On Jan 21, 9:01 am, [EMAIL PROTECTED] wrote: > Hi all, > > It's great how many different sorts of solutions (or almost solutions) > this puzzle has generated. Speedwise, for reference my solution posted > above takes about 40 seconds on my 1.8GHz laptop, and the less elegant > version (on my webpage linked to in the original post) takes about 15 > seconds. It seems to me like surely this problem can be more > efficiently solved than that?
I haven't had the time to look at your solution yet. I wanted to have a go without being influenced. > Arnaud: I haven't had time to play with your solution yet - how quick > does it run? I can't run test from here (work) but last night it seemed of the order of a second on my 2GHz MacBook Pro (it was late and I was quite tired so I didn't have the energy to time anything...). It depends if you stop when you hit the first solution or you want to go through all of them. I guess it will run quicker if I don't build a string representation of each calculation. You should run a test on your own machine though to make comparisons meaningful. > My fantasy is that there is a solution that isn't TOO slow where you > can just look at the code and go 'Oh yes, of course that works!' and > understand it immediately. Maybe that's too much to ask even of > Python! ;-) It's a laudable goal. We might get there :) -- Arnaud -- http://mail.python.org/mailman/listinfo/python-list
