Crafty Strength, Development, and Questions for Bob

[FlavusSnow]

I don't know. I post this out of frustration because when I was little I remember a Crafty that destroyed all other engines. I remember a Crafty that was the Rybka of its time, with fanboys just the same. I now look at IPON (and other rating lists): Crafty ranked #22 at 2601 ELO. 400 ELO away from #1. I miss the Crafty of my childhood.

I know Crafty is still active, I mean, there is a release every now and then, but certainly not the 100's of ELO gain in the past year like we've seen with more than a few other engines. I realize that version 23.1 was a large jump in ELO but that was over a year ago and was probably the largest boost in ELO in the past several years.

So to get to it: Is there still a focus for significant ELO gains? I'm not talking about cut-paste from others. I'm talking about understanding how others are getting these huge gains and then implementing some of the 'new' ideas.

I don't know all the technical details, but if Crafty is 'maxed out' then what about starting a new engine? Obviously it wouldn't be starting from scratch, but maybe it would allow for more innovation to step outside the Crafty box a little.

[bob]

I don't know. I post this out of frustration because when I was little I remember a Crafty that destroyed all other engines. I remember a Crafty that was the Rybka of its time, with fanboys just the same. I now look at IPON (and other rating lists): Crafty ranked #22 at 2601 ELO. 400 ELO away from #1. I miss the Crafty of my childhood.

I know Crafty is still active, I mean, there is a release every now and then, but certainly not the 100's of ELO gain in the past year like we've seen with more than a few other engines. I realize that version 23.1 was a large jump in ELO but that was over a year ago and was probably the largest boost in ELO in the past several years.

So to get to it: Is there still a focus for significant ELO gains? I'm not talking about cut-paste from others. I'm talking about understanding how others are getting these huge gains and then implementing some of the 'new' ideas.

I don't know all the technical details, but if Crafty is 'maxed out' then what about starting a new engine? Obviously it wouldn't be starting from scratch, but maybe it would allow for more innovation to step outside the Crafty box a little.

Crafty is still being actively developed by a group of us. However, we are not "cut-n-paste" types and do not take every new idea in every open-source engine and stick it into Crafty. That is not a very satisfying way of developing a chess program. We are still improving. Will we catch those at the top? Who knows. We are still moving up. And are testing new ideas daily/weekly...

I don't see anything problematic about the overall design we have. Bitboards are as good as anything for developing a chess engine. We are certainly not too slow. The search is cleanly coded (now). There is still work to do, of course.