I would have to agree that a tile based game is not the same creature as a room based game. Furthermore, I think that we (the community, or even the people on the thread), are pretty aware of the differences between such creatures. Tiles are not containers usually. They are merely coordinates, and usually in a tile based game everything is linked more by coordinates than which room they are in. Each "place", "object", "creature", "whatever" has it's own coordinates variables (x,y, [and possibly z if you have depth or something]). Someone above mentioned that tiles represent an exact space whereas "rooms" can represent inches, feet, or miles. I feel like this is a pretty good description of their differences.

FWIW, KaVir's use of 'tile' (meaning a collection of smaller units) is somewhat non-standard and specific to his game; a tile – as I nearly always hear it used by people other than him – is nothing more than a place that other things occupy. As he said this is MudBytes, so it's not GW2Bytes. :wink: So, I think it's a little unfair to blame all of this on some bizarre usage from the RL community when in fact the RL definition (as has been presented here) is the standard definition (see many RTS games, for example), as opposed to something one developer happened to choose for his game.

MUDs in general don't tend to think about tiles, because tiles usually imply coordinates and most MUDs are room-based, not coordinate-based. I think it is coordinates, not "containerness", that distinguishes tiles from rooms. (Indeed, a tile can be taken to be the container of the things on its coordinates; you could very easily imagine things being actually implemented this way.) A tile is assumed to be one block that has a very clear spatial relation to other blocks around it. If you had a room system where each room was taken to be of equal size, and furthermore Euclidean geometry were enforced on exits, then you would have a case of rooms being essentially identical to tiles.

By the way, if you don't believe me regarding the common definition of tiles, just read the Wikipedia article on tile-based games. By that understanding, Tavish's usage of tile is entirely normal. The only definition that comes close to KaVir's is perhaps the notion of a tileset, but that's more a collection of sub-images (see example on Wikipedia article) than something you paint the world with.

FWIW, my usage of the term 'tile' (meaning a collection of multiple distinct units) has been in use in the mud community since the 90s, isn't unique to my muds, and was first used in room-based muds. It's also consistent with many other tile-based games such as Dominos, Blokus, Gheos, Tantrix, etc, etc, etc, although in my case it drew more inspiration from games like:

Sorcerer's Cave: "The players start by choosing several generic character types, such as Man, Woman, Dwarf or Giant, differentiated by their Strength and Magic abilities to join their party based on a points system. Once selected, a marker is placed on the central "Gateway" tile. The party then explores the cave by drawing cave tiles."

And:

Space Hulk: "The game is set on a board made up of various corridor and room tiles which can be freely arranged and locked together like a jigsaw puzzle to represent the interior of derelict space ships."

It's also common in RTS games for tiles to include multiple terrain types, such as edges of coast or forest, or internal features like bushes or ponds. Conversely, in a mud, a "room" usually consists of just one terrain type (called a "sector" in Diku muds) - thus a tile system in a room-based mud will typically involve a tile respresenting multiple rooms.

Tile-based is a graphical technique of displaying space. Even in Rogue-likes it's not at all necessary, as many still use and prefer ascii character maps.
The common mud directed graph (or often multi-graph) where edges are exits and vertices are rooms could well be displayed as graphical tiles.
But using tiles doesn't say much about the underlying data structures of a game.

KaVir, those are board games, not computer games… I just think that if you're going to admonish somebody for using terminology from another community, you should be pretty sure that the terminology you're using is in fact very standard in this community.

RTS games do use multiple terrain graphics on a single tile, however that doesn't make them similar to yours; you cannot occupy each sub-section of terrain on the tile because the single tile is still an atomic container for units, etc.
Well, if you have other examples of all these MUDs that take your definition as standard vs. computer games at large, I'd be happy to hear them out of curiosity. Again, my point here was that I thought it was unfair to criticize somebody for using a definition that was entirely reasonable, consistent with the computer game industry at large, but just happened to disagree with how you conceive of tiles in your game (and some board games). I don't really care if your definition is 'right' or 'wrong' (it's sensible too), I just wanted to clear up what was an unnecessarily rejecting remark to somebody from another community.

A fair point, but the technique itself (and even the reaso... is much the same as I've seen used in muds. In fact the approaches I've seen used ASCII graphics for representing the tiles anyway, so I think it's pretty much the same concept.


And thus my confusion when it was suggested that tiles and rooms are the same thing.

What is "much the same" here? A tile set, as given in that link, is rather different from your GW2 tiles. It is a collection of sub-images that are meant to be displayed separately. It is not something that you put down on the world as a whole. In other words, it is not an atomic unit; it is a collection of atomic units. Even the little picture they give in that link (which I provided earlier) shows that it's a collection of various sprites. :thinking:

Perhaps the question is academic at this point, but I'm still curious about this.


They can be, given the constraints I stated earlier.

The funny thing is that I'm not even from the RL community, I've been a MUD dev for 10 years now. I've read tons of game design information from across several spectrums of the gaming industry over the years. By and large the concept of a tile or tile space was fairly common to represent the location in the world. Until this thread I don't really recall tiles being really associated with MUD discussions in any meaningful way that would suggest a different definition when relating to the genre.

Mark it down as a difference of semantics and replace my original use of tile with grid location or area coordinates in the tile = room equation.

No biggy - just I'm pretty interested in movement systems and world representations, which is why I jumped at the prospect of a decent discussion about alternative designs.

David, since you're pressing the point of it being somehow obscure: The term 'tiles' has been used in that context with RPGs since the 1980, and refers to the actual tile drawn for a single segment of terrain. The idea of 'tilesets', in that context (still used in 2d games to this day, and some 3d ones), comes from the idea of a collection of terrain tiles.

Wikipedia has a more indepth look on the subject: http://en.wikipedia.org/wiki/Tile-based_...

Boardgames like BattleTech, that predate computer games, also used the idea of tile-based game design. It's a concept that's been around for a while, and I don't really think it's that obscure knowledge, as you seem to make it sound to be.

Maya/Rudha

Rudha, you're giving the same usage of 'tile' that I did. You also gave the exact same Wikipedia link I gave… :smile: I'm not sure what I said prompted you to write that post?

I laid tile in my bathrooms. Bathrooms predate board games. Do I need to link wikipedia for indoor plumbing?

Can you clarify this statement?

I'm afraid I am going to have to ask you to do this. :wink:

Sure. When I was working on my mud that had handwritten areas, building those areas myself always felt more like a chore that had to be done than something that I actually wanted to do. Sure there were times that I would have some inspiration and enjoyed the process, but for the most part it annoyed me. Working with builders was even a bigger PITA. I'm sure that was just mainly the builders I've worked with, but if I recall correctly I ended up with 2 usable areas produced out of nearly a dozen different builders. The hassle vs production wasn't really worth it and I'm happy right now staying away from going back down that road.

I've found over the years that builders are much more likely to have a strong work ethic when its a project already live. A friend of mine has a builder port and a system for nonstaff to work and that's where he gets a majority of content.

Also, an in game currency for helping in development, testing, player-moding, reporting bugs can incentivize. Of course in this case you must come up with a fair system to players not earning. Ie not direct benefits to the prime directive of your game. In the past I've used item restrings and cosmetic additions as well as the ability to start global quests and events.

I always had that problem as well, which was one of the main reasons why I took such an interest in generated descriptions. However I've had some success with mixing the two approaches, adding a touch of hand-written content here and there to make the areas feel a bit less generic.

When you talk about "creating a roguelike world and bringing it to MUDs", are you talking about a full roguelike interface? Or some sort of side-by-side description/map hybrid?

The interface will be very similar to roguelikes. The game concept is also very similar in that if you are in a dungeon admiring the view instead of searching for a stratgeic place to defend yourself then chances are you will be dead shortly.

How will you deal with real-time position information when sending over a telnet connection?

I just wouldn't redisplay unless a player requests it. Instead I'd display that monsterx is approaching and let them decide if they need to repeek.