I don't understand why you're so confused about why the first alias is slower than the second. Matching against over a dozen wildcards versus matching a single wildcard anchored at the end of a string is obviously going to produce vastly different results.
Your logic comparing matching wildcards versus using tokens doesn't make sense. Using tokens the algorithm knows that there is a token separator and what it is. Using wildcards it knows no such thing, all it knows is that there are given characters separated by any number of characters (including that character), so it's not at all like providing tokens. I have no idea if mIRC's internal wildcard algorithm attempts to match * greedily or ungreedily, I'd guess it's greedy and maybe if that were changed it might improve your results somewhat. But then it might also make other wildcard matches far slower. Of course this is all pure speculation on my part.
It's not the fact that the first alias is slower than the second that 'confuses' me. What -surprises- me is the difference in speed.
How does it come that a wildcard match, even with 100 * wildcard symbols and 1000 chars long, takes 6 sec to go through 15 lines?
How many cpu cycles is that on todays average cpu's?
And indeed, I used $ctime instead of $ticks and missed that completely.
