Chen Shuang

I don't have a 356i Play. So I'm not able to reproduce the bug right now.

There are amounts of factors that might affect the performance degradation when use multi-threading. Most of them depends on CPU archtechture, e.g. cache miss rate, branch prediction miss rate, cpu...

Would you mind providing your WCA ID? In addition, this may be caused by cstimer+ (not maintained by me). It is recommended to disable cstimer plus and try again.

csTimer's server has only received less than 5k solves. As for why other solves were not uploaded, we are not sure whether it is related to csTimer+.

Are there any specific screenshots?

I'd love to add these features, is there any more documentation?

Well, actually, the next scramble is generated in 1~2 frame(s) from the prev solve.

Implememted by an optimal solver in

The algorithm is not designed for finding the optimal solution. More specifically, it uses many tricks to improve solving efficiency for random state. And due to these tricks, the optimal...

I think the best solution to this bug now is to write/use the TurnBuilder because I haven't found a reasonable way to process move canceling between stages.