At an earlier point this was possible; should be easy enough to parameterize. The idea is that when we call permutations on qf (and friends, pf, pf2, pf3), perhaps we only want individual fields, or pairs of fields or triplets of fields. We might not want all fields. This would allow measurement of "the best we can do if we only use 2, 3, 4 or n fields".
At an earlier point this was possible; should be easy enough to parameterize. The idea is that when we call permutations on qf (and friends, pf, pf2, pf3), perhaps we only want individual fields, or pairs of fields or triplets of fields. We might not want all fields. This would allow measurement of "the best we can do if we only use 2, 3, 4 or n fields".