Difficulty

[demagogue]

It works in situations when one of your goals is to make a replayable map, and player foreknowledge would gut most the mission on replay, like the combination of a safe or precise location of the objective item when most of the mission is working to figure out the combination or location.

 

This same idea applies to "gamey" levels, like that idea to remake Clue, or Komag's MindMaster, one of the more "artistic" randomization mechanics in recent memory. (You might actually be disappointed to know that a lot of what you did in that blueroom could now be handled by NV's randomization script and a few other custom scripts.)

 

As for just generic loot ... really depends on what your goal with the map is. If you want to design it around a lot of replayability, like a timed maze like Run Thief Run where you have to get X loot in time, or a more puzzle oriented maze, those are situations when it makes sense.

Edited December 1, 2007 by demagogue

[smithpd]

As a player, I find the idea of random loot of limited usefulness. As sparhawk has said, lack of repeatability can cause problems and frustration. Examples include: ghosting, where one player may find success and another not, for mechanical reasons only, not skill; obtaining advice from other players; playing in a collaborative mode, where two or more people are comparing notes; speed runs, where you need to develop an optimum route. I also agree that mission design is what causes good replayability.

 

That said, I agree that a random number generator, rand(), is a good thing to have. But I would make it something that has to be implemented in a script rather than having "random loot" a standard feature of the tools. I would not make it so easy. Designers should know that if they use this feature certain game play elements as described above will be destroyed.

 

This thread has gotten way off topic. The original post, about graded awareness and alert status, was very good. I'd definitely like to see some adoption of that.