I really do not know where to stick this, and I need some semi-open place to discuss idea's and steps for my little first tile project. I have talked to some of you directly, but now that school is heavily under way I cannot remember some of the important details, so even if you told me something before please repeat it.
Goal: 1. Learn the basics of tileset work
2. Create a very basic tileset that has no "features" just land at different heights, smooth or cliff, costal, sandy etc. So that the "features" can be done as placeables (to experiment on later)
At first I was thinking of having many kinds of ground basis, and then working to integrate them, but while driving to and from school all last two weeks and watching the leaves change and the ground change, I realized it would be best to have two kinds of ground texture. A "grass" texture and a "dirt" texture, and then have a set of placeables the same shape as the tiles with different grass, dirt, leaves and possibly even snow to go over the tiles to alter the appearance (without walkmeshes of course)
Since that might be somewhat monumental in nature that leads me to think keeping the variety with in tile types would be good. And maybe it would be easy to copy just the surface of a tile and turn it into a placeable and then reskin that often.
Don't worry if you think in general that this tileset is impractical or ridicules for game use, it is an experimental project that is being done in my spare time (very little of that have I) and I think a couple of other equally crazily busy people also wanted to play around on this, and well I wanted to let anyone who wants to help create or constructively criticize help me. I hope this is an ok topic to place here.
I think it was Bannor suggested I get help from people here and since no one seems to be prattling in the forums when I am around, and my timeline slowed to a crawl (the arthritic turtle is going faster then I here) I hoped placing a post here would help.
Rebecca
Placeables Tileset
-
Bannor Bloodfist
- Posts: 1244
- Joined: Fri Oct 09, 2009 11:45 pm
- ctp: Yes
- dla: Yes
- TBotR: Yes
- nwnihof: Yes
Ok, here goes. First off, I do NOT want to discourage you at all.
Couple things that you need to consider right from the start:
Placeables in general for tilesets, is gonna cause you great pain.
Why you ask?
Engine limits, lag for example, is going be very high. Placeables just don't work well for tilesets.
You first have to create a base tile, then add a placeable on top of it. This is two seperate objects, for every tile in every area. Your total item count is going to be very high, before you even get to real items like walls, buildings, etc. That 'wastes' a minimum of 256 item counts in a standard 16x16 area. That number adds up pretty quickly, especially when you start adding things like plants, trees, walls/fences etc...
It's a nice idea, but it just does not work with the aurora game engine. It was not designed for that. You could attempt it in the same fashion that Lord of Worms created his seasonal forest, which is to have overlay haks, that change the textures only... however, even then, you won't necessarily accomplish what you are looking for. The primary problem with an overriding texture hak, is UVW mapping. The mapping is stored on the tile mdl itself, NOT on the texture. So, you get once chance to map your texture down onto your object, and changing the texture name, does not allow you to change the mapping. It's pretty obvious in LoW sets that the mapping gets pretty bad when you have multiple textures assigned to the same object using overrides.
Next issue is placeable buildings. These would require absolutely flat tiles to place them onto. Then engine does not recognize 'Z' at all, and once a placeable is dropped onto the surface, if you adjust anything, you lose it's original placement. The objects, (buildings, walls, whatever) tend to drop to Zero Z. This might not be an issue except for the fact that the engine does not allow a placeable to cut a hole into the walkmesh. The best you can accomplish is to 'stack' walkable surfaces higher than the original Zero Z.
This also means that your walkmesh, is worthless on the tile for pathfinding, as the engine can't tell that you have placeables when it does the pathfinding algorithms. Basically this means that every tile would have to have an 'A' pathnode to start with. Which is fine, except that pathfinding doesn't work when you have placeables in the way (fences, buildings, trees, etc). So, npc's get stuck, your henchmen can't follow the pc, the pc can't click half a map away and 'run' to that location, as he gets stuck when he/she runs into the first placeable in the way.
Basically, what I am saying is that tiles as placeables just doesn't work with NWN in any practical aspect. Oh, you can accomplish it, and make it 'look' pretty, but it just won't function very well, if at all, in a module.
Couple things that you need to consider right from the start:
Placeables in general for tilesets, is gonna cause you great pain.
Why you ask?
Engine limits, lag for example, is going be very high. Placeables just don't work well for tilesets.
You first have to create a base tile, then add a placeable on top of it. This is two seperate objects, for every tile in every area. Your total item count is going to be very high, before you even get to real items like walls, buildings, etc. That 'wastes' a minimum of 256 item counts in a standard 16x16 area. That number adds up pretty quickly, especially when you start adding things like plants, trees, walls/fences etc...
It's a nice idea, but it just does not work with the aurora game engine. It was not designed for that. You could attempt it in the same fashion that Lord of Worms created his seasonal forest, which is to have overlay haks, that change the textures only... however, even then, you won't necessarily accomplish what you are looking for. The primary problem with an overriding texture hak, is UVW mapping. The mapping is stored on the tile mdl itself, NOT on the texture. So, you get once chance to map your texture down onto your object, and changing the texture name, does not allow you to change the mapping. It's pretty obvious in LoW sets that the mapping gets pretty bad when you have multiple textures assigned to the same object using overrides.
Next issue is placeable buildings. These would require absolutely flat tiles to place them onto. Then engine does not recognize 'Z' at all, and once a placeable is dropped onto the surface, if you adjust anything, you lose it's original placement. The objects, (buildings, walls, whatever) tend to drop to Zero Z. This might not be an issue except for the fact that the engine does not allow a placeable to cut a hole into the walkmesh. The best you can accomplish is to 'stack' walkable surfaces higher than the original Zero Z.
This also means that your walkmesh, is worthless on the tile for pathfinding, as the engine can't tell that you have placeables when it does the pathfinding algorithms. Basically this means that every tile would have to have an 'A' pathnode to start with. Which is fine, except that pathfinding doesn't work when you have placeables in the way (fences, buildings, trees, etc). So, npc's get stuck, your henchmen can't follow the pc, the pc can't click half a map away and 'run' to that location, as he gets stuck when he/she runs into the first placeable in the way.
Basically, what I am saying is that tiles as placeables just doesn't work with NWN in any practical aspect. Oh, you can accomplish it, and make it 'look' pretty, but it just won't function very well, if at all, in a module.
-
Winterhawk99
- Posts: 1648
- Joined: Thu Oct 08, 2009 12:00 am
- ctp: Yes
- nwnihof: Yes
- Location: Pa.
- Contact:
Harvest Moon uses lots and lots of tile placeables and sunjammers tilemagic. So this can be done, HOWEVER..... it takes a very competent builder that can work for hours to get it correct so that players can feel an sense the illusion your createing without walkmesh and other issues. We do accomplish this for the most part with blocking tiles placeables that cover gaps and other tricks of advanced building including some scripting. No matter if you use these techniques or not you will never get 100% with the walkmesh issues alone. there will always be places where the placeable walkmesh with overide the original and create weird effects some of these glitches can be used to your advantage most take away any semblence of immersion when pcs find that gap you didnt see or take care of.
Believe me i know from experiance. When I use tileplaceables it can take me upwards of 80 hours for a 12x13 area to think i got it right only to find out i have to spend another 40 hours fixing my own gaps where the pc can go into a "percieved tile" or the pc jumps to the top of a mountain or sinks below the visual tile.
Believe me i know from experiance. When I use tileplaceables it can take me upwards of 80 hours for a 12x13 area to think i got it right only to find out i have to spend another 40 hours fixing my own gaps where the pc can go into a "percieved tile" or the pc jumps to the top of a mountain or sinks below the visual tile.
I doubt I will really do much with this. Everyone tells me why it won't work, except pstmarie, and so few people want to experiment and push NWN to see if you can do new things with it... and it always gets back to that "how playable and will players like it" that I don't care about... I mean if they do cool, if they don't I still learned and tried new things....
But well, I am in a very intense school program now, and if when I want to do stuff with this to experiment and play, all I get back from the community is all the problems I will face, I get discouraged and find distraction elsewhere.
I have been thinking about these really cool idea's around a placeable-based tileset and ways around those very issues you mentioned everyday on the way to and from school and I got excited about it enough today to want to discuss the idea with someone...
Now I am totally deflated and remember why I keep giving up on NWNs. I don't understand this community anymore. I remember to much how it was pre-expansions when everyone wanted to try and experiment with it. And several of those successes people now do not believe would work even though I have seen them in action I cannot convince people would work in NWNs. NWN is still as cool as it was...
This community is loosing something... that is excitement, innovation and it is turning into everything has to be this way or that.... I keep hearing those kinds of sentiments from people in all sorts of ways... and I am not interested in that game, I like NWNs because I could do something completely different with it... If the attitude towards it keeps up, I will inevitably join in the mass of people abandoning it.
I know a lot of limitations that come from placeables... but tilesets come with a certain set of limitations that everyone just got used to... I want to try trading some of the tileset limitations for placeable ones... I think that ways to make a tileset based placeable that is not hugely laggy can be done. I do not know if what I want to do can be done lag-free, or not, but I wanted to face lag issues with attempts to work around them, with things like walk-mesh free placeables that rest on top of tiles at .01 increments higher that can layer on textures for more ground cover coloring, and plants with walk meshes that do not match the outside skin exactly, but are much simpler to save on poly count, and have the set of placeables be specifically designed to go with the tileset, including curves, as well as a non flat bottom to items that is only visually non-flat so that when placed on curved surfaces would look more natural.
Of course I do not know if I can do any of these things, and I especially cannot all on my own, I have way to much darned math homework. Hours and hours of it daily. I mean I have 5 math classes this semester. And I find that imagining ways to work through placeable lag issues seems to have been very fun and relaxing for me lately.
I may have totally over-reacted to that... I may just have over-reacted a little bit. This comes from that fact that I really respect your opinion Bannor, and admire your work and like you. I guess hearing pretty much the same sort of "this is a bad idea because" from you was harder to take then from most. I don't know...
Maybe I really should give this game up...
Please do not be offended... oh please don't... clearly my idea's for this game are not exciting people... and I am sick of fighting against some main stream, and dealing with people who have this sort of fixed idea about what NWN is supposed to be. I am sick of seeing people I like have stupid flame wars in forums. I am sick of all the telling everyone what is wrong with their idea's and not what is good about them... I know your advice was well meant, of course I do and I appreciate it...
I like NWN because I can do something different with it.... so I do not want to just do another tileset like other people do... I like imagining and trying out new things. I always have. Most of what I imagine never goes far, but when idea's rattle around and won't be silenced I want to share them with other people...
That was why I posted earlier...
I must have been excited about it cause I had tried before on the more main forums but felt pretty discouraged by those, and tried here... a place I respect for showing decorum and respect for people. And I do not feel offended by what you said... I just feel hopeless about it, and the game...
And clearly I am something of upset as I have not stopped typing... I feel very sad about NWNs now... and wonder if it is still even worth it...
Rebecca
But well, I am in a very intense school program now, and if when I want to do stuff with this to experiment and play, all I get back from the community is all the problems I will face, I get discouraged and find distraction elsewhere.
I have been thinking about these really cool idea's around a placeable-based tileset and ways around those very issues you mentioned everyday on the way to and from school and I got excited about it enough today to want to discuss the idea with someone...
Now I am totally deflated and remember why I keep giving up on NWNs. I don't understand this community anymore. I remember to much how it was pre-expansions when everyone wanted to try and experiment with it. And several of those successes people now do not believe would work even though I have seen them in action I cannot convince people would work in NWNs. NWN is still as cool as it was...
This community is loosing something... that is excitement, innovation and it is turning into everything has to be this way or that.... I keep hearing those kinds of sentiments from people in all sorts of ways... and I am not interested in that game, I like NWNs because I could do something completely different with it... If the attitude towards it keeps up, I will inevitably join in the mass of people abandoning it.
I know a lot of limitations that come from placeables... but tilesets come with a certain set of limitations that everyone just got used to... I want to try trading some of the tileset limitations for placeable ones... I think that ways to make a tileset based placeable that is not hugely laggy can be done. I do not know if what I want to do can be done lag-free, or not, but I wanted to face lag issues with attempts to work around them, with things like walk-mesh free placeables that rest on top of tiles at .01 increments higher that can layer on textures for more ground cover coloring, and plants with walk meshes that do not match the outside skin exactly, but are much simpler to save on poly count, and have the set of placeables be specifically designed to go with the tileset, including curves, as well as a non flat bottom to items that is only visually non-flat so that when placed on curved surfaces would look more natural.
Of course I do not know if I can do any of these things, and I especially cannot all on my own, I have way to much darned math homework. Hours and hours of it daily. I mean I have 5 math classes this semester. And I find that imagining ways to work through placeable lag issues seems to have been very fun and relaxing for me lately.
I may have totally over-reacted to that... I may just have over-reacted a little bit. This comes from that fact that I really respect your opinion Bannor, and admire your work and like you. I guess hearing pretty much the same sort of "this is a bad idea because" from you was harder to take then from most. I don't know...
Maybe I really should give this game up...
Please do not be offended... oh please don't... clearly my idea's for this game are not exciting people... and I am sick of fighting against some main stream, and dealing with people who have this sort of fixed idea about what NWN is supposed to be. I am sick of seeing people I like have stupid flame wars in forums. I am sick of all the telling everyone what is wrong with their idea's and not what is good about them... I know your advice was well meant, of course I do and I appreciate it...
I like NWN because I can do something different with it.... so I do not want to just do another tileset like other people do... I like imagining and trying out new things. I always have. Most of what I imagine never goes far, but when idea's rattle around and won't be silenced I want to share them with other people...
That was why I posted earlier...
I must have been excited about it cause I had tried before on the more main forums but felt pretty discouraged by those, and tried here... a place I respect for showing decorum and respect for people. And I do not feel offended by what you said... I just feel hopeless about it, and the game...
And clearly I am something of upset as I have not stopped typing... I feel very sad about NWNs now... and wonder if it is still even worth it...
Rebecca
-
Winterhawk99
- Posts: 1648
- Joined: Thu Oct 08, 2009 12:00 am
- ctp: Yes
- nwnihof: Yes
- Location: Pa.
- Contact:
Tell you what Rebecca if you really want to do this come to HM this weekend and Ill show you how we set them up and you can see how we do it. There are techniques the will work well and I can show you some of the pitfalls. Then it may help you with some good techniques and ways to approach building with placeable tiles.
-
Bannor Bloodfist
- Posts: 1244
- Joined: Fri Oct 09, 2009 11:45 pm
- ctp: Yes
- dla: Yes
- TBotR: Yes
- nwnihof: Yes
Rebecca, I specifically posted that I did not want to discourage you!
But I do feel that tilesets as placeables just won't work. The engine can't perform what you are asking of it with tiles as placeables.
You CAN do buildings and such, that has been done several times already.
Changing ground textures can also be accomplished, but typically only works on flat terrains and no trees, hills, buildings in the way.
Tilesets in general require major work. Lots of time. The more 'stuff' you wish to add, the more time it takes, exponentially. For each 'crosser' like streams, roads, fences, you MUST add all the specifics for every terrain type etc... it gets real painful, REALLY fast. I know this because of the different times I listened to Mask and 'added' new walls into a set, then had to add all the connecting tiles... It can take forever to even think of all the combination's that are actually required.
It is why TNO is over 1200 tiles... mainly because of the different terrain/crosser tiles that are required to make them work. Many of them will never actually be seen by folks, as they only show up in specific situations/combination;s of tiles... like a wall, intersecting with a corner hill, with a gate etc...
Thrikreen is working on adding some features into nwmax, to help you create 'blanks' that auto-fill when you add new terrains and crossers, just to tell you what tiles you must actually create. That will be a huge help to tileset authors for NWN.... when he releases the update that is.
But I do feel that tilesets as placeables just won't work. The engine can't perform what you are asking of it with tiles as placeables.
You CAN do buildings and such, that has been done several times already.
Changing ground textures can also be accomplished, but typically only works on flat terrains and no trees, hills, buildings in the way.
Tilesets in general require major work. Lots of time. The more 'stuff' you wish to add, the more time it takes, exponentially. For each 'crosser' like streams, roads, fences, you MUST add all the specifics for every terrain type etc... it gets real painful, REALLY fast. I know this because of the different times I listened to Mask and 'added' new walls into a set, then had to add all the connecting tiles... It can take forever to even think of all the combination's that are actually required.
It is why TNO is over 1200 tiles... mainly because of the different terrain/crosser tiles that are required to make them work. Many of them will never actually be seen by folks, as they only show up in specific situations/combination;s of tiles... like a wall, intersecting with a corner hill, with a gate etc...
Thrikreen is working on adding some features into nwmax, to help you create 'blanks' that auto-fill when you add new terrains and crossers, just to tell you what tiles you must actually create. That will be a huge help to tileset authors for NWN.... when he releases the update that is.
Oh I wasn't going to make streams and roads placeables, just the trees and rocks and ground texturing.
The idea, originally that my daughter and I had was for a sort of persistant world where things where more flexible. I am not sure placeables en mass are a great idea, but we wanted to give it a try, and well I have this idea about a tree that is a multi-part placeable that has the branches separate from the leaves so that the leaves could be scripted to change through a cycle to simulate fall.
I am not sure I care if its all that playable, I am curious to see if it can be done at all. Besides the real goal is learning more about modeling, and even if I have a useless tileset, the knowledge from how to build it will be useful.
I certainly don't want to go through the trouble of trying to build something that does not interest me, cause then I would never spend any time on it. And this has fascinated me for long enough that I think I may have enough interest to accomplish it.
As for my reaction, I am sorry. I have not been able to properly sleep for a long while not and it is wrecking havoc with my attitude. I really should avoid talking to people when sleep deprived, unfortunately my judgment also gets off when sleep deprived... *shakes head* I want to sleep...
The idea, originally that my daughter and I had was for a sort of persistant world where things where more flexible. I am not sure placeables en mass are a great idea, but we wanted to give it a try, and well I have this idea about a tree that is a multi-part placeable that has the branches separate from the leaves so that the leaves could be scripted to change through a cycle to simulate fall.
I am not sure I care if its all that playable, I am curious to see if it can be done at all. Besides the real goal is learning more about modeling, and even if I have a useless tileset, the knowledge from how to build it will be useful.
I certainly don't want to go through the trouble of trying to build something that does not interest me, cause then I would never spend any time on it. And this has fascinated me for long enough that I think I may have enough interest to accomplish it.
As for my reaction, I am sorry. I have not been able to properly sleep for a long while not and it is wrecking havoc with my attitude. I really should avoid talking to people when sleep deprived, unfortunately my judgment also gets off when sleep deprived... *shakes head* I want to sleep...
[quote=Rebecca;14910]Oh I wasn't going to make streams and roads placeables, just the trees and rocks and ground texturing.[/quote]
What a relief - trying to cut out all those cobblestones and make something that meshed together adequately was making me cross-eyed
I've been playing around with the microset and placeables - laying down as many as I can and seeing if I can give the game a migraine. So far - no luck. Although pathfinding can be a nightmare at times, I've noticed that if the placeables are clustered the engine has an easier time calculating pathfinding. That is to say you can actually move around - albeit without much of the finesse mentioned earlier by Bannor. Pathfinding also seems to be easier with placeables that have a simpler pwk geometry - 2 faces seems to be ideal - which would make sense of course.
Not sure about scripting in leaves that change - when the placeable spawned you'd have to script in it positioning as well, making it more difficult to align the foliage with the trees. What might work better is doing a tree for season, swapping them out through a script...
Basically you'd drop waypoint where you want the tree spawn then make a script that spawns a seasonal tree at the waypoint's location. You could also set the tree to match the orientation of the waypoint - in effect, using the waypoint to store the placeable's orientation....
What I'd do is paint a tree first, note its x/y/z position and rotation, then drop a waypoint at the same coordinates. Once the waypoint is painted, just delete the tree and you're all set.
These lines of code would create the placeable tree. You'll need to add some function for tracking what time of year it is...
The script named SetMyFacing would need the following lines of code...
There may be some errors in the code as I'm just pulling it from off the cuff so to speak.
What a relief - trying to cut out all those cobblestones and make something that meshed together adequately was making me cross-eyed
I've been playing around with the microset and placeables - laying down as many as I can and seeing if I can give the game a migraine. So far - no luck. Although pathfinding can be a nightmare at times, I've noticed that if the placeables are clustered the engine has an easier time calculating pathfinding. That is to say you can actually move around - albeit without much of the finesse mentioned earlier by Bannor. Pathfinding also seems to be easier with placeables that have a simpler pwk geometry - 2 faces seems to be ideal - which would make sense of course.
Not sure about scripting in leaves that change - when the placeable spawned you'd have to script in it positioning as well, making it more difficult to align the foliage with the trees. What might work better is doing a tree for season, swapping them out through a script...
Basically you'd drop waypoint where you want the tree spawn then make a script that spawns a seasonal tree at the waypoint's location. You could also set the tree to match the orientation of the waypoint - in effect, using the waypoint to store the placeable's orientation....
What I'd do is paint a tree first, note its x/y/z position and rotation, then drop a waypoint at the same coordinates. Once the waypoint is painted, just delete the tree and you're all set.
These lines of code would create the placeable tree. You'll need to add some function for tracking what time of year it is...
Code: Select all
void main()
{
object oWP = GetWaypointByTag("");
location lLoc = GetLocation(oWP);
float fFacing = GetFacing(oWP);
string sResRef;
int nSeason = ;
switch (nSeason)
{
case 1: sResRef = "Tree_Spring"; break;
case 2: sResRef = "Tree_Summer"; break;
case 3: sResRef = "Tree_Fall"; break;
case 4: sResRef = "Tree_Winter"; break;
}
object oTree = CreateObject(OBJECT_TYPE_PLACEABLE, sResRef, lLoc);
SetLocalFloat(oTree, "MyFacing", fFacing);
ExecuteScript(oTree, "SetMyFacing");
}
Code: Select all
void main ()
{
float fFacing = GetLocalFloat(oTree, "MyFacing");
SetFacing(fFacing);
}
Has anyone ever tried the following :
1/ keep 1 wokmesh per tile, with no visible geometry
2/ create 1 placeable out of the fusion of all the tiles
3/ place the placeable in the area at the load of the module ?
For a 16x16 area, it would have to be a huge (!) placeable, I realize that, but a static one, so not so demanding on the engine.
The gist would be that we could create 1 placeable for each season, for example, that could be dynamically exchanged.
Footstepsounds would not match, and I am not sure NWmax would accept to export it, but hey, since we're brainstorming...
PS : In any case, thank you for starting the topic, Rebecca
1/ keep 1 wokmesh per tile, with no visible geometry
2/ create 1 placeable out of the fusion of all the tiles
3/ place the placeable in the area at the load of the module ?
For a 16x16 area, it would have to be a huge (!) placeable, I realize that, but a static one, so not so demanding on the engine.
The gist would be that we could create 1 placeable for each season, for example, that could be dynamically exchanged.
Footstepsounds would not match, and I am not sure NWmax would accept to export it, but hey, since we're brainstorming...
PS : In any case, thank you for starting the topic, Rebecca
-
maxam.maxam
- Posts: 13
- Joined: Sun Jun 17, 2007 5:08 am
- ctp: No
- dla: No
- TBotR: No
- nwnihof: No
- Contact:
[quote=Nissa_Red;14918]Has anyone ever tried the following :
1/ keep 1 wokmesh per tile, with no visible geometry
2/ create 1 placeable out of the fusion of all the tiles
3/ place the placeable in the area at the load of the module ?
[/quote]
Ok, as a placeable this would not work because placeables do not have walkmeshes per-se. Their equivalent is a 2d plane that marks the area that a placeable is "unwalkable".
This is why you never see placeables that you can walk on (only through).
HOWEVER
This is a great idea and it could be done along similar lines...
For example, you could make your "seasonal" area as a 16 x 16 tile group, and make winter, summer, autumn and spring versions (4 groups). You could then create an area for each appropriate season and create an entry/exit script that directs traffic to the appropriate area depending on the season.
1/ keep 1 wokmesh per tile, with no visible geometry
2/ create 1 placeable out of the fusion of all the tiles
3/ place the placeable in the area at the load of the module ?
[/quote]
Ok, as a placeable this would not work because placeables do not have walkmeshes per-se. Their equivalent is a 2d plane that marks the area that a placeable is "unwalkable".
This is why you never see placeables that you can walk on (only through).
HOWEVER
This is a great idea and it could be done along similar lines...
For example, you could make your "seasonal" area as a 16 x 16 tile group, and make winter, summer, autumn and spring versions (4 groups). You could then create an area for each appropriate season and create an entry/exit script that directs traffic to the appropriate area depending on the season.