I will start by introducing myself - I'm also known as Valdair on Reddit, and have been part of the RCT community there for a few years. I've been actively involved in the online RCT community ever since a little website called RCTcompetition back in 2002 and migrated to NewElement in 2004. I have extensive experience with Hyper Rails, Scream Machines, Sim Theme Park, RCT1, 2, and 3, Ultimate Ride, and NoLimits 1 and 2, including experience with Newton2, FVD++, and Elementary. Just so you know my background and the lens through which I am thinking about how to make Planet Coaster the absolute best theme park simulator that has ever existed (which it is poised to be, if the pace of the alpha has shown us anything).
My thinking is this: Yes, the coaster builder is incomplete and not even technically official yet. But we know that it is a node-based system with some basic pitch/roll/yaw control, just like RCTW and basically just like Hyper Rails from back in 2002. While this is serviceable, a game that prides itself on spectacle (after all, a big part of the draw of 3D theme park simulators is riding your rides, and if they are jerky and painful because of the constraints of the builder, it's just not as fun as something that's really polished and feels real, a la really good NL1 work or most any NL2 work). The best way to accomplish this is, I think, two-fold:
- Leave the node-based system in the game. It's intuitive. Perhaps some basic grid snapping in all three dimensions so that constant-radius turns and pitch changes are easy for pre-lift sections, lift hills, and things like that. Easy straightening tools for complicated brake+transport sections as well as a variety of catwalk toggles (left/right/both/none). Allow powerful spline smoothing so that inexperienced players are comfortable to slap some nodes down, place some pre-built inversions, smooth it out and open to the peeps.
- More complex FVD and geomtry-based building. This should never be the only method of building in a game for the general sim-playing public - this is part of why NoLimits is viewed as so complex and esoteric, but the tools are undeniably powerful. The best part is, the code to do all of this already exists. Newton2 translated FVD into NoLimits 1's old node-based system and FVD++ does the same for NoLimits 2's new system. The game should already know what the friction value and associated heartline value are for a given ride so these can be built right into the FVD engine. Geometric building allows you to do low-speed pre-lift and end-of-ride sections to join brake sections, as well as launch and lift transitions, while the FVD allows you to execute buttery smooth inversions and other complex track. Planet Coaster has a history (perhaps brief, but getting longer) of being the theme park sim of the current generation that continuously rewards you for dedicating more time to it. This system would be incredibly powerful for those looking to do recreations of real rides or even who are simply passionate about real-world roller coasters and who want to make their work emulate that with finer control.
As an additional thought, the support engine is a bit messy at the moment. There's surely a lot of work planned to be done yet, but my thought here is that you might include a toggle and a new type of scenery category that is ride-specific: toggle the pre-built supports on or off, and once they are off, allow the placement of an array of supports that are appropriate for that coaster type. They can snap to the track and have track connectors follow along to link the two, oriented however the track is, like this (Mantis at Cedar Point). This would probably not be very useful for wooden coasters, but the woefully undersupported loops and inversions in the screenshots and videos I've been seeing are very eye-catching.
I have two more questions/suggestions. Obviously custom scenery/user-generated content will be a big part of this game's life. In RCT2, custom objects are saved and exported with the save file itself to allow easy sharing and viewing of one another's parks. RCT3 suffered from a complete lack of this, and it has all but killed the "competitive" creative community surrounding the game. Despite the absolutely enormous array of CSO out there for the game, I can't view my friend's park if he used a single object that I don't have without hunting it down and giving that to me separately. I really hope it will not be difficult to implement this. If you go through a hosted website, perhaps it wouldn't be too difficult to have the used packs link from the uploaded park and be downloaded alongside it if anyone downloads it. This could also be done through the Steam Workshop, or simply saved within the game, or there could be a special project format that could be output, like a package that Planet Coaster knows how to open which contains all of the necessary files (this is exactly what NL2 does). A lot of us in the creative community are incredibly excited about Planet Coaster, as it promises to be the next step forward for everything we love about these games. Including this feature will ensure it fits right into the way we already structure our sharing of content.
Finally, paths. People have railed against the snap system, but I know that will be fixed so I'm not so concerned about it. The bigger issue is the ideas behind how they work in these modern titles - a snake of texture that links one node to another. Real path is just the ground, extending from ride to ride, shop to planter, regardless of their layout. See:
Ironically, the older grid-based system actually accomplishes this better than the systems that have been devised for PC and RCTW. Such as:
(this is my own work, I'm not taking it from anyone)
People will want to be able to push your game's engine to be the absolute best representation of reality there is. People are already running into issues with queues looking wrong because you have this snake of path with grass in the middle where real queues in real parks are generally just concrete/rock with true switchbacks, or are indoors in which case it's just whatever the floor is. I'm not sure of the best way around this. The historical strategy has been for the creative community to just make invisible path and then use scenery to get around the problem, but I would really love to see Frontier recognize this and tackle it to give us an intuitive system that feels perfectly real when we walk through it. I'm confident this can be done.
Sorry for the essay everyone. I hope what I've said makes sense.
My thinking is this: Yes, the coaster builder is incomplete and not even technically official yet. But we know that it is a node-based system with some basic pitch/roll/yaw control, just like RCTW and basically just like Hyper Rails from back in 2002. While this is serviceable, a game that prides itself on spectacle (after all, a big part of the draw of 3D theme park simulators is riding your rides, and if they are jerky and painful because of the constraints of the builder, it's just not as fun as something that's really polished and feels real, a la really good NL1 work or most any NL2 work). The best way to accomplish this is, I think, two-fold:
- Leave the node-based system in the game. It's intuitive. Perhaps some basic grid snapping in all three dimensions so that constant-radius turns and pitch changes are easy for pre-lift sections, lift hills, and things like that. Easy straightening tools for complicated brake+transport sections as well as a variety of catwalk toggles (left/right/both/none). Allow powerful spline smoothing so that inexperienced players are comfortable to slap some nodes down, place some pre-built inversions, smooth it out and open to the peeps.
- More complex FVD and geomtry-based building. This should never be the only method of building in a game for the general sim-playing public - this is part of why NoLimits is viewed as so complex and esoteric, but the tools are undeniably powerful. The best part is, the code to do all of this already exists. Newton2 translated FVD into NoLimits 1's old node-based system and FVD++ does the same for NoLimits 2's new system. The game should already know what the friction value and associated heartline value are for a given ride so these can be built right into the FVD engine. Geometric building allows you to do low-speed pre-lift and end-of-ride sections to join brake sections, as well as launch and lift transitions, while the FVD allows you to execute buttery smooth inversions and other complex track. Planet Coaster has a history (perhaps brief, but getting longer) of being the theme park sim of the current generation that continuously rewards you for dedicating more time to it. This system would be incredibly powerful for those looking to do recreations of real rides or even who are simply passionate about real-world roller coasters and who want to make their work emulate that with finer control.
As an additional thought, the support engine is a bit messy at the moment. There's surely a lot of work planned to be done yet, but my thought here is that you might include a toggle and a new type of scenery category that is ride-specific: toggle the pre-built supports on or off, and once they are off, allow the placement of an array of supports that are appropriate for that coaster type. They can snap to the track and have track connectors follow along to link the two, oriented however the track is, like this (Mantis at Cedar Point). This would probably not be very useful for wooden coasters, but the woefully undersupported loops and inversions in the screenshots and videos I've been seeing are very eye-catching.
I have two more questions/suggestions. Obviously custom scenery/user-generated content will be a big part of this game's life. In RCT2, custom objects are saved and exported with the save file itself to allow easy sharing and viewing of one another's parks. RCT3 suffered from a complete lack of this, and it has all but killed the "competitive" creative community surrounding the game. Despite the absolutely enormous array of CSO out there for the game, I can't view my friend's park if he used a single object that I don't have without hunting it down and giving that to me separately. I really hope it will not be difficult to implement this. If you go through a hosted website, perhaps it wouldn't be too difficult to have the used packs link from the uploaded park and be downloaded alongside it if anyone downloads it. This could also be done through the Steam Workshop, or simply saved within the game, or there could be a special project format that could be output, like a package that Planet Coaster knows how to open which contains all of the necessary files (this is exactly what NL2 does). A lot of us in the creative community are incredibly excited about Planet Coaster, as it promises to be the next step forward for everything we love about these games. Including this feature will ensure it fits right into the way we already structure our sharing of content.
Finally, paths. People have railed against the snap system, but I know that will be fixed so I'm not so concerned about it. The bigger issue is the ideas behind how they work in these modern titles - a snake of texture that links one node to another. Real path is just the ground, extending from ride to ride, shop to planter, regardless of their layout. See:
Ironically, the older grid-based system actually accomplishes this better than the systems that have been devised for PC and RCTW. Such as:
(this is my own work, I'm not taking it from anyone)
People will want to be able to push your game's engine to be the absolute best representation of reality there is. People are already running into issues with queues looking wrong because you have this snake of path with grass in the middle where real queues in real parks are generally just concrete/rock with true switchbacks, or are indoors in which case it's just whatever the floor is. I'm not sure of the best way around this. The historical strategy has been for the creative community to just make invisible path and then use scenery to get around the problem, but I would really love to see Frontier recognize this and tackle it to give us an intuitive system that feels perfectly real when we walk through it. I'm confident this can be done.
Sorry for the essay everyone. I hope what I've said makes sense.
