Computing these has been an occasional fascination of mine ever since I wrote my first program in assembly language on a 2MHz 8080 (8-bit microprocessor) in 1978. That program, using a classic recursive backtracking algorithm, eventually found one solution after running for about a week.
The problem is interesting because there are so many possibilities to test and the many blind alleys which lead almost to a solution are so long and time-consuming. (A solution is basically a 64 digit base 8 number). I find it fascinating that this problem, which is so extremely difficult for a person (or even a straightforward computer program) to solve, turns out to have so many solutions. Using my best program in 1992 on 486-class hardware I easily found tens of thousands of cyclic solutions and 1.2 million non-cyclic solutions. Writing a good program is a fun exercise.
Notice that the solutions (at least the ones I find) all have a somewhat similar appearance but like the proverbial snowflake, each is unique.