a book about Rodent (in French)

a kind soul going by the name of Rob Robinson, maintaining a web site "Echecs et informatique" as well as a page on free computer games, decided to write a book on Rodent IV. It is a user manual, better than I could ever prepare, supplemented by an attempt to classify Rodent personalities and to add one or two to the roster. The book can be preordered at

echecs-et-informatique.franceserv.fr/rodent-iv.html

So far, only the French version is being prepared, but there are some chances for a translation.

Nice.

He did a great job at presenting in detail the Rodent engine on his site. And, btw, it is one of the best sites about computerchess I've seen in a long time. Thanks, Pawel!

@matejst, You are right - Rob Robinson prepared his book very diligently, almost in scientific manner. I wish I had been able to prepare such high quality documentation myself. And his website is similarly good.

I played a few games against Rodent at 1600 and 1500 elos. The time controls were about 1h for 40 moves + a bonus. I had the impression it was several hundreds elos stronger than the declared strength -- perhaps more than 300 elos.

Let's say that it calculates variations too well. Almost no blunders. And its positional choices, even at 1500, are perhaps not first hand, but relatively solid. I managed to create some advantage, but it felt too resilient. OTOH, Rodent was quick to punish my own errors.

matejst wrote:

I played a few games against Rodent at 1600 and 1500 elos. The time controls were about 1h for 40 moves + a bonus.

Rodent weakening scheme has been tuned in blitz games, where low depth hurts a bit more, and scaled against CCRL ratings, which are probably too low in that rating range. Perhaps it makes sense to go a bit lower with rating on long time controls.

I found it virtually impossible to calibrate Elo across different time controls, or let alone across different personalities. For example, personalities that are clearly weaker for full strength Rodent, sometimes perform suprisingly well when weakened.

Thanks for the answer, Pawel. I continued playing and tried it at 1350 elos. Anyway, after the first surprise, I find it to be an excellent sparring partner: clever, but not too much, makes mistakes, but not too obvious. Even at 1350, his play is sound, at the level of a strong club player.

I will be able to give a more insightful review in a month, after 20, 30 games at different levels. I will also record the games, in case it is useful to you.