Release of XBoard 4.9.0 soon
Moderators: hgm, Rebel, chrisw
-
- Posts: 1525
- Joined: Tue Jul 15, 2014 12:47 pm
Re: Release of XBoard 4.9.0 soon
And where the program can be downloaded and is there anything else I need to additionally download to play checkers 8x8.
-
- Posts: 730
- Joined: Fri Oct 19, 2012 2:23 am
-
- Posts: 1525
- Joined: Tue Jul 15, 2014 12:47 pm
Re: Release of XBoard 4.9.0 soon
What do I need to download from this site, so to play checkers.JoshPettus wrote:http://hgm.nubati.net/alien.html
-
- Posts: 1525
- Joined: Tue Jul 15, 2014 12:47 pm
Re: Release of XBoard 4.9.0 soon
I do as I ask for very detailed instructions on how it all started.I would add that the first time in my life to run this program.
-
- Posts: 125
- Joined: Tue Jan 17, 2012 8:08 pm
Re: Release of XBoard 4.9.0 soon
Can you swear on the Bible Xboard 4.9.0 is bug free
-
- Posts: 27809
- Joined: Fri Mar 10, 2006 10:06 am
- Location: Amsterdam
- Full name: H G Muller
Re: Release of XBoard 4.9.0 soon
It is easy enough to swear on the bible. But that of course doesn't make it true.
I fact I discovered and fixed a new bug only yesterday evening: one-click moving was not always working with engine-defined pieces, because in disambiguating SAN moves, when a move is ambiguous under the engine-specified rules, XBoard tries to decide based on the built-in 'native' move of the piece, rather than the move redefined by the engine. (Making the assumption that it could be dealing with a game saved before engine-defined piece moves existed.) But one-click moving is done by disambiguating moves like g1??, and in general these are ambiguous, so that it waits for the user to click a to-square. But with engine-defined pieces XBoard then first tried to move the piece by XBoard's own rules, and sometimes the piece did have only a single move according to those. And then it would attempt that, and immediately reject it as illegal, so you could not move the piece at all. It should of course only use this back-up when parsing SAN from a file (where the to-square is always fully specified), not when deciding about one-click moving (which should always strictly follow the engine-specified rules.)
Another problem that currently is not solved yet is that when used with a tiling window manager like i3wm, the board does not immediately expand to fill the window that the WM assigned, and only does that after the user slightly resizes that. It seems that i3wm 'silently' makes the initial window assignment, i.e. without telling XBoard or GTK that the window has a different height from the requested one.
So it is like always: this version has fewer bugs in features that the previous version already had, but can have new bugs in features that did not exist before. Which seems an improvement, because in the previous version you could not use these features at all, so it doesn't really matter how buggy they are. I mean, not many users would be interested writing Japanese kanji over their Chess pieces, so why would they care if the kanji were sometimes written in the wrong color?
I fact I discovered and fixed a new bug only yesterday evening: one-click moving was not always working with engine-defined pieces, because in disambiguating SAN moves, when a move is ambiguous under the engine-specified rules, XBoard tries to decide based on the built-in 'native' move of the piece, rather than the move redefined by the engine. (Making the assumption that it could be dealing with a game saved before engine-defined piece moves existed.) But one-click moving is done by disambiguating moves like g1??, and in general these are ambiguous, so that it waits for the user to click a to-square. But with engine-defined pieces XBoard then first tried to move the piece by XBoard's own rules, and sometimes the piece did have only a single move according to those. And then it would attempt that, and immediately reject it as illegal, so you could not move the piece at all. It should of course only use this back-up when parsing SAN from a file (where the to-square is always fully specified), not when deciding about one-click moving (which should always strictly follow the engine-specified rules.)
Another problem that currently is not solved yet is that when used with a tiling window manager like i3wm, the board does not immediately expand to fill the window that the WM assigned, and only does that after the user slightly resizes that. It seems that i3wm 'silently' makes the initial window assignment, i.e. without telling XBoard or GTK that the window has a different height from the requested one.
So it is like always: this version has fewer bugs in features that the previous version already had, but can have new bugs in features that did not exist before. Which seems an improvement, because in the previous version you could not use these features at all, so it doesn't really matter how buggy they are. I mean, not many users would be interested writing Japanese kanji over their Chess pieces, so why would they care if the kanji were sometimes written in the wrong color?
-
- Posts: 125
- Joined: Tue Jan 17, 2012 8:08 pm
Re: Release of XBoard 4.9.0 soon
I have 2 questions :
Which pre release should I take ? Yours : http://hgm.nubati.net/cgi-bin/gitweb.cg ... ;a=summary
or http://alpha.gnu.org/gnu/xboard/xboard- ... 402.tar.gz
How do I install a translation language from the "PO" directory ?
Thank you.
Which pre release should I take ? Yours : http://hgm.nubati.net/cgi-bin/gitweb.cg ... ;a=summary
or http://alpha.gnu.org/gnu/xboard/xboard- ... 402.tar.gz
How do I install a translation language from the "PO" directory ?
Thank you.
-
- Posts: 27809
- Joined: Fri Mar 10, 2006 10:06 am
- Location: Amsterdam
- Full name: H G Muller
Re: Release of XBoard 4.9.0 soon
Use the one from GNU, it has the latest translations.
The translations should be built into XBoard automatically, I think. (Could be that you would have to enable this with "./configure --enable-nls", but I thought that was default. It should automatically adapt to the locale of your system. You can overrule that by prefixing the xboard command with LANG=...
The translations should be built into XBoard automatically, I think. (Could be that you would have to enable this with "./configure --enable-nls", but I thought that was default. It should automatically adapt to the locale of your system. You can overrule that by prefixing the xboard command with LANG=...
-
- Posts: 2872
- Joined: Wed Mar 08, 2006 10:09 pm
- Location: Germany
- Full name: Werner Schüle
Re: Release of XBoard 4.9.0 soon
[quote="hgm][/quote]
Hi,
can you please show the settings in the tournament manager:
I want to play 100 games between 2 engines, I have 50 opening positions like nunn.pgn. Each opening is played with black and white.
I tried (with 4.80) and every game from the RR (0) is played with the same opening line.
Werner
Hi,
can you please show the settings in the tournament manager:
I want to play 100 games between 2 engines, I have 50 opening positions like nunn.pgn. Each opening is played with black and white.
I tried (with 4.80) and every game from the RR (0) is played with the same opening line.
Werner
-
- Posts: 4610
- Joined: Wed Oct 01, 2008 6:33 am
- Location: Regensburg, Germany
- Full name: Guenther Simon
Re: Release of XBoard 4.9.0 soon
Haven't used the internal tourney manager for a long time, but AFAIK youWerner wrote: Hi,
can you please show the settings in the tournament manager:
I want to play 100 games between 2 engines, I have 50 opening positions like nunn.pgn. Each opening is played with black and white.
I tried (with 4.80) and every game from the RR (0) is played with the same opening line.
Werner
should tick those two boxes:
Code: Select all
1. Step through lines/positions in file
2. Use each line/position twice