@sarciness actually a typical tablebase will not tell you whether help mates are possible or not as a TB typically only looks at best continuations, not worst. Though such an alternate tablebases could easily be built with a very similar algorithm, someone would have to put in the computing power though. Not sure it's worth it though, up to 5 pieces I think general rules will cover basically all cases correctly. (I mean, even for more pieces they will almost always be correct, but there you could construct hypothetical positions where they fail)
A 6 piece help mate tablebase is rather doable but probably not very useful yet either. And 7 men hmtb will be too expensive to generate. (unless someone has a spare computer with one tera byte of RAM; but that would be pretty much a waste of computing power)
You could look at existing files and see if mates with a material configuration exist. But the problem there is you don't know if they are reachable then, e.g. something like White pawns a4 c4 f4 h4, Black pawns a5 c5 f5 h5, kings on their own side. Clearly drawn but the material distribution has winning positions of course.
A 6 piece help mate tablebase is rather doable but probably not very useful yet either. And 7 men hmtb will be too expensive to generate. (unless someone has a spare computer with one tera byte of RAM; but that would be pretty much a waste of computing power)
You could look at existing files and see if mates with a material configuration exist. But the problem there is you don't know if they are reachable then, e.g. something like White pawns a4 c4 f4 h4, Black pawns a5 c5 f5 h5, kings on their own side. Clearly drawn but the material distribution has winning positions of course.
This topic has been archived and can no longer be replied to.