Hello,



You all know about nice features in Chessbase as they are regularry
announced on their homepage. One of them is the position explainer. Here is
a nice screenshot of how "usefull" this feature is:

http://www.playwitharena.com/common/...serklaerer.jpg (In German it
says, white has a constricted position...) hmm really ?


Also you surelly heard about the reliable cheating detecting software. We
have some screenshots of this software in action:

http://www.playwitharena.com/common\pic\fritz11.gif (who cheated here ?)
http://www.playwitharena.com\common\...truggling5.gif (take a look at
the score)
http://www.playwitharena.com\\common...truggling2.png here the
cheating detection software took away the points from white for ranking
manipulation, as the programm indicated an advantage of +23.16

This features are not aproblem of course, it is the support/Chessbase sysop
reaction if you tell them about the bugs. Remember you all pay for this crap
!

Michael Diosi
http://www.playwitharena.com

Michael Diosi wrote:
Also you surelly heard about the reliable cheating detecting
software. We have some screenshots of this software in action:

http://www.playwitharena.com/common\pic\fritz11.gif (who cheated here ?)
http://www.playwitharena.com\common\...truggling5.gif (take a look at
the score)
http://www.playwitharena.com\\common...truggling2.png here the
cheating detection software took away the points from white for ranking
manipulation, as the programm indicated an advantage of +23.16

Those URLs got mangled. The correct ones are

http://www.playwitharena.com/common/pic/fritz11.gif
http://www.playwitharena.com/common/...truggling5.gif
http://www.playwitharena.com/common/...truggling2.png

To be honest, I'm not sure what the first two are supposed to show.
It is impossible to say that anyone has cheated in the first one; has
Fritz asserted that somebody did? In the second one, why is it a
problem that the GUI quotes an evaluation of exact equality in a
position where the game is over? Sure, it would be nice if it blanked
that part of the display but it's hardly a serious problem.

The third one is bad, however -- the engine evaluating a position as
+23.16 when every legal move stalemates black. (By the way, the
knight on d5 looks redundant to me; does the problem still occur if
that piece is removed?)


Dave.

--
David Richerby Perforated Strange Tongs (TM): it's
www.chiark.greenend.org.uk/~davidr/ like a pair of tongs but it's totally
weird and full of holes!

Hello David,

Thank you for correcting the URL's. The point is that Chessbase and their
sysops say the software is RELIABLE when they kick people from the server
for what they call "cheating". But their software is not reliable. So I will
let the people know before buying such a software they should know that
ChessBase sysops will accuse them for cheating if they don't like their
face...


If you take away the knight everything is o.k.

Michael

David Richerby wrote:
Michael Diosi wrote:
Also you surelly heard about the reliable cheating detecting
software. We have some screenshots of this software in action:

[snip broken URLS]
cheating detection software took away the points from white for ranking
manipulation, as the programm indicated an advantage of +23.16

Those URLs got mangled. The correct ones are

http://www.playwitharena.com/common/pic/fritz11.gif
http://www.playwitharena.com/common/...truggling5.gif
http://www.playwitharena.com/common/...truggling2.png

To be honest, I'm not sure what the first two are supposed to show.
It is impossible to say that anyone has cheated in the first one; has
Fritz asserted that somebody did? In the second one, why is it a
problem that the GUI quotes an evaluation of exact equality in a
position where the game is over? Sure, it would be nice if it blanked
that part of the display but it's hardly a serious problem.

AFAIK Fritz and CB GUI software in general gives game terminal nodes in
checkmate an eval of 0.00/0. I don't know why. I would prefer +/- Inf.

The third one is bad, however -- the engine evaluating a position as
+23.16 when every legal move stalemates black. (By the way, the
knight on d5 looks redundant to me; does the problem still occur if
that piece is removed?)

I suspect it takes the extra piece to tip it over the edge. Strangely
weaker engines do better on this test by responding instantly.

Shredder10 in infinite analysis mode quickly got to 99/99 and locked up
requiring a system level kill process to regain control. Most engines
behaved OK in normal play. I don't have Fritz 11 to test.

The most annoying ChessBase bug for me is the massive slow down
encountered when the engine is doing a wide search of multiple lines in
some puzzles and the search ply depth starts to find mate in N. When the
ply depth exceeds 2N the rate of progress becomes glacial the program
spends massive amounts of time on already solved PVs.

CB insist there is no problem (what a surprise).

Regards,
Martin Brown
** Posted from http://www.teranews.com **

Michael Diosi wrote:
The point is that Chessbase and their sysops say the software is
RELIABLE when they kick people from the server for what they call
"cheating". But their software is not reliable.

Thanks for the clarification.


Dave.

--
David Richerby Surprise Devil Drink (TM): it's like
www.chiark.greenend.org.uk/~davidr/ a refreshing juice beverage that's
possessed by Satan but not like
you'd expect!