Forums

any1 ever take the Bratko-Kopec test?

Sort:
Validior

http://kopecchess.com/?page_id=55

 

Supposed to be an accurate ratings estimate based on chess knowledge etc. I am about to take it and see how it compares to my USCF rating.

It would be interesting if others took it and compared also

TheGreatOogieBoogie

Don't let the results go to your head, I got 2200 on it.  The last problem was refreshingly easy after a few really hard ones though. I still need to work on tactics apparently.  Still, I have a tendency to overprepare when the goal can be immediately accomplished.  I recognized the thematic Nd5 sac in the Sicilian so maybe I shouldn't even count the half point it gave. 

Validior

I gotta admit, I was a little dissapointed.

First off, the diagrams are absolutely tiny. No one in their right mind would use such a small diagram for serious chess work.

Secondly. half the time it shows you the diagram with white on bottom, yet its blacks turn to move! Again, 99% of us will be uncomfortable with such an arrangement

 

Some of the tactics were pretty esoteric, throwing a piece into a nest of pawns etc

 

I scored 9.08, cest la vie. Maybe could have done better with larger diagrams and more than 2 minutes per position.

 

for rating it tells me I am rated between 1500-1900. Wow, thanks Einstein, even a blind monkey could guess within a 400 point range.

 

BTW im currently 1745 USCF. Do you have an OTB rating?

 

Again, pretty dissapointed with the quality of the test. I know Kopec has a good reputation etc but it just seems asanine to try to solve positions based on tiny backwards diagrams

Validior

oh yeah, every position is some locked chain, closed position with almost all the pawns still on the board. I guess thats supposed to be "advanced." I suppose a KID type player would be at home with those diagrams but a kings gambit/scotch player wouldnt.

 

I purposely play so as to AVOID those megalocked positions so I was very disadvantaged trying to size them up in 2 minutes each

TheGreatOogieBoogie

Yeah there was a definite imbalance in the kind of positions shown.  I think the endgame ones were fair though but somewhat obvious, but then I was studying the endgame for the previous month so I should know the common themes. 

Totally agree on the diagrams being small and needing to show black to move from black's perspective.  When reading a "black to play and win" position I flip the board and study from that side after setting up the pieces.  Chess stuff really isn't noted for its excellent coding and even Chessbase stuff is very buggy.  This is even with great hardware too.

Yeah, should be given more time so as to exercise a thinking system, note imbalances, evaluate, then calculate candidates. 

I think people should get points for reasonable moves as long as they maintain a clear advantage not just the best one. 

davecom

You need to understand the history of the test to appreciate it.  It was created in academia in the early 1980s to evaluate the strength of computers based on human results and for a while became the de facto standard test for assessing machine strength. There has to be a time limit - without a time limit it's pretty pointless as an evaluation of rating. Even a beginner will eventually find the right move given days of analysis and the ability to try every combination. The 2 minute time limit in particular allows for a large number of positions in a reasonable amount of time so that the results are statistically significant. The fact that you ran out of time on some positions is a good thing - it means the positions were appropriately challenging for your ability.

The original academic paper and other pieces about the test are linked to from here:

http://chessprogramming.wikispaces.com/Bratko-Kopec+Test

I agree with you (on this particular implemention - there are many) about flipping the board for black and the size of the diagrams - but on the other hand it sounds pretty entitled to be so down on a free educational tool. There's a book that has both the test as well as many other more modern tests:

http://www.amazon.com/Test-Evaluate-Improve-Your-Chess/dp/1483991571/ref=sr_1_1?ie=UTF8&qid=1410713826&sr=8-1&keywords=test+evaluate+and+improve+your+chess

Ziryab

Have the solutions stood up to more recent analysis?

I'm in the 1%. I like White on bottom. I insist upon that in all published diagrams. I will concede, though, that in onlive solving exercises, it is helpful to flip the diagram when it is Black to move. Even so, I have turned this feature off in a iOS app that gives me the option of keeping White on bottom. 

davecom

I can tell you that many modern engines still utilize the test (and other tests).  Not necessarily anymore as a rating check (since they are beyond 2400 in rating) but more as a sanity check.  And it is still administered to humans on a regular basis too.

Jeromepassion

Bonjour a tous,

Je suis français.

J'aimerai avoir les 4 solutions proposées pour chacun des 24 diagrammes du test de kopec ainsi que lattribution des points fractionnes en fonction des reponses. 

1point Dd1

0,75 point xx

0,5 point xy

0,25 point xz

Jai seulement les 24 diagrammes mais pas toutes les solutions secondaires, moins bonne que le meilleur coup !!

. Mon java ne fonctionne pas sur l'ordi malgre tous les essais....

Pouvez vous svp me les indiqués ?

Merci pour votre aide.

A bientôt

 

testaaaaa

john bartholomew took such a test rated him around 2200 lol

vesna10

hey testaaaaa

testaaaaa

hi vesna

Jeromepassion

Hello