chessbase or Shredder, can you then bother them with unlimited feature requests that they will implement for you within a week for you?
And this is compared to what? I ran away from Winboard just because of the clutter on the loading dialog, I don't want to deal with that every time I want to load and engine. Your answer? It's already an optimal implementation and I'd have to become a programmer just to fix that.
HGM: "Screwed? Because someone is friendly enough to make it an almost a full-time job to to give non-programmers what they want, if it is feasible and not too silly?"
Yes indeed. You merit much Kudos for this. Truthfully, absolutely no one has a right to complain.
Much here is a matter of taste methinks. Also note. There certainly is no harm in offering suggestions to improve (beauty is in the eye of the beholder - yes?)the GUI .
Now. If we were only just as good at our verbal transactional abilities as we are at chess.....
Yes indeed. You merit much Kudos for this. Truthfully, absolutely no one has a right to complain.
Much here is a matter of taste methinks. Also note. There certainly is no harm in offering suggestions to improve (beauty is in the eye of the beholder - yes?)the GUI .
Now. If we were only just as good at our verbal transactional abilities as we are at chess.....
Compared to what??? Why should it be compared to anything? It was an absolute question, which can be answered by a simple "yes" or "no".
But what you say is not entirely accuratel. Because of your input on this:
1) WinBoard can now load recently used engines directly from the menu, so that you don't even have to open that dialog that you so abhor in most cases.
2) I put it on the to-do list to replace the Select Engine combo-boxes in the Load Engine and Tournament dialog by list boxes, as soon as they become available.
That reminds me I still had a question about the Shredder Load dialog: why is there a button 'Load' in the lower-right corner? Does it have to be clicked after you select an engine from the listbox, or does the engine load automatically when you select it?
But what you say is not entirely accuratel. Because of your input on this:
1) WinBoard can now load recently used engines directly from the menu, so that you don't even have to open that dialog that you so abhor in most cases.
2) I put it on the to-do list to replace the Select Engine combo-boxes in the Load Engine and Tournament dialog by list boxes, as soon as they become available.
That reminds me I still had a question about the Shredder Load dialog: why is there a button 'Load' in the lower-right corner? Does it have to be clicked after you select an engine from the listbox, or does the engine load automatically when you select it?
> That reminds me I still had a question about the Shredder Load dialog: why is there a button 'Load' in the lower-right corner? Does it have to be clicked after you select an engine from the listbox, or does the engine load automatically when you select it?
User can click the button to load the selected engine, or double click the selected engine to load it.
So if I count right, installing an engine that is initially visible in the listbox (because it was recently used, assuming that the Recent checkbox is ticked by default) takes 3 clicks: one to summon up the Load dialog, (e.g. on the engine name in the analysis window), one to select the engine (in the listbox), and one to load it (again on the engine, or on the load button).
Well, in WinBoard the same action would take only two clicks: on the Engine main menu, and then on the engine. In XBoard it even takes only a single click, because there the down-click can be used to open the menu, and the corresponding up-click to select the item from the menu.
I don't consider that inferior enough to the Shredder method to completely re-design things. In fact, it is what I would call objectively better, because 2 < 3... But of course this is just because of my inexperience with other GUIs, and I am sure you can explain us why 2 is in reality is much larger than 3.
Well, in WinBoard the same action would take only two clicks: on the Engine main menu, and then on the engine. In XBoard it even takes only a single click, because there the down-click can be used to open the menu, and the corresponding up-click to select the item from the menu.
I don't consider that inferior enough to the Shredder method to completely re-design things. In fact, it is what I would call objectively better, because 2 < 3... But of course this is just because of my inexperience with other GUIs, and I am sure you can explain us why 2 is in reality is much larger than 3.
Of course, I was talking about a Winboard version that didn't show engines to switch on the menu.
Powered by mwForum 2.27.4 © 1999-2012 Markus Wichitill