Wouldn't it make more sense if - for games downloaded from TWIC (and
other similar sites) - game information presented in Scid simply
contained clickable link (maybe just in Source tag)?
This would give some intended visits...
Larry Isaacs
writes:
> I'm using Ubuntu 8.04 LTS 32bit with Wine 1.2.2 and am able to run the
> Houdini 1.5a exe. For the Command setting, the exe path (containing no
> spaces) works for me.
It depends on system configuration (whether given Linux runs .exe's via
wine by default, ol not).
Safe
> I like the ICC format, just a comment with the remaining time ( h:mm:ss )
> For example:
>
> 1. e4 {1:40:30} c6 {1:40:25} 2. d4 {1:40:02} d5 {1:40:50}
The problem here starts once you also have some true annotations.
I don't really like
1. e4 {1:40:30 Most popular opening move}
or
1. e4 {M
Below there is a patch for sc_import.tcl which adds an extra way to run
this command. If run with
sc_import somedatabase -
the command will read the names of PGN files to import from standard
input.
Why?
--
First, it makes it possible to do things like
find . -name '*.pgn' -a -mt
As a sidenote bonus, here is the game I discovered some time ago. Both
players are currently GMs (white was not at the time the game was
played, he was a very talented youngster then).
The game was played during European Rapid Chess Championship 2006.
Note moves 10, 26, and 31 :-)
[Event "Ampli
> You can't. Scid enforces the laws of chess here and does not accept
> invalid moves. In your case this is actually a drawback, in most others
> it is definitely an advantage.
Well, one need not exclude the other. Enforcing laws normally but having
some "force invalid move" option/button/whatev
I am just using Scid to enter the games my kids played today on their
first slow time tournament. Unfortunately, kids are not perfect. And
scid could help me a little bit more than it does...
Trouble:
a) Opp of my son castled under check at move 7. My son missed this and
they happily continued t
ine
there is an error (I routinely use this as "compilation" routine)
--
----------
| Marcin Kasperski | I have a very small head and I had better learn
| http://mekk.waw.pl | to live with it and to respect my limitations,
|
B, and
> seg fault is pending.
Maybe it would be possible to add some check for it, even bare
refusing to load such a game...
--
------
| Marcin Kasperski | A complex system designed from scratch
| http://mekk.waw.pl
Nevertheless...
--
------
| Marcin Kasperski | The goal of a software process is the
| http://mekk.waw.pl | production of software (...) Intermediate
|| artifacts are not the goal
"wine TogaII.exe" was noticeably faster than linux-compiled Toga
(the same version).
--
------
| Marcin Kasperski | The cost of a few uncorrected non-critical
| http://mekk.waw.pl | human errors is less then the cost impose
"Benoit St-Pierre" <[EMAIL PROTECTED]> writes:
> Incidently, that seems to offer a challenge for centralized, reference
> databases, like the Centriscid.
In case of my personal archive, I ended up keeping the database
splitted by eco (take not-so-large PGN files one by one, pgnscid them,
eco_clas
ime
I used to send 'date' ;-)
--
------
| Marcin Kasperski | If Staff, Scope and Schedule are all fixed,
| http://mekk.waw.pl | managers will have no options, other than
|
; commentary code within the body. (...)
I doubt anybody cares much about it when move emails are
considered. Some cleanup is needed mostly while submitting the
game score,
--
------
| Marcin Kasperski | Working overtime sucks the spi
d that ./configure scripts
allow to specify all important directories separately.
--
------
| Marcin Kasperski | You have the right to produce quality work
| http://mekk.waw.pl
> > The point of this file was to make it possible to build scid
> > in Linux "packaged" way, where binaries and data are not
> > installed together, but we use sth like /usr/bin/scid (or
> > /usr/games/scid) and /usr/share/scid/data/
>
> Given that books and bases directories are now user
> co
es and data are not installed
together, but we use sth like /usr/bin/scid (or /usr/games/scid) and
/usr/share/scid/data/
--
------
| Marcin Kasperski | The cost of a few uncorrected non-critical
| http://mekk.waw.pl | human erro
> So, it looks like all positions are cached? Or just those up
> to some move?
Up to some move. IIRC it is configurable while building the
direct tree, the standard is up to move 20 or so.
> What about matching games? Are they displayed
> immediately, or does it require some processing?
Some
similar could be probably achieved using
EPD files in the background (just providing functions to build them
from databases with statistical info, and integrating them in the
interface better).
--
------
| Marcin Kasperski
disk spinning, without an
opportunity of incremental build)
Well, but once the direct tree is here, it works very, very, very fast.
--
--
| Marcin Kasperski | Users typically do not know exactly what
| http://me
> I found that the last version that works painlessly in wine
> is Fritz 8, (V6 and 7 are also working). From V9 onwards a
> recent version of DirectX is required and I did not fiddle
> into this. Enginewise V7 and V8 can handle the Fritz engines
> up to V9, from V10 onwards CB seems to have change
> But note that it must not
> necessarily compile and stuff like that. (Well cvs ;)
That's obvious.
BTW: it could make sense to place cvs tags while building
releases ("cvs tag scid-3_6_19b2" etc)
--
----
hess.scid.user
--
------
| Marcin Kasperski | In any large change, 1/3 will think it is
| http://mekk.waw.pl | great, 1/3 will think it is stupid, and 1/3
|| w
> I remember a discussion on the issue for chessx and there is no doc
> about how to read cbh. All DOS-tools that use it just call a DLL
> from ChessBase (had a detailed discussion of the issue with Stefan
> from Shredder, why he does not support cbh in [EMAIL PROTECTED], and
> about data formats
Well, I decided to take a look at development version of scid2 in CVS
and ... had noticeable problems to find it. The website is not mentioning
CVS repo in any way, bah - it explicite says that there is no version
on sourceforge.
My suggestions:
- add a sentence about CVS repository on the Downl
> Does this mailing list actually support both?
Well, it is recently used mostly for development discussions
about scid continued. Not sure whether anybody active here is
still using original scid.
> Is one preferred over the other?
Well, is chessbase 9 preferred over chessbase 6? Or so...
O
ot of work)
--
------
| Marcin Kasperski | Users typically do not know exactly what
| http://mekk.waw.pl | they want and are unable to articulate all
|| that they kno
Alexander Wagner <[EMAIL PROTECTED]> writes:
> Marcin Kasperski schrieb:
>
> Hi!
>
> >> Hi Alexander, I added a "silence" checkbox.
> >
> > See my earlier post, this is really bad name.
>
> How'd you name it?
I'd set both thos
hose lines out, or issue "sought" from
time to time.
--
----------
| Marcin Kasperski | A process that is too complex will fail.
| ht
ntly observed, this could probably be of
> some interest for a later analysis.
He he, ...
http://mekk.waw.pl/mk/watchbot/index
(btw, watchbot PGNs fairly often face scid's 64kb limitation,
personally I compile scid patched to handle larger games)
--
---
ed by the opportunity.
--
------
| Marcin Kasperski | If Staff, Scope and Schedule are all fixed,
| http://mekk.waw.pl | managers will have no options, other than
||
moves as Scid...
No, this is just ideology. Jin don't get the moves because Jin author
considers current behaviour to be a FICS bug and refuses to workaround
it in Jin.
--
------
| Marcin Kasperski | Systems built by humans
to publish repository, let
somebody clone it, accept patches etc.
--
----------
| Marcin Kasperski | If Staff, Scope and Schedule are all fixed,
| http://mekk.waw.pl | managers will have no options, other than
||
t show which rook may castle.
It is nevertheless not important for the initial position (there we always
have one rook) and for the complete game replay (there we know which rook
has already moved).
--
------
| Marcin Kasperski |
there are more chances to get
people support while extending it.
--
------
| Marcin Kasperski | A complex system designed from scratch
| http://mekk.waw.pl | never works and cannot be patched to make
|
> Point is: currently scid really refuses every "illegal"
> move, simply cause scids input methods are bullet proof in
> the sense that they do not allow for illegal moves. "It can
> not be there what can not be there" ;)
This is both good and bad.
It is good, as it makes scid perfect tool to v
> > AFAIK current tendency is to use just 'FEN' (not X-FEN) for
> > chess960
>
> FEN can't express all of the castling scenarios in chess960,
> but can certainly handle a variety of starting positions.
> X-FEN is probably more important in terms of exporting
> positions later in the opening or midd
Dnia czwartek, 6 grudnia 2007 23:48, Todd A. Jacobs napisał:
> How well does scid currently support chess960--or other
> variations, for that matter? Specifically, I don't believe
> that scid currently supports X-FEN, and I vaguely remember
> having a lot of grief trying to import PGN games marked
ther possible obstacles?
--
------
| Marcin Kasperski | You have the right to produce quality work
| http://mekk.waw.pl |
> Can wine run cb-light?
I tried, but failed, it just crashes immediately
(in the same environment in which I run Fritz8 without problems)
--
--
| Marcin Kasperski | Systems built by humans are always subject
| h
ine (I tested Rybka)
works OK.
--
------
| Marcin Kasperski | If we are to be successful, we must still have
| http://mekk.waw.pl |the courage to put our faith in people as
|| opposed to a p
on the system.
Depending on the paths, after you do 'make install' you will either
overwrite that scid, or install your in separate location. In the
latter case you must be careful with your PATH settins to pick correct
scid...
--
-------
reason is ... lack of something (make, gcc, tcl
libraries, tcl development headers, ...). At least this was the reason
of all failures I've seen so far. Error message you get should make
it possible to guess what you lack.
--
-------
> > [Save and continue] [Discard changes] [Back to the game]
>
> But [Save and continue] is too dangerous to be default. Now
> it is far too easy to make some random changes when playing
> through the game and pressing Enter means old game is lost
> forever.
Discard is also dangerous, imagin
to
illustrate the point...)
--
------
| Marcin Kasperski | We want to know as early as possible whether
| http://mekk.waw.pl | the project will succeed. Thus we attack the
||high-r
nd continue] [Discard changes] [Back to the game]
--
----------
| Marcin Kasperski | I have a very small head and I had better learn
| http://mekk.waw.pl | to live with it and to respect my limitations,
|
> On windows, use Scid 3.6.19 without setup (just unzip anywhere).
> On Linux, compile and run anywhere. But the scid configuration files
> will be common between the 2 scids (not a big problem).
Also worth noting, that bases are binary compatible, so you may open
same scid base in new scid, modi
> > Some stupid heuristics like extracting capital letters or
> > taking first 2-3 letters could do.
>
> Dee, Tog ? ;)
Shr, Tog. Prefix Deep may be always safely removed ;-)
-
This SF.net email is sponsored by: Splunk Inc
> > Friendly google took me to:
> > http://www.muonics.com/FreeStuff/TkPNG/
> > Maybe it would be possible to take a look at it?
>
> I admit to be a bit conservative. But doing a quick package
> search on debian stable revealed: "nothing". (Same for
> testing and even unstable.)
Not particularl
Whenever I change some game and then try to leave it (say, press
Ctrl-Down to go to the next one) I get dialog box:
Scid: New Game
This game has been altered.
Do you really want to continue and
discard the changes made to it?
[Yes][No]
I hate this dialog.
First, i
>> a) Is it necessary to use .gif? .png has nice benefit of handling
>>alpha-blending (= far better looking piece edges)
>>
> "image create photo" only supports GIF and PPM/PGM (yes, it still
> exists).
Friendly google took me to:
http://www.muonics.com/FreeStuff/TkPNG/
Maybe it would
>> Exactly. While moving back, you have calculated and deepened
>> lines which actually happened in the game. While moving forward,
>> you have many lines which are no longer relevant.
>>
> You don't imagine the tricky stuff I had to do to make Scid analyse a
> whole game, recursively taking into
>* Probably one could shorten the engines names in the
> comments? The legend coud be given in the Annotators
> field. Something like
>
> [Annotator "DeepShredder 11 (A), Toga II 1.3.1 (B)"]
>
> and then later on
>
> A: +0.35, B: +0.27
+1 ;-)
But it would be even nice
>> - Analysis: It would be worth considering to run the
>>analysis of a game from then end to the start. This allows
>>the engine to reuse internal hash tables. Stefan told me
>>that at least within Shredder this improves the analysis
>>quite a bit (...)
>
> Currently engines don't
> - ./configure: it would be nice to implement --prefix
./configure BINDIR=/usr/local/bin SHAREDIR=/usr/local/share/scid
let you configure the same. Of course one can set those two variables
somehow when --prefix is given, but considering scid configure is
written in TCL, I am not volunteering ;-
> First of all I did not manage to subscribe for this list for
> whatever reason till now (was always refused or did not get
> any answers from the listrobot but now it seems to work) so
> I was not able to follow the discussion (still I've not
> found a way to access the archives )...
For about a
> Let's let Scid-pg take over the Scid project at SourceForge and end
> this silly feud.
Bah, as far as I know Pascal would like (and wanted) to do it.
SourceForge refused.
-
This SF.net email is sponsored by: Splunk Inc.
S
Well, those flamewars fortunately are more or less over,
at least I haven't seen them for quite a lot of time.
So - there is no reason to recover them.
And let us stop on it.
-
This SF.net email is sponsored by: Microsoft
D
> I’m curious as to what the differences are between version 3.5.x and
> 3.6.1 of SCID. Is it mostly bugfixes, or are there significant
> functional enhancements?
Well I do not know. But there are a LOT of important enhancements
between scid 3.6.1 (which is also ancient) and scid 3.6.18 (th
>> > PGN is "Portable Game Notation", a standard designed for the
>> > representation of chess game data using ASCII text files.
>>
>> Well, this is false for long, long years. Plenty of games annotated in
>> German, French, Russian, Polish, Hungarian,
>
> I can put a comment in french, in Asc
> I am not sure Scid uses UTF-8 : if I enter white player as "ùà" for
> example, it will be exported in PGN as 2 chars (not 4).
I suspect locale can be of some importance here. Try
export LANG='fr_FR.UTF-8'
or sth similar...
>> It even seems that utf-8 encoded pgn file is properly imported vi
> PGN is "Portable Game Notation", a standard designed for the
> representation of chess game data using ASCII text files.
Well, this is false for long, long years. Plenty of games annotated in
German, French, Russian, Polish, Hungarian,
> As far as I know Scid always uses UTF-8 encoding for PGN files. It
> is a bit broken on export, but that's probably Tcl fault, not
> Scid's.
Wow. Nice.
It even seems that utf-8 encoded pgn file is properly imported via
pgnscid and exported back via scidpgn.
The only thing which does not work
Well, maybe I open pandora box, but ... more and more often
I happen to meet PGN files containing non-ascii characters
in comments. Sometimes PGNs are UTF-8, sometimes use different
native encodings (I'm Polish so latin-2 is frequent, I read
Russian so ... PGN you can download from
http://online
>> No, no, no, it should be continue!
>>
>> This is just semantics of handling possible empty lines. In case of
>> continue, we just ignore empty lines and keep reading stdin. In case
>> of break, first empty line abort the process.
>> The process is over when stdin is finished, fgets returns fa
> I integrated your patch, and added your example and more
> details in the usage() function.
>
> But correct me if I am wrong, "continue" should be "break"
> here :
No, no, no, it should be continue!
This is just semantics of handling possible empty lines. In case
of continue, we just ignore em
The following patch (written against scid-3.6.19-rc2) adds new
way of running scmerge. One can now write
scmerge newbase -
and supply the list of merged bases on standard input (one by
one, separated by newline).
I wrote it after getting 'argument list too long' error once more
(not that
> Try to load ECO file from custom location (via Options menu),
> save current configuration and restart Scid.
>
> It doesn't work correctly here.
Ugh, it seems that we have incorrect order. The instruction
which sets default ecoFIle was moved and landed after options
file loading. So scid loads e
Dnia wtorek, 11 września 2007 22:46, Michal Rudolf napisał:
> I have a problem with 3.6.19rc1: ECO file isn't loaded
> properly on startup. It seems to be always falling to default,
> so if the file is in other location, it has to be loaded
> manually every time.
There was some ugly hardcoded defa
> * Best Games windows keep opening too, cluttering GUI
Mhm, I agree, this is somewhat irritating. I would prefer to open
tree window only, and just open best games manually when I want
it.
> * Tree window is not open immediately, so sometimes user
> doesn't know what is happening
This probabl
> - A Tree window can be opened for each base through the
> command tools-> open base as tree.
Cool and nice, thank you!
One possible improvement: history. Some history of recently
opened trees would be nice I am not sure about GUI, though.
One could also consider adding File/Open another
In the patch attached (made against 3.6.19rc1), a few extra small
makefile corrections:
1) ratings.ssp was not installed and uninstalled. Now it is
2) old-install target was removed (whatever it was, it was not
maintained)
3) Additional makefile targets were added
make all_scid
make ins
> The points 1. and 3. are done now ("open base as tree" and
> "stop refreshing the tree when board changes"). I will release
> a Scid 3.6.19rc1 sometime soon.
Nice :-)
> Concerning the "multiple Trees", I tend to think 2 Tree
> windows is enough (like engines windows). What do you think
> about
> always use Scid from the
> compilation dir, and maybe others do the same (I understand
> people that are reluctant to issue a make install as root).
If somebody on Linux is not wishing to install as root, I would
recommend sth like
./configure BINDIR=$HOME/bin SHAREDIR=$HOME/share/scid
or
> So, for people that don't issue a correct :
> ./configure BINDIR=$HOME/scid SHAREDIR=$HOME/scid
> (...)
> Scid will fallback to the "classical" directories, and this
> cannot hurt.
It should not be necessary, if somebody issues just
./configure
it is equivalent to
./configure BINDIR=/
I think that the *source* distribution would be noticeably
smaller, if it did not contain the following files (the list
below according to the scid-3.6.18.tar.bz2):
Makefile
Makefile.bak
engines/Phalanx-XXII/book.lrn
engines/Phalanx-XXII/position.lrn
engines/crafty-20.14/book.lrn
engines/cra
>
> There is already some cache for the tree window, and despite
> not being as fast as books, I find the Tree window faster in
> Scid than its equivalent in Chessbase.
I've never used chessbase, but in ChessAssistant the tree tab
refreshes instantly. There is some cons there - after changing
t
The attached patch (due to my laziness it is in two parts as I just maintan it
in this way, apply both in order)
changes scid so it complies to the typical Linux way of organising data files
(data files are separate from
the binary, all paths in use are set-able by ./configure switches) - but it
> > It would be nice to have sth like File/Open Tree or
> > File/Open as Tree, what would do just the same - open
> > selected database, open it's tree, and lock it - while
> > keeping the currently loaded game active.
>
> You are right, and also that seems pretty easy to do
> (...)
(keeping fin
Well, a few separate suggestions regarding the tree window.
(background) I am using scid to keep reference database(s) (split
by eco) and search for opening ideas
1. Open the tree easier.
Let's say you are viewing some game, and would like to take a
look at reference tree with respect to t
> This is normal as I put a check on the game size in file
> tkscid.cpp (line 7330) :
> (...)
> You should also remove this, and it *may* work.
Unfortunately it does not, although differently: while loading
the game, scid does not report any problems, but the game is
open incomplete, and when I
> As there is an overflow with a game, my guess is that the
> games before this one should be imported (at least I think so,
> if I remember well the code),
Yes, they were. In fact this is how I found troublesome game in
large file (the first after last imported)
> Concerning a game of 200 kB,
Note that I tried recompiling scid with Length variable type
changed from ushort to uint (this is how I understood the
suggestion in changelog, is it right?) and it did *not* help.
PS I won't send the game to the list as it belongs to the
commercial offering, but I think I can provide it direct
I just once more got scid failure while importing PGN file due to
the large game. This time it is reasonably mainstream case
- ultra083.pgn file from UltraCorr database
(http://www.chessmail.com/shopcart/catalog01.html, this is
according to many the best correspondence games
archive available).
> Scid by itself uses another null move notation (--). Let it
> be, but maybe PGN import could recognize Z0 and handle it?
I took a deeper look and it turned out that patching this is
fairly trivial. The patch (against scid 3.6.16-2) is attached.
I tested it and it seems to work without any probl
ChessAssistant uses Z0 (Z and zero) notation for null moves (also
in generated PGN files). I sometimes want to import to scid
something exported from CA and fairly frequently face
Error: Unexpected symbol: Z0
Scid by itself uses another null move notation (--). Let it be,
but maybe PGN
> > It would be nice to have some keyboard shortcut, which 'just
> > saves' the current game. By 'just saves' I mean what most
> > people usually mean while saving document, game, whatever:
> > if there were previous version, replace it, if not, save it
> > as sth new. So one can press this Ctrl-S
Well, minor annoyance, but...
It would be nice to have some keyboard shortcut, which 'just
saves' the current game. By 'just saves' I mean what most people
usually mean while saving document, game, whatever: if there
were previous version, replace it, if not, save it as sth new.
So one can pres
> Location of "tcl.h": not found
> Location of "tk.h": not found
> Location of Tcl 8.4 library: /usr/lib
> Location of Tk 8.4 library: /usr/lib
> Location of X11 library: /usr/lib
Seems you installed libraries, but did not install development
files. To compile scid yourse
SCID has functions to patch player/event names so they are shown
in uniform way, and to remove duplicate games. That's nice, they
are useful (well, in fact more usable than in some commercial
tools).
But, still, there is some room for improvement. It would be nice,
if scid could make some use
Dnia wtorek, 15 maja 2007 20:00, Stu Owen napisał:
> HI,
> I find if I enter a variation, it automatically deletes the
> rest of the main line and any other variations stemming from
> it without warning.
> This is with version 3.6.14 and happens consistently.
This is managed by the option Option/M
Stu Owen napisał(a):
> Hi,
> I'm trying to anaylse with crafty on Kubuntu 7.04. But I keep getting
> an error "The analysis engine terminated without an error". Crafty
> runs fine from the the command line. Its version is "Crafty v20.14 (1
> cpus)".
>
What about other engines? Do togaII work? O
Two separate issues.
1) I just installed 3.6.14. It seems sth changed in the way variations
are handled, and I do not like the change. Two problems
a) I enter long variation, then move back a couple of moves in it, make
another move and ... buum, silently whole long line gets thrown away.
Prev
93 matches
Mail list logo
