"Distance from entrance" discussion post

[Minim]

To explain my most recent contest post in detail, the "distance from entrance" is measured in pixels. It records a certain pixel on the landscape that the required number of lemmings is able to walk onto. Let's call that point Z. The coordinates of the entrance's trigger area can be given the letter W. Finally the "distance" is given the letter D. The formula is:

D = Z - W

Given that this is the first time I brought this up I might only record the X coordinate to simplify things. Here're the details from the current competition level: The X coordinate of the entrance's trigger area is 21. (W) Let's say that the player can only get 80 lemmings to walk as far as pixel 251 (Z) - the last pixel before the large wall. Putting the formula to effect, the number will be recorded as 230. (D)

I hope my explanation is clear enough. This criterion probably won't be necessary on those short levels, but on the long ones like the one we have today I thought this criteria would be better than "saving 1 or 2 lemmings". If it's agreeable enough then we can discuss about adding to the formula by putting in the Y-coordinates and recording distance away from exit, but just remember, it only applies to solving contests, and also to players who fail to solve it.

[Proxima]

This is an interesting idea, but firstly, it's unlikely to be necessary to rank non-solving players, as they will always rank below anyone who does solve the level. Secondly, because levels can have very complex terrain, it's an extremely complex task to get as close as possible to an exit in absolute distance, and time-consuming for the contest organiser to check what was the closest distance reached by enough lemmings. If the solver has time to optimise this metric, they would probably be better off putting that time into looking harder for a solution

Finally, this metric seems unfair because absolute distance can have very little correlation with difficulty of reaching a certain distance (remember "The Far Side"?) I think a better metric for ranking non-solutions would be (1) number of lemmings saved; (2) whatever the criteria for the current contest. Note that number of lemmings would always rank ahead of specific criteria, even if there are some criteria that rank ahead of it for solving replays. The reason is that saving a single lemming while maximising (or minimising) the contest's specific criteria is further from being a valid solution than saving one or two below the save requirement.