Nope, that's alright. The difference is how the distinct part of circuit is enforced. Getting a difference in propagation is not that straightforward if only using assignments (try to impose Hall sets and you will see the difference).
Cheers Christian -- Christian Schulte, www.gecode.org/~schulte Professor of Computer Science, KTH, [email protected] Expert Researcher, SICS, [email protected] -----Original Message----- From: [email protected] [mailto:[email protected]] On Behalf Of Kish Shen Sent: Wednesday, August 26, 2015 08:30 PM To: [email protected] Subject: [gecode-users] circuit constraint: differences between ICL_DOM and ICL_VAL I am writing more unit tests for our Gecode interface, and for the circuit constraint (without costs), I would like to write tests for the supported propagation levels (ICL_DOM and ICL_VAL), and see some difference between the two levels, but I am unable to do so: I tried with circuits with up to 7 nodes, and then assigning a value to one of the nodes. For both ICL_DOM and ICL_VAL, and both seem to produce the same domains for the variables. Is this expected? If so, what do I need to do to get a difference between the two propagations? If the propagation is supposed to be different, this may indicate there is a problem in the interface and the constraint is not actually posted with the stated propagation levels... Thanks and cheers, Kish _______________________________________________ Gecode users mailing list [email protected] https://www.gecode.org/mailman/listinfo/gecode-users _______________________________________________ Gecode users mailing list [email protected] https://www.gecode.org/mailman/listinfo/gecode-users
