Original scenario design question

[NateTheDBA]

I guess this is morbid curiosity, but.. If large sprawling swaths of paths are bad (for both staff and guests), as I’ve heard, then WHY do many of the original game scenarios contain them? I’m thinking of parks like Roaring 20’s Alcatraz, the Great Wall of China, the giant American city-surrounded-park coaster challenge, the Rio Mardi-Gras carnival beach, etc. (some of these happen to be “no money” scenarios... and the latter two specifically have this “We need to mimic real paved city streets” thing going on that seems completely counter-productive.) =P

Perhaps a better question, a “tip or trick”, is, when playing such scenarios, what’s the best thing to do with those huge pathing areas? Just start deleting stuff until you have nice single-wide discrete paths, 2-wide boulevards, and/or grids?

[NateTheDBA]

Couple path-only screenshots for example.

[jensj12]

The scenario's you named are all from the expansion packs, which are known to be of poor general quality. Single-wide paths are always the best option. Junctions are what make pathfinding bad, and anything wider than 1 tile automatically inserts a junction every ~10 tiles.

[NateTheDBA]

Ah I see! Thanks for the reply. In this context, what is a "junction"? Like a normal crossing of 2 perpendicular pathways, a T or X? Or does it have some more specific meaning in RCT pathing? PS: I tried reading some stuff but I'm still not 100% sure what the "junction problem" really means.

Edited June 16, 2020 by NateTheDBA

[zaxcav]

The wiki page you refer to has a good description of the path finding under the section "Pathing Systems". It leaves out some details to keep the explanation simple.
Junctions are indeed intersections of paths, such as a T or X. Other things that are considered junctions are listed on the wiki page.

The main additional point to know is that the pathfinding searches through at most N junctions on a potential path before it stops. It chooses the direction of the various potential paths that ends closest to the destination. The value of N is around 10-15 if I remember correctly. Peeps that own a map have a slightly large value than those that do not, so should be a little better at finding their way around.

If it weren't for the path thinning algorithm that assists the path finding, every path tile on a wider path would be a junction and the path finding would reach the search limits very quickly indeed. This may be what you refer to as the "junction problem". The "junction problem" might also refer to various other undesirable outcomes of limiting the search in this way.

There is an option to view the result of the path thinning in game. When path finding, peeps always follow the thinned paths. RCT1 had no path thinning.

Probably the best way to help the path finding is to design paths that follow the same principles of a real road system. Build a network of main thoroughfares (like highways) with minimal junctions to get the peeps most of the way to their destination. Few rides or shops should be connected directly to these. Then fill the zones between the main thoroughfares with smaller roads and laneways to which the rides and shops are connected.

 

[NateTheDBA]

Thanks so much for the reply Zaxcav! Very informative. I wonder if anybody's posted like a sample "best practices" for path layout in a medium-large park.   Cheers all!