Yes and yes.
Possible Future Direction (personal view)
- Thread starter Koshling
- Start date
Yes and yes.
primem0ver
Emperor
Correct me if I am wrong but I think the main reason for using tiles on a grid as opposed to grid less movement in a game like this is so that a graph can be easily defined and navigated (in path finding).
Using a grid in civ4 makes it easy to tell where routes exist, landscape types are, features are etc... which are all taken into account in pathfinding. That way the most efficient route to any one position can be found.
RTS pathfinding is particularly bad in my experience because of the lack of a grid. When one gives orders to where they want units to go (without defining a specific path), they will take the shortest spatial route, regardless of hindrances, even when those hindrances are obvious such as a huge body of water.
I also want to point this out... We don't have to use tiles per se. We could also use points. The points on a ISEA grid actually don't make up hexagons and pentagons. They make up triangles.
Using points rather than tiles would have distinct advantages. For one thing they would get rid of the need to put rivers on tile edges. Technically they would still be on the edge of a triangle, but that edge would no longer prevent a unit from traveling the river since travel is done along edges as well. We could define textures and features on triangles rather than more complex shapes. This would allow edges themselves (lines between points) to be used as paths and the be the boundaries between terrains, features, etc. Points would also serve as bonus locations.
As far as the AI is concerned, which side do we use for pathfinding? That depends on the purpose. We can assume that humans are smart enough to take the woods if they want to use stealth; or the plains if they want to move quickly. We would simply use the value that matches the AI purpose (mission).
Using a grid in civ4 makes it easy to tell where routes exist, landscape types are, features are etc... which are all taken into account in pathfinding. That way the most efficient route to any one position can be found.
RTS pathfinding is particularly bad in my experience because of the lack of a grid. When one gives orders to where they want units to go (without defining a specific path), they will take the shortest spatial route, regardless of hindrances, even when those hindrances are obvious such as a huge body of water.
I also want to point this out... We don't have to use tiles per se. We could also use points. The points on a ISEA grid actually don't make up hexagons and pentagons. They make up triangles.
Using points rather than tiles would have distinct advantages. For one thing they would get rid of the need to put rivers on tile edges. Technically they would still be on the edge of a triangle, but that edge would no longer prevent a unit from traveling the river since travel is done along edges as well. We could define textures and features on triangles rather than more complex shapes. This would allow edges themselves (lines between points) to be used as paths and the be the boundaries between terrains, features, etc. Points would also serve as bonus locations.
As far as the AI is concerned, which side do we use for pathfinding? That depends on the purpose. We can assume that humans are smart enough to take the woods if they want to use stealth; or the plains if they want to move quickly. We would simply use the value that matches the AI purpose (mission).
primem0ver
Emperor
In addition to the advantages mentioned above, their would be HUGE asset advantages.
Instead of coming up with 16 + separate graphics for each route type, all we would need is one main graphic and 4 connector graphics (8 if we take the angles of pentagons into account).
The same logic could be applied to forest graphics, beach graphics, river graphics, etc.. Instead of coming up with a necessary number of graphics in order to accommodate connection needs, we could focus on creating a variety of graphics to add realism to the look.
Instead of coming up with 16 + separate graphics for each route type, all we would need is one main graphic and 4 connector graphics (8 if we take the angles of pentagons into account).
The same logic could be applied to forest graphics, beach graphics, river graphics, etc.. Instead of coming up with a necessary number of graphics in order to accommodate connection needs, we could focus on creating a variety of graphics to add realism to the look.
Hydromancerx
C2C Modder
Yeah Zoo Tycoon 2 uses triangles for placing objects. You can still have linear roads as well as hexagons.
Also you could have units, terrain features and improvements that take up different size tiles.
Attachments
Yeah Zoo Tycoon 2 uses triangles for placing objects. You can still have linear roads as well as hexagons.
Also you could have units, terrain features and improvements that take up different size tiles.
I don't particularly like the Zoo Tycoon isosceles right triangles because they don't tesselate onto a globe - micro-equilateral triangles are the way to go. Not only is pathfinding easier and smarter because the playing field is ultimately quantized, the player need never even know that the tiles exist because they can be only a few pixels in size or smaller. These would also go a long way towards providing a "snap-to" functionality.
primem0ver
Emperor
Don't forget that Zoo Tycoon doesn't have world maps. The isosceles right triangles worked perfectly for what they needed. We need something different so naturally we aren't going to use the same kind of triangles.
Also don't forget that the triangles aren't the main point. They are a convenient side effect of focusing on points rather than tiles as I explain above. Triangle tiles would be annoying for a number of reasons, the most obvious being the number of surrounding tiles. But points as a focus of data and movement is advantageous for many reasons as mentioned in my previous post. Each point on an ISEA grid is surrounded by 6 other points except on the 20 starting points which have only 5 points.
What is the difference between using hexagons (and some pentagons) and claiming to use the center points of those plots?
Nothing. Underlying data should always be points, with the tiling schema implicitly providing quantization. Whether the UI then explicitly displays the timing is essentially a UI issue then, though if it does it will need to understand the quantization scheme being used.
There is a difference - the points at the vertices of hexagons and pentagons will not be used if we just use hex and pentagon tiles. And given that we're trying to move away from tiles anyway, we should remove from end users any hint that there is quanitzation at all (with the exception of snap-to functionality as necessary).
primem0ver
Emperor
I am not sure you understood what I am trying to point out. I am not saying anything about using the center of those plots. Basically I am suggesting we get rid of traditional plots (at least in terms of positioning) and use points instead. Centers are irrelevant with points because there is no "center" point.
I am saying we should do away with plots and use points instead. It makes a BIG difference in terms of drawing choices and options of placement. I have already pointed out some of these differences in several posts above.
More locations to chose from: (a step in the RTS direction)
First, a pentagon is made of 6 points (5 points around the side and the center); a hexagon 7. The latter gives us 7 options of placement as opposed to 1 (Though obviously that point will also belong to another polygon).
Graphics for landscape features are simplified drastically
Currently, any tree feature in the game requires 15 layout tiles to cover all the different layout possibilities (four different quarters to cover and all the different possible combinations). These different layouts would no longer be required. There would only be one necessary layout in the shape of a basic triangle (which comprises one fifth of a pentagon and one sixth of the hexagon).
Graphics for routes are simplified almost as drastically
If we decided to keep routes bound to a methodological layout, we would now place them on edges between two points. This means that theoretically speaking, only one graphic would be necessary for routes as well that would be placed on the edges between two points.
However, this would gives us very mechanical looking honeycombed routes. To ease that, we would probably want to make one edge graphic, and 2 connection graphics to get rid of sharp corners (one double ended connector and one triple connector). Either way, both the algorithm for drawing roads and the need for graphics is drastically simplified.
Rivers on plots are now irrelevant. They can be navigable
Several people have expressed how nice it would be to be able to have ship units navigate rivers and complained about them being on edges. Some have suggested we have rivers that go through plots to create this possibility. Since rivers now run along edges between points, the distinction is unnecessary (and graphics are made much easier).
In addition, river deltas can now be expanded to be larger or smaller as well as more complex (with braided stream being a texture/feature that can cover a triangle).
Broader range for movement values
Since we are now dealing with distances that amount to half of a plot size, we now can chose slower movement speeds (half a "plot") as well has have a larger range of movement speeds to chose between.
OVERALL GRAPHICAL IMPLICATIONS: In the end, the number of graphics is now about variety rather than meeting possibility requirements
It is true that with all these advantages comes a significant weakness.
Points along borders between different types of entities don't truly belong to either one or could be considered to belong to both. However, this won't affect much when it comes to the AI since we can assume that humans are intelligent and can pick the side of the boundary that suites their purpose. For example, in the case of traveling along an edge between woods and bare plains, if speed is what is important, the lower movement cost is used (i.e. the units travel on the side of the line that is the plains rather than the side that is the woods). For stealth units, the AI would be forced to pick whatever movement cost offers the best stealth (i.e. the woods value).
Other issues will arise as well, but in the end it is easier to program rules for dealing with ambiguous circumstances (especially for this team) than it is to create graphics.
Ah, now I understand what you are suggesting: Always walking along the edges of the terrain and feature plots (unless you also want to allow moving to the center point which you probably want as it is a bit weird otherwise). That would be a triangular meshing if plots and movement would be aligned.
So where would you be allowed to build cities? In the plot center only?
EDIT: One weird thing is also that the center point has higher connectivity compared to the border points.
So where would you be allowed to build cities? In the plot center only?
EDIT: One weird thing is also that the center point has higher connectivity compared to the border points.
@ primem0ver: It certainly is an intriguing idea to use points instead of tiles. I'm not a programmer, so I won't comment on the technical aspects - Koshling and the others know much more about this than I do.
In terms of game design, it would probably be the most radical depature yet suggested. Tiles have always been one of the holy cows of the Civilization series. They are taken to be a self-evident, integral part of that game, like the six attributes and saving throws are in D&D. A game without tiles would, in essence, be perceived to be a totally different type of game.
Personally, I'm not necessarily opposed to that - I'm all for trying out something new and innovative, like using globes as I suggested. But eventually, as you make more and more departures, the question arises: what type of game do Koshling et al. want to make? One that clearly displays its Civilization heritage, and stays roughly in that ballpark, or one that pretty radically departs from it in almost every department?
For instance, one vision for a game I have had for a long time is something along the lines of SimLife or Evolution: The Game of Intelligent Life (which Hydromancerx mentioned), but with polities instead of lifeforms. Civilization via the SimLife route, if you will. The player would be controlling and all the time changing overarching parameters, but the control over his polity / civilization would be largely indirect. It would be more like: "Experiment with the parameters of society, and watch how your people develop!"
I think it would be a very interesting type of game, but it's obviously totally different from Civilization, and very probably too much of a departure for many players here on Civfanatics to get excited about it.
In terms of game design, it would probably be the most radical depature yet suggested. Tiles have always been one of the holy cows of the Civilization series. They are taken to be a self-evident, integral part of that game, like the six attributes and saving throws are in D&D. A game without tiles would, in essence, be perceived to be a totally different type of game.
Personally, I'm not necessarily opposed to that - I'm all for trying out something new and innovative, like using globes as I suggested. But eventually, as you make more and more departures, the question arises: what type of game do Koshling et al. want to make? One that clearly displays its Civilization heritage, and stays roughly in that ballpark, or one that pretty radically departs from it in almost every department?
For instance, one vision for a game I have had for a long time is something along the lines of SimLife or Evolution: The Game of Intelligent Life (which Hydromancerx mentioned), but with polities instead of lifeforms. Civilization via the SimLife route, if you will. The player would be controlling and all the time changing overarching parameters, but the control over his polity / civilization would be largely indirect. It would be more like: "Experiment with the parameters of society, and watch how your people develop!"
I think it would be a very interesting type of game, but it's obviously totally different from Civilization, and very probably too much of a departure for many players here on Civfanatics to get excited about it.
makotech222
Warlord
- Joined
- Mar 18, 2012
- Messages
- 240
Laskaris, check out europa universalis 3 or victoria 2.
I own both games. Actually, I've done some modding work for EU3 a while ago.
However, these games are not in any way like what I outlined above. They don't have any similarities to SimEarth or SimLife. They also have a very different design philosophy compared to Civilization. Unlike Civilization, Paradox Interactive games are real-time strategy. They also cover a much shorter timespan - EU3 covers some 350 years, Victoria 2 only 100.
So I don't quite understand why you are bringing them up? I suppse Victoria 2, with its POP system, has an element of "indirect control", but that is the only vague resemblance I can see with the kind of game I described above.
primem0ver
Emperor
Well... sometimes but not most of the time. It would be where borders occur... but there are far fewer borders than points. Most points would be inside a "border" (whether that be cultural, terrain, features or other).
Anything that is positioned would need to be/should be centered on a point. This includes bonuses, cities, terrain improvements (yes... we would need programmed rules for improvements and terrain improvements could have some flexibility as far as point versus triangle), units. Cities would necessarily need to be centered on (or at least start on) points. since we would want to station units there. The nice thing though is that cities don't have to be confined to a single point (and could hence include "plots" or "tiles"). Keep in mind that cities in a plot system occupy entire plots which in this system would include 3-6 or more triangles. The actual city can grow to consume more points with population and building.
Things like peaks and hills should also be centered on points. Things that require textures and graphical extension that don't require a "center" would be composed of triangular "plots" such as borders, terrains, and features.
Actually this isn't true. All points on the grid have the same number of connecting points (six) except for those 20 original points of the icosahedron which would have 5 instead of 6 neighboring points. To see what I mean check out the graphic below (which is a flat hexagonal grid made of the points and triangles of which I speak)
primem0ver
Emperor
Yes and no. It is a radical difference from all previous civilization type games in terms of placement and map structure. However, it would not change much about the nature of the game or how it is played overall. Only some of the rules of play. Appearance would still be remarkably similar... to the point that it would be possible to have circumstances that make it still look like a tile based game. The only thing that would make it obvious that it wasn't in a screenshot would be to have entities such as bonuses or units sitting on an edge of a forest or some such border or positioned at consecutive points which can be avoided in screen shots. In such screen shots, the Civilization heritage would still be apparent.
LOL. I think is totally hilarious how alike we are in some ways. For a long time this has been my dream game. I was even thinking about it the other night; how it would be possible using an engine like we seem to be envisioning to combine a game like spore with civilization, with something like Master of Orion, in a sequence of escalating scale.
Imagine controlling the evolution of a species in an early game stage... but not necessarily being the winner.. the winner of each stage controlling the starting conditions for the next stage. For example, if the winners of the evolution stage are aquatic, then for the rest of the stages, an aquatic species is used; it is the oceans that are inhabited rather than land during the the "civilization" phase until technology grants the oceanic species the ability to go out of water.
The winners of the civilization phase control the type of society/culture that exists for the space expansion game. I even thought about a single celled stage in which the winning team picks the type of life that develops (individual, versus communal, versus single mind over several separate part types of life where "individuals" are all part of one creature).
But that is down a long road... right now we should keep are heads on the ground and take Civ to the next level (if this ever gets off the ground).
By God that would be the best game ever made. Why haven't the developers at major gaming companies made it yet? If we do it we could gain a name in the gaming industry.
I'm not sure. I think you under-estimate the difference in "feel" between the two. With tiles, you basically have a digital boardgame (which is a pretty apt description for what Civilization is). Throw tiles out, and I'd say you largely lose that digital boardgame feel, and change it to something quite different.
I'm torn on the issue. I see the advantages you cite for using points (and I'm interested in what the other techies here have to say about this). On the other hand, I have to say... I do like tiles! I find myself in the unusual role of the stodgy old conservative for once, but there you are. I think there's just a certain, endearing "old-fashionedness" about tiles...
I think so, too!
Well, my dream game is more modest than that. It would "only" cover the development of homo sapiens from the Stone Age to the first manned landing on Mars (i.e. the point in time where we become an interplanetary species). Of course the idea of also having the evolution of life and, for the endgame, large-scale space expansion is fascinating... But as a historian, I guess my focus is more on human history and potential alternate histories. I'd be happy to forgo evolution and the space game and instead model the development of human civilization on the planet in greater detail.
Hmm, well, actually there is a new hexagonal grid centered on the points then which is smaller than the original one on which you could consider the point based things to occupy.
You are right. I forgot some connectivity.
In total it is related to my original suggestion of a smaller grid with objects occupying more than one subplot.
And in this case it also has some hierarchy of implicit meshes.
Similar threads
