mesilliac's terrain graphics thread
Moderator: Forum Moderators
Forum rules
Before posting critique in this forum, you must read the following thread:
Before posting critique in this forum, you must read the following thread:
Re: mesilliac's terrain graphics thread
The stone wall was commited to trunk already, but I backported it to my campaign. Just FYI.
Author of the unofficial UtBS sequels Invasion from the Unknown and After the Storm.
- Aethaeryn
- Translator
- Posts: 1554
- Joined: September 15th, 2007, 10:21 pm
- Location: Baltimore, Maryland, USA
Re: mesilliac's terrain graphics thread
Yet very rewarding also in the area of outdoor walls (for RPGs for instance) or other kinds of barriers such as fences... Maybe GSoC idea next year?Boucman wrote:between hex would be horribly hard to implement...
Aethaeryn (User Page)
Wiki Moderator (wiki)
Latin Translator [wiki=Latin Translation](wiki)[/wiki]
Maintainer of Thunderstone Era (wiki) and Aethaeryn's Maps [wiki=Aethaeryn's Maps](wiki)[/wiki]
Wiki Moderator (wiki)
Latin Translator [wiki=Latin Translation](wiki)[/wiki]
Maintainer of Thunderstone Era (wiki) and Aethaeryn's Maps [wiki=Aethaeryn's Maps](wiki)[/wiki]
Re: mesilliac's terrain graphics thread
No, no, no and no.Aethaeryn wrote:Yet very rewarding also in the area of outdoor walls (for RPGs for instance) or other kinds of barriers such as fences... Maybe GSoC idea next year?Boucman wrote:between hex would be horribly hard to implement...
- Aethaeryn
- Translator
- Posts: 1554
- Joined: September 15th, 2007, 10:21 pm
- Location: Baltimore, Maryland, USA
Re: mesilliac's terrain graphics thread
Just an idea, probably too much effort to be worth the results.zookeeper wrote:No, no, no and no.
Aethaeryn (User Page)
Wiki Moderator (wiki)
Latin Translator [wiki=Latin Translation](wiki)[/wiki]
Maintainer of Thunderstone Era (wiki) and Aethaeryn's Maps [wiki=Aethaeryn's Maps](wiki)[/wiki]
Wiki Moderator (wiki)
Latin Translator [wiki=Latin Translation](wiki)[/wiki]
Maintainer of Thunderstone Era (wiki) and Aethaeryn's Maps [wiki=Aethaeryn's Maps](wiki)[/wiki]
Re: mesilliac's terrain graphics thread
Its probably a lot of effort if this requires an overhaul of the system. So unless you do it yourself, don't expect it to be on the game.Aethaeryn wrote:Just an idea, probably too much effort to be worth the results.
About the thread. I think that these test of concept looks nice, and the perspective is rather nice, looks like higher ground to me. At least is a refreshing image even if it never get included. So props for the work spent on it.
"Mysteries are revealed in the light of reason."
Re: mesilliac's terrain graphics thread
*sigh*
Did no one other than me notice that the cliffs aren't intended to function as walls between hexes at all, but that
Did no one other than me notice that the cliffs aren't intended to function as walls between hexes at all, but that
i.e. only flying units can get above the cliffs anyway?...mesilliac wrote:In terms of game mechanics, they are also an alias of chasm.
For I am Turin Turambar - Master of Doom, by doom mastered. On permanent Wesbreak. Will not respond to private messages. Sorry!
And I hate stupid people.
The World of Orbivm
And I hate stupid people.
The World of Orbivm
- Darker_Dreams
- Posts: 608
- Joined: February 1st, 2008, 5:26 pm
Re: mesilliac's terrain graphics thread
I can't really tell from the posted pic, but just to be sure, what's happening with the cliffs is not that the rock-face is occupying a whole hex?
If it's not, the problem I see is what happens when someone wants to do a level where you start on the clifftop, or when someone creates a path (bridge) between the two? Also, how does the block work on the backside. Reflecting, the backside cliffs suggest that there is a set of edge-hexes being treated as chasm but which will create nonsense situations if a non-flier were to get up there.
Visually, it's beautiful, and I'll have no compunctions about using it if I come up with something within its apparent constraints.
If it's not, the problem I see is what happens when someone wants to do a level where you start on the clifftop, or when someone creates a path (bridge) between the two? Also, how does the block work on the backside. Reflecting, the backside cliffs suggest that there is a set of edge-hexes being treated as chasm but which will create nonsense situations if a non-flier were to get up there.
Visually, it's beautiful, and I'll have no compunctions about using it if I come up with something within its apparent constraints.
Re: mesilliac's terrain graphics thread
I'll clarify a bit.
Sangel almost completely summed it up with those 2 options. My cliffs here are option 1: cliffs drawn between hexes. However they have the caveat that walking units may never get on top of them. This sidesteps the question of inter-hex barriers completely . Turin was quite right in that this is exactly what I meant when I said they were an alias of chasm.
I have some WML which would probably draw option 2 sensibly. Zookeeper had a prototype version a while back, too (http://forum.wesnoth.org/viewtopic.php?f=2&t=10081). The main blocker for me, is the issue with transitions to other terrains. If the cliff wall is running across a hex, and there is grass on one side and snow on the other, there is no easy way to make one half of the tile snow and the other half grass.
That means cliff images (or a whole new set of transitions) would have to be drawn for every distinct base terrain .
On the whole I'd probably prefer (2), but unwalkable (1) can be implemented easily.
Sangel almost completely summed it up with those 2 options. My cliffs here are option 1: cliffs drawn between hexes. However they have the caveat that walking units may never get on top of them. This sidesteps the question of inter-hex barriers completely . Turin was quite right in that this is exactly what I meant when I said they were an alias of chasm.
I have some WML which would probably draw option 2 sensibly. Zookeeper had a prototype version a while back, too (http://forum.wesnoth.org/viewtopic.php?f=2&t=10081). The main blocker for me, is the issue with transitions to other terrains. If the cliff wall is running across a hex, and there is grass on one side and snow on the other, there is no easy way to make one half of the tile snow and the other half grass.
That means cliff images (or a whole new set of transitions) would have to be drawn for every distinct base terrain .
On the whole I'd probably prefer (2), but unwalkable (1) can be implemented easily.
-
- Moderator Emeritus
- Posts: 2232
- Joined: March 26th, 2004, 10:58 pm
- Location: New York, New York
Re: mesilliac's terrain graphics thread
Yes, that's pretty much as I had feared - the snow on one side, grass on the other problem. I'm impressed that you solved the WML problem already; that leaves only the problem of the graphics. There might actually be a feasible workaround here.mesilliac wrote:I'll clarify a bit.
I have some WML which would probably draw option 2 sensibly. Zookeeper had a prototype version a while back, too (http://forum.wesnoth.org/viewtopic.php?f=2&t=10081). The main blocker for me, is the issue with transitions to other terrains. If the cliff wall is running across a hex, and there is grass on one side and snow on the other, there is no easy way to make one half of the tile snow and the other half grass.
That means cliff images (or a whole new set of transitions) would have to be drawn for every distinct base terrain .
The following relates to Option 2 - that that cliff tiles fall in the middle of a hex, representing an impassible hex.
Suppose that the cliffs are drawn as the top-most layer of the base terrain (ignoring castles etc for the moment). Assuming that the cliffs themselves are quite wide, they'd thereby cover up a good portion of the hex. Now, what if we could create an algorithm that automatically carves up hexes along certain lines, putting one terrain on one side, and another on the other? Once you placed the cliffs down on top, the actual (and abrupt) division between the two would be completely obscured. Voilà - snow on one side, grass on the other, without creating whole new grass-cliff and snow-cliff tiles for every tile in the cliff set.
There would doubtless be considerable work needed on the cliffs to make them transition smoothly to multiple terrain types, but it would be n-transitions, rather than n-squared-transitions.
Is this line of thought making sense, or is it overly complicated/impractical?
"Pure logic is the ruin of the spirit." - Antoine de Saint-Exupéry
Re: mesilliac's terrain graphics thread
I'm sure this request was made with this issue in mind. I'm pretty sure it'd be feasible if that masking thing was implemented.Sangel wrote:Is this line of thought making sense, or is it overly complicated/impractical?
Re: mesilliac's terrain graphics thread
I put some thought into making cliffs (of type 2) some time ago, and it *is* difficult to do without essentailly having special cases/graphics for each possible combination.Sangel wrote:Suppose that the cliffs are drawn as the top-most layer of the base terrain (ignoring castles etc for the moment). Assuming that the cliffs themselves are quite wide, they'd thereby cover up a good portion of the hex. Now, what if we could create an algorithm that automatically carves up hexes along certain lines, putting one terrain on one side, and another on the other? Once you placed the cliffs down on top, the actual (and abrupt) division between the two would be completely obscured. Voilà - snow on one side, grass on the other, without creating whole new grass-cliff and snow-cliff tiles for every tile in the cliff set.
There would doubtless be considerable work needed on the cliffs to make them transition smoothly to multiple terrain types, but it would be n-transitions, rather than n-squared-transitions.
Is this line of thought making sense, or is it overly complicated/impractical?
Masks would be a neccessary feature, but we would still need a way to assign these half-tiles (unless we add one rule for each (base) terrain). But it would still be feasible as long as there is only one terrain on each side of the cliff. As soon as there are two different terrains on one side we have transitions between *those* to deal with and it gets really ugly.
An easier way out would be to make the cliffs wide enough so that they occupy the whole hex.
Aurë entuluva!
Re: mesilliac's terrain graphics thread
Urg, I hadn't thought about what to do when there are two different terrains on one side of the cliff.mog wrote: Masks would be a neccessary feature, but we would still need a way to assign these half-tiles (unless we add one rule for each (base) terrain). But it would still be feasible as long as there is only one terrain on each side of the cliff. As soon as there are two different terrains on one side we have transitions between *those* to deal with and it gets really ugly.
An easier way out would be to make the cliffs wide enough so that they occupy the whole hex.
The problem with making the cliff graphic take up the whole hex, is that I can't think of a way to draw it that wouldn't look either climbable or silly . The problem is most obvious when you think of a north-facing cliff. There the perspective means that you have a lot of tile to fill with something.
Here is an example of what I have in mind when I think of cliffs for wesnoth: http://images.quickblogcast.com/34670-3 ... Valley.jpg (for similar, search google images for "glacier carved valley").
The best idea I came up with for making the cliffs take up more room is to make them "canyon walls"... with big water-carved rock pillars taking up most of the space on the tile. But really, it doesn't seem right for wesnoth's geography.
The second-best idea I had was to terrace them, heavily. It's hard to imaging not being able to climb them in this case, but it would probably look quite nice.
Every time I think of all this, unwalkable option (1) really seems like an incredibly good option to me .
Re: mesilliac's terrain graphics thread
Come to think of it... just making a new base terrain for the cliffs would probably be ok. "rocky grass" and "snowy rocks" would fit almost everywhere, and the cliffs could be an overlay on top of that. Cliffs would need an additional snowy version anyway. Hmmmmmm...
-
- Art Contributor
- Posts: 410
- Joined: October 30th, 2006, 4:55 am
- Location: The Big Ö (a.k.a. Austria)
Re: mesilliac's terrain graphics thread
Awesome sauce: mnln plz kthxbi
You'd also need a cave version though. However! If you do the cliff edge as an overlay, it can easily be placed on arbitrary terrains in the terrain def file, just as hill/mountain villages are now (I think? ).
You'd also need a cave version though. However! If you do the cliff edge as an overlay, it can easily be placed on arbitrary terrains in the terrain def file, just as hill/mountain villages are now (I think? ).
that little girl's parents were attacked by ninjas - generic npc
hee hee! - little girl
hee hee! - little girl