Thanks!
Click a file in file manager or open with scidb behaves differently
Error message is: ::scidb::engine::snapshot 0 all 0 InsertMoves .application.nb.board.__frame__analysis:1__3.analysis:1.mw.main.tree all 0 ::namespace inscope ::application::analysis {InsertMoves .application.nb.board.__frame__analysis:1__3.analysis:1.mw.main.tree all 0} ::tk::menu::WidgetProc .application.nb.board.__frame__analysis:1__3.analysis:1.mw.main.tree.__menu__ invoke active ::tk::menu::WidgetProc .application.nb.board.__frame__analysis:1__3.analysis:1.mw.main.tree.__menu__ invoke active...
Append all variations in a game with no moves -> application error
I mean right arrow key in the keyboard.
Error in game ending with checkmate
Some untranslated texts
Sorry about the not-so-exact error report! r1435 r1436 scidb-beta --first time new 3check game click locomotive button select Stockfish-> In the center of engine window appears the message: "This engine does not support variant 'Three-check Chess'" In normal games engine works, and Sjeng works normally.
No hurry! Yes, I get the message in the title when clicking the locomotive button when three-check-chess game is open in r1435. Could it then be the -O2 (if I remember correctly) compiling option of scidb I'm using?
Stockfish does not support Three-Check chess
This problem seems to be solved. First the overlapped row is shown, but it corrects by itself.
r1414 compilation error in Knoppix 7.7.1
The errors were in r1531.
I tried every option in the tree window configuration menu, and I didn't see any change in the appearance of the header files no matter what I tried. None of optimize/fit/shrink/expand squueze did anything, nor was there error messages. Changing minimal and maximal widths did not help. Here I could set minimal width to 20 and maximal to 10 and it was accepted. Maybe that's a genuine bug? It didn't affect the display though. Configuring the colours did gave error messages, which I give below. But...
I tried installing latest Linux Mint to USB stick and it seems to run pretty well, but compiler didn't work out of the box, if there was one. But I only have internet connection on public Windows computers to download Linux updates, and many distros seem to require a connection even just to install the distro. knoppix@Microknoppix:~$ cat /proc/cpuinfo processor : 0 vendor_id : AuthenticAMD cpu family : 15 model : 44 model name : Mobile AMD Sempron(tm) Processor 3000+ stepping : 2 cpu MHz : 800.000...
The downloads in the Slax homepage are some years old, and I don't have an internet connection to update the software. I didn't realize that. My mistake, there's probably no point ot solving this problem.
Suggestions for Tree window column headers
Game list error: Scidb->contains(database, variant)
Previous/next/first/last game buttons very slow
Opening of database is postponed with no obvious reason
r1336 Works fine now!
r1336 I cannot reproduce it. I recreated the game but with the correct crazyhouse board it didn't happen.
Yes I get readyok message from engine log. There's nothing wrong, my mistake, sorry.
Compiling error in Slax r1336
r1336 has the same bug: (-analysis is on, 4 tabs) -close tree window -start new normal or variant game ->error can't read "Vars(frame:tree)": no such element in array (func) invoke (file) tcl_base.cpp:677 (what) can't read "Vars(frame:tree)": no such element in array (type) tcl::Error === Backtrace ============================================ /usr/local/bin/tkscidb-beta() [0x84096ee]
r1336 still has this: right-click-menu-untick show coordinates right-click-menu-tick use small letters right-click-menu-tick show coordinates ->shows big letters also click-menu-tick use small letters click-menu-untick show coordinates click-menu-untick use small letters click-menu-show coordinates ->small coordinates are used, but not ticked
Closing undocked analysis window doesn't close it
It looks the same in r1336.
This error possibly came when board was not showing hand piece racks: select stockfish (ch game, no moves made) ->error can't read "Vars(number)": no such variable loadArray1 ::application::analysis::restartAnalysis 1 StartEngine 1 .setupEngine1.top.engine.list ::namespace inscope ::engine {StartEngine 1 .setupEngine1.top.engine.list} 3 3 3 3 ->start sjeng no error ->start stockfish no error
Window grab failed
Closing tree window produces error
Undocked analysis window disappears after starting engine
r1320 screenshot
Game header overlaps opening infromation row
Crazyhouse hand piece racks not shown
Thanks, I didn't notice that.
Sorry, I meant the window that is shown in the beginning on top of information tab when you start scidb --first-time. That has a lot of white space. Unfortunately Knoppix is the only one of many linuxes that I have tried that runs smoothly and beautifully in my hardware. The popular ones are very slow. Arch linux dropped official 32-bit support and I couldn't even figure it out yet how to install it to USB memory without linux machine that has an internet connection, with not advanced linux expertise....
I couldn't quickly reproduce it either. I must have had another normal game open that Stockfish analysed, is that possible? I'll check.
Help window too high?
Open database error
Stockfish uses 45% CPU even though it doesn't support variant
board options not updated
Starting analysis adds an extra tab to analysis window
r1313
board misaligned when changing window size?
I don't use VMs now. I can try another Linux, thanks for the advice. Maybe it is not useful anymore to report the error messages I get, but I noticed that when I have unsaved games I get a different error message (when selecting board options /select them) If there are no unsaved games, but a game history listing (twice the same error message): unknown option "-state" while executing "$w.__hist__ $command {*}$args" (procedure "::game::history::WidgetProc" line 29) invoked from within "::game::history::WidgetProc...
Maybe this is unrelated: board options - theme tick colors change background and border to wood patterns hover mouse on buttons tooltip shows for border "Solid Color: #999999 (Gray 60) even though border is wood01.jpg
r1285 If I have a position with promoted marked pieces on the board and change the mark from disc to star/bullet, the mark goes under the piece until cursor is moved onto the board. If I then change the mark back to disc, the disc comes on top of the piece until cursor is moved onto the board.
I can select themes, boards, and sets without any errors, only ticking on "colors" produces the errors. I can select there square patterns and they seem to work, but clearing them again with the eraser produces the errors. Here are some new ones, too: r1282 scidb-beta --first-time board options/ select theme tick Colors ->error can't read "colors(user,coordinates)": no such element in array while executing "llength $colors(user,coordinates)" (procedure "SetColorTooltips" line 16) invoked from within...
r 1282 I still get errors when starting with scidb-beta --first-time or removing .scidb-beta. Some new ones, too. I have a quite old laptop, too slow for many brands of linux, but knoppix is fine in general. Is it possible it's not up to demands of Tcl/Tk in performance? I have noticed computer display freeze for too long periods - window opening or closing may take many many seconds. Highlighting a file and trying to delete it asks a confirmatory question, but for a different file name. Can this...
r 1282 I still get errors when starting with scidb-beta --first-time or removing .scidb-beta. Some new ones, too. I have a quite old laptop, too slow for many brands of linux, but knoppix is fine in general. Is it possible it's not up to demands of Tcl/Tk in performance? I have noticed computer display freeze for too long periods - window opening or closing may take many many seconds. Highlighting a file and trying to delete it asks a confirmatory question, but for a different file name. Can this...
The same problem in r1253. After this I cannot select themes anymore without reinstalling scidb. main menu-theme-alt Board options / select theme basic theme -magnetic tick colors (bottom left) ->error can't read "colors(user,coordinates)": no such element in array while executing "llength $colors(user,coordinates)" (procedure "SetColorTooltips" line 16) invoked from within "SetColorTooltips" (procedure "ToggleLock" line 39) invoked from within "ToggleLock " (in namespace inscope "::board::options"...
Also after editing position more than once, some discs are on top, some below:
promotion discs on top of piece
After installing r1253: About Scidb - System: Tcl/Tk version: 8.6 (pl6) Operating System: Linux 4.7.9 OS Identifier: linux-glibc2.24-ix86 System Architecture: i686 Windowing System: x11 Vendor String: X11R0 The X.Org Foundation 11804000 Total Memory: 1891.2 MB (Copying the above from the "about" window didn't work, typed it.)
After installing r1253: About Scidb - System: Tcl/Tk version: 8.6 (pl6) Operating System: Linux 4.7.9 OS Identifier: linux-glibc2.24-ix86 System Architecture: i686 Windowing System: x11 Vendor String: X11R0 The X.Org Foundation 11804000 Total Memory: 1891.2 MB
Output of tk "make test". I didn't yet do the interactive tests. Attached file below.
output of "make test": attached file in the next post.
Tk test log file attached:
Should have atttached the log file, sorry. The Tcl test log file attached:
Output of tk "make test". I didn't yet do the interactive tests. gcc -c -O2 -pipe -Wall -fPIC -DBUILD_tk -I/home/knoppix/tk8.6.6/unix/../unix -I/home/knoppix/tk8.6.6/unix/../generic -I/home/knoppix/tk8.6.6/unix/../bitmaps -I/home/knoppix/tcl8.6.6/generic -I/home/knoppix/tcl8.6.6/unix -DPACKAGE_NAME=\"tk\" -DPACKAGE_TARNAME=\"tk\" -DPACKAGE_VERSION=\"8.6\" -DPACKAGE_STRING=\"tk\ 8.6\" -DPACKAGE_BUGREPORT=\"\" -DSTDC_HEADERS=1 -DHAVE_SYS_TYPES_H=1 -DHAVE_SYS_STAT_H=1 -DHAVE_STDLIB_H=1 -DHAVE_STRING_H=1...
output of "make test": gcc -c -O2 -pipe -Wall -fPIC -DBUILD_tcl -I"." -I/home/knoppix/tcl8.6.6/unix -I/home/knoppix/tcl8.6.6/generic -I/home/knoppix/tcl8.6.6/libtommath -DPACKAGE_NAME=\"tcl\" -DPACKAGE_TARNAME=\"tcl\" -DPACKAGE_VERSION=\"8.6\" -DPACKAGE_STRING=\"tcl\ 8.6\" -DPACKAGE_BUGREPORT=\"\" -DSTDC_HEADERS=1 -DHAVE_SYS_TYPES_H=1 -DHAVE_SYS_STAT_H=1 -DHAVE_STDLIB_H=1 -DHAVE_STRING_H=1 -DHAVE_MEMORY_H=1 -DHAVE_STRINGS_H=1 -DHAVE_INTTYPES_H=1 -DHAVE_STDINT_H=1 -DHAVE_UNISTD_H=1 -DHAVE_SYS_PARAM_H=1...
I tried first r1248 and then 1247. Identical error messages in compiling. I will try r1253.
Couldn't check because tried to reinstall everything, but will try to reproduce this after I get a working version.
From the tk documentation I got the impression that tk test errors are expected, probably not serious.
I tried to reinstall everything, but r1248 didn't compile. Forgot to check "system" before. I ran tests after installing Tcl/Tk as usally. Tcl didn't give errors, but Tk gave a couple as always. I wonder if it is due to mismatch with compiz window manager? I have reinstalled everything from scratch a few times because of errors that appear after scidb freezes/crashes. I don't know if they are related. There have been sed errors with etc/default/keyboardXXXXXX file when logging out, sometimes keyboard...
compiling error in r1247 and r1248
r1239 I'm not able to select theme anymore. Every other restart theme selection is not active in the menu, every other restart it gives an error. I did not reinstall yet. board options/select theme basic (untick show border, show coordinates) theme edit list select - preview - close tick show border, show coordinates ->error ->ok ->board options/select theme is greyed out from menu can't read "colors(user,coordinates)": no such element in array while executing "set tip "${::mc::Color}: extendColorName...
Sounds great, thanks!
I added possibly theme-related error messages and a screenshot in bug report #147 in r1239.
I tested different sets, boards and themes. I don't know what was selected when this happened but will check. In r1239 I got a theme related bug.
Crazyhouse engine sjeng - assertion failed error
Crazyhouse test game freezes scidb
Stop trial mode exits scidb
Selecting file type when exporting pgn - error
Setting promotion mark in crazyhouse, it shows in normal chess
Styles - ticking color or coordinates gives error
dragging PGN file from desktop - opening postponed
marks-palette error on empty move list
The idea of using only one king in the diagram and play is perhaps too far from the original idea of scidb. Engines probably don't support it either, even though the positions would be made for human study and learning. May I take back this request of one-king positions? I feel it will be a distraction for you.
The idea is that user setups the pieces on the board and hand pieces for the player to move (who typically in a tactical teaching position only has a few hadn pieces). The rest of the pieces would be given to the other colour. This is tradition in shogi, all pieces not shown in a tactical problem position are assumed to be defender's hand pieces. They are not shown in a problem diagram Only the defender's king is shown (losing side). All the pieces contribute to the attack, there are no unnecessary...
16 pawns in holding, for example: opponent captures all you pawns, then drops them on the board. Then you capture all the pawns.Not very likely, but legally possible. Other values 2/4/4/4 are right, if you only want to allow positions that can be reached from the normal setup (or fewer pieces).
Possibly I had the promotion button selected when clicking on squares.
Crazyhouse setup board errors
Promotion marks not cleared
Easier setting up of crazyhouse positions
Yes, now position with nine black pawns on the board can be setup. But holdings can only have 8 pawns of one colour, when 16 is possible.
Apologies, my math is incorrect. I thought about pieces on the board. Pawns in hand can be as many as 16 for one colour. Other figures are of course wrong.
Crazyhouse Set Start Board does not accept legal positions
Crazyhouse Set Start Board errors
r1224 compiles in Knoppix.
Thanks, I was wrong.
I have the same error in Knoppix in r1220. Also several earlier revisions had the same error, I think at least r1215, r1217. Did not try r1221.