Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Neonivek]

Hahaha, joke's on you! This IS still Ultima VII data.

Neonivek: agreed, I've even stated that myself.
Just wishful thinking, easily fixed too.
(though I don't know where the walls are not single tile objects.)

It really makes me impatient for the time Toady finally allows objects to be done in Art! perhaps even with quality differences!

Goodness I really can't wait

Does Toady currently have a "Blank" color... or rather a color (with shade) that has a color as defined by the raws? That way you can have Swords have a different color depending on material much easier... then again the colors are currently quite limited.

Maybe by then, and when my computer is fixed (I am using a different one), I could put my hand into the ring for foods maybe!

[Mikademus]

Hahaha, joke's on you! This IS still Ultima VII data.

Neonivek: agreed, I've even stated that myself.
Just wishful thinking, easily fixed too.
(though I don't know where the walls are not single tile objects.)

Well, joke's on Tormy really... Lying bastard that he is!

Anyway, I can't help but notice how you sidestepped the entire Ultima VI request... *hopes*

[Mike Mayday]

No way man, 7 is too much better ;>

[Khelz]

HMmm great ideas going around.

Concerning the tile-hiding issue: What about a transparency system triggered locally where there's something behind a wall? (Character or object).

[Mikademus]

No way man, 7 is too much better ;>

U7's tiles are gorgeous, but I find U6's simply cosier. Ah, happier times...

[penguinofhonor]

Also in the Zelda pic... The "O"s should be "I"

Depends on whether it's a wall or a support. A single wall is an O and a support is an I.

I like the idea of having tons of different graphics styles for DF. The only problem I see is it being a lot of work, but Toady can just add them in one at a time when he feels like it. Even if I'll always use ASCII, I still think all these ideas are great.

[Nadaka]

I had a rather interesting idea this weekend for graphics...

The problem with isometric view is that a block wall obscures what is behind it. Also of minor note is that diagonal blocks seem to be solide, even though dwarves can walk between them. Also of minor note is that diagonal walls are very jagged while horizantal or vertical walls are very smooth.

What if we used short octagonal blocks instead of square? Doing so allows at least a quarter of the tile behind a wall to be visible. It also makes some diagonals appear to support visual gaps and all walls will be undulating to some degree. Another advantage is that multi level views are less confusing because the grid for each level is somewhat skewed from the one below it.



This will also have a performance advantage over the "rotated cavalier" view because each tile can be drawn without consulting what is in adjacent tiles. Whereas with the rotated cavalier, in order to get the right graphic for any wall you have to check all 8 of its surrounding tiles.

However, it may still be advantageous to do that kind of extensive checking. The tile group on the far right uses a lighter border coloring  to "fill" in gaps in the walls and make a clearer distinction between solid walls and diagonals that can be passed through. This would require the same kind of checking and produces a more visually appealing view.

Also trying to figure out a good way to display ramps, that is a work in progress.

[Captain Mayday]

I'd like to add my input here.
To me, the Xcom style of Isometric would seem to be the way to go for the following reasons.
First point: Doors are always visible.
Second point: It's fairly easy to rotate isometric, in terms of 90 degree rotations.
Third point: It's built to allow viewing of multiple z levels. At the moment, DF allows you to view say... 2 tiles down, by fading the terrain colours into dots. With Isometric we should be able to do that much inside and far more outside.
Fourth point: Xcom had selection boxes which boxed down to the ground level directly below in order to show height. This would be ideal.
Fifth point: creatures could be somewhat larger than their tile's borders, allowing a more interesting display.

Now, the issues in regards to this.
1. Interface. The controls are going to need to be made intuitive here. More mouse support would really go a long way to help here.
2. Display. Multiple creatures on the same tile. This isn't an issue for flying creatures, since only one creature can be standing on a given tile. If a creature isn't 'standing' whilst flying, they instantly plummet towards the ground. Given this, I suppose it wouldn't be too awful to have a similar sprite cycling method.
3. blocked by walls. Isometric walls can block you from seeing things on the other side of them. Half-height walls could be the answer for current level, here.

[Tormy]

I'd like to add my input here.
To me, the Xcom style of Isometric would seem to be the way to go for the following reasons.
First point: Doors are always visible.
Second point: It's fairly easy to rotate isometric, in terms of 90 degree rotations.
Third point: It's built to allow viewing of multiple z levels. At the moment, DF allows you to view say... 2 tiles down, by fading the terrain colours into dots. With Isometric we should be able to do that much inside and far more outside.
Fourth point: Xcom had selection boxes which boxed down to the ground level directly below in order to show height. This would be ideal.
Fifth point: creatures could be somewhat larger than their tile's borders, allowing a more interesting display.

Now, the issues in regards to this.
1. Interface. The controls are going to need to be made intuitive here. More mouse support would really go a long way to help here.
2. Display. Multiple creatures on the same tile. This isn't an issue for flying creatures, since only one creature can be standing on a given tile. If a creature isn't 'standing' whilst flying, they instantly plummet towards the ground. Given this, I suppose it wouldn't be too awful to have a similar sprite cycling method.
3. blocked by walls. Isometric walls can block you from seeing things on the other side of them. Half-height walls could be the answer for current level, here.

The isometric idea has been dismissed already. [I've suggested the rotatable iso display method before, but read what Mike has posted about it: http://www.bay12games.com/forum/index.php?topic=27508.msg341677#msg341677 ... he is correct, even if we could rotate the camera, there would be problems.]

[Nadaka]

I'd like to add my input here.
To me, the Xcom style of Isometric would seem to be the way to go for the following reasons.
First point: Doors are always visible.
Second point: It's fairly easy to rotate isometric, in terms of 90 degree rotations.
Third point: It's built to allow viewing of multiple z levels. At the moment, DF allows you to view say... 2 tiles down, by fading the terrain colours into dots. With Isometric we should be able to do that much inside and far more outside.
Fourth point: Xcom had selection boxes which boxed down to the ground level directly below in order to show height. This would be ideal.
Fifth point: creatures could be somewhat larger than their tile's borders, allowing a more interesting display.

Now, the issues in regards to this.
1. Interface. The controls are going to need to be made intuitive here. More mouse support would really go a long way to help here.
2. Display. Multiple creatures on the same tile. This isn't an issue for flying creatures, since only one creature can be standing on a given tile. If a creature isn't 'standing' whilst flying, they instantly plummet towards the ground. Given this, I suppose it wouldn't be too awful to have a similar sprite cycling method.
3. blocked by walls. Isometric walls can block you from seeing things on the other side of them. Half-height walls could be the answer for current level, here.

The isometric idea has been dismissed already. [I've suggested the rotatable iso display method before, but read what Mike has posted about it: http://www.bay12games.com/forum/index.php?topic=27508.msg341677#msg341677 ... he is correct, even if we could rotate the camera, there would be problems.]

Unless you reduce the height of the wall, or use non-square shapes, as I demonstrated above.

[Impaler[WrG]]

Whereas with the rotated cavalier, in order to get the right graphic for any wall you have to check all 8 of its surrounding tiles.

That computation would not need to be done each rendering cycle, rather at the time the wall is constructed or tile mined out a check is made to the surrounding tiles and a the 'flavor' of wall graphic will be applied to the new construction.  The renderer will then use this value to fetch the appropriate graphic and display it.  Likewise deconstruction would trigger updates in the surrounding tiles.  Finishing of a terrain alterations is so rare any computation that can be put their instead of the render cycle will virtually disappear, I'm proposing the same method in the lighting arc thread as well.

As for your suggestion, I agree this would be a fine way to represent the corners that are movable along with pillars but contiguous walls should appear flat.

[Tormy]

Unless you reduce the height of the wall, or use non-square shapes, as I demonstrated above.

Nadaka, could you make a mockup using this octagonal method? Example: Create a medium sized room with walls around it.

[Mikademus]

Unless you reduce the height of the wall, or use non-square shapes, as I demonstrated above.

Nadaka, could you make a mockup using this octagonal method? Example: Create a medium sized room with walls around it.

Seconded, I too am intrigued.

[Andir]

I really like the Octagonal method... except:

Look at the bottom 2x3 grouping of octagons demonstrating ramp methods.  In DF, dwarfs can move diagonally.  It looks as though a dwarf would have to jump over a pit to move diagonal.

I love the idea though.  It makes a unique solution with it's own decorative flair.  The only way I know of to get around that right off the top of my head would be to make the walls of the octagon "sticky" so that they automatically draw in the gap if passage is possible.  This could be done with edge checking.  (if 3 of the 4 tiles at that intersection are walls, render the square.  If it's less, leave the gap and maybe render the floor above.)

Sorry for multiple edits... but the thing I just thought about.  What material are the filler squares made of?  How would it render if I had a group of contructed walls near a rock wall.

[Nadaka]

Hmm... Now that it has been discussed a bit, using octagons does have its own set of limitations like all the other "fancy pseudo-3d" graphical display methods. I'll see if I can whip up a demonstration that addresses a demonstrates or addresses these issues. If those squares in the corners between octagons get filled in while walkable, it would mostly negate the advantage of being able to see most of a hidden tile behind a block.



I personally would rather not display the covered squares between pillars, at least not for the upper most displayed layer.