Skeleton Lords


Skeleton Lords is a game which I wanted to do a few years back. It's based on Warlords for the Amiga, and it's an army/empire building TBS. There are no scenarios, instead the entire campaign is just one big map with all of the factions duking out. I like uninterrupted gameplay like that, because your actions and progress carry over from place to place. With scenarios, you often build a huge army only to start on the next map with very little. IIRC, Homeworld didn't do this though.



World map

The world map continent layout is a paraphrase of ours. Some thought have been put into the placement and adjacency of the factions. Exactly what the thoughts were, I have forgot.

After a while I decided to expand the map from a square to a larger rectangle, so it could be wrapped around an entire globe. I don't mind dead areas (like desers). Even distribution makes things unbelievable and boring.

As for the gameplay, I was thinking of... using some kind of resource or city nodes. Not terribly innovative, but a proven concept. The map is hexagonal and upgraded cities will expand outwards concentrically, making use of the resource types of nearby cells. This makes each city unique in terms of production, as the surrounding terrain varies.

Also, the different factions are 'fueled' by different things, so they might be at a disadvantage in certain regions. Since I have over 40 factions, I can afford making some unbalanced (weaker) ones, and factions with very odd behavior, such as weak humans who turn into powerful werewolves cyclically. Some units prefer to fight in certain terrain, and are handicapped elsewhere. A long pike or lance is probably less useful in a dense jungle, for example.

There will also be temples, sacred sites and ruins which yield artifacts and bonus army units.

World map


I went nuts and drew up some 40 factions which inhabit a large (perhaps hexagonal) world map. I tried to include all of the most common fantasy species (which I like). Blue text color indicates a major faction, and yellow a minor one.


Humans lose a lot of resources and time on bureaucracy, religious paranoia and such things. However, they are highly adaptive. This is reflected in their troops as well as in their ability to terraform (ruthlessly exploiting natural resources).

The Longswords are typical white knights who oppress the peasants. Their armies are a mix of elite knights and weak peasants.

The Barbarians are tall and muscular. Skilled in hand to hand combat, but lacks a sense of tactics. Does not know how to sack cities, only plunder. They are kind of haphazard, lacking organization, so they can not control what units they build.



The Valkyries are all female humanoids. The Yaksi uses a gray clay bodycream to give them a statue-like appearance.

Valkyries, amazons


The Summoners have 4 or so unique summoner characters which they are reliant upon. Once those are dead all demons become wild and it’s practically over for the summoner player.

Undead, skeletons, vampires


Yeah, there's some paraphrasing going on here.



The Gnomes sometimes just slack. In fact, they’re not terribly interested in war. They have random periods (turns) when they get inspired to build stuff.

Dwarfs and gnomes


John Bauer is a big inspiration here. The trolls are ancient creatures possessing the wisdoms of the old world. They're also jaded and unwilling to keep up with the new world. Because of this they are close to extinction. Many troll have already left the world by turning themselves into stone. A form of transcendence or mass suicide? Who knows? The remaining ones might have other plans.



Orks fight among themselves and don’t follow orders too well. Sometimes they wander off to wage their own personal war. It's difficult to control the Orks, but they make up for it in numbers.

Orks and imps.

Beastly humanoids

Beastly humanoids.




Ducks and giants

Wildlife, Beasts and Guardians


The story (and possibly gameplay) needs to explain how the many factions could have coexisted for a long time without annihilating each other. Maybe an ’Age of Plenty’ begins, suddenly giving the factions resources to expand. Maybe a powerful artefact appears, forcing the factions to crawl out from under their rocks.

Setting up a game

To what extent should the player be restricted in changing the design of the game universe? I don't know really. Allowing the player to play on random maps with arbitrary faction selections and starting conditions could make the universe feel less solid. It could also be fun and increase replay value. However, I'd like to think that I've designed such a large world that there's enough variation as it is. I want to create immersion, and allow the player to explore the same game from different perspectives (factions).

One solution could be to use text files to set up the game rather than a very convenient ingame setup menu. Editing a text file make you feel more like you're changing the game than using an ingame setup menu. Game difficulty / AI settings should be in an ingame setup menu though.

Ally traits - match making

I'm not sure if factions can ally, but it would be possible to device a system where each faction has a number of traits which are used in match making:

Troops and generals

I haven't settled on anything regarding heroes, generals and troop construction, or even if you should see stacks of troop or just generals on the map. Thinking about the different possibilities, this one seems the most appealing atm.:

There's only a few movement classes per race, such as foot soldier, riders, flying. While this puts limits on variation, it's quite possible that you actually need to be simple and clear with movement, and that finer differences between units are of needlessly fine resolution and even confusing. Units of the same movement type can still get different terrain bonuses in terms of combat, like pike-men having troubles in forests or rocky terrain.

So, all variants of cave troll foot soldiers would march slowly at the same pace and with the same terrain penalties for movement (e.g. mountains and rocky terrain doesn't slow them down). All foot soldiers can be grouped up with each under under a general, and you can't mix riders with foot soldiers.

On the map you'd see a general, carrying a banner indicating, say a medium sized group of foot soldiers. The exact composition of the force could actually be invisible, for better or worse. A form of fog of war. Maybe scout units could allow you to click on nearby enemy stacks and see the exact composition.

Upkeep? I don't know. Perhaps some troops, like the barbarians, could live off the land, whilst other troops like knights require upkeep.

As for balance, I'll just wing it. Master of Orion 1 is 'unbalanced' racially, with the Psilons being a huge threat and Mrrshans who are much less of a threat. I think it's more interesting with unbalanced games. I will try to balance the few major factions against each other though, so either of them can win.

Map cells

  + + +
 + * * +
+ * o * +
 + * * +
  + + +

I can think of a few cell properties / flags which would be useful when it comes to unit combat/movement advantages/penalties and such. Snow bunnies tolerate cold, whilst desert people have adapted to hotter climates. A pike man will prefer open terrain. Some units may hide in chaotic terrain, etc. Mountain trolls may have adapted to mountains and rocky terrain.

Most tiles could be made in 3 variants and there could be 3 superficial styles as well, to make certain regions distinct. For example, a region could have a certain geology which makes the mountain look a certain way. Another example is forests, which, while in the same temperate region, could be dominated by different types of trees.

The style of the cities and castles vary by faction. The style of ruins and temples vary by region (pyramids in the desert, stonehenge up north, etc). The roads and agricultural areas are artificial. Both offer a movement bonus for most factions/units. None can exist in the icy areas (you don't grow crops or build roads on top of an ice sheet with constant snow fall on it).

. Dead Plains Wetlands Forest Rocky Mountains
-10 Ice field None None None
+4 Blacklands Tundra Swamp Mostly pine trees Glacial erratic Flattened by inland ice
+20 Wasteland Grasslands Swamp Mostly leaf trees Sometimes snow tipped
+40 Desert Savannah Swamp Palm trees, Rain forest

Resources around the cities and castles

Each terrain cell has a few resource values which a nearby city or castle can exploit.

Resource gathering

Forests Elves might be skilled hunters and get more resources from adjacent forest tiles. Humans would prefer plains and agriculture. Maybe my Asai (Asians) are skilled fishermen. Races used to hot or cold climates may be somewhat restricted because they have troubles adapting to the opposite climate. Some resource types requires an initial investment of time and funds before they can be exploited.

I liked the system used in Colonial Conquest, where you could put little men into the surrounding terrain, and they'd grab resources every turn. Then you could build, say, a mining machine or a green house on a map cell, and the men working there would be much more effective.

Colonial Conquest

Construction of units

I haven't elaborated much on this yet. Batch production could be faster and/or cheaper. The drawback is that more resources has to be tied up. Some special units can be found in the wild or at certain sites.


I haven't thought much about heroes or special characters. I'd like to use Generals who gives units under their command certain light bonuses. I don't like heroes much because... I'd rather use troops than a single pimped out ubermench. Still, it's fun to have a special character who can run around and have... some watery tart threw a sword at them.

Fun with hexagons

Warlords is a TBS with a square grid, I think. I'm thinking of using a hexagonal solution, since: Having six clean movement directions seems interesting. I'm doing land masses, not square rooms. I wanted to try it.

Also, structures (buildings, trees) often look better at an angle. I know you can cheat a bit with perspective and do side view and top view at the same time, but I want to try and go for a more accurate perspective (no diminishing point though). Ideally the view angle should not be too top down (can't show height) or too side view (hard to estimate distances)

After much experimenting, I came up with this practical solution:

hexagons on a grid, in perspective

I found the polygon/hex tool in PS, and made a big hexagon. Hexagons, if rotated right, have 2 straight sides and 4 angled sides. Most games do the angled sides as 2:1 slopes because it looks good. While clean, these 2:1 slopes are not true to the actual angles of hexagons. But this doesn't matter as much to me since I won't be using a top down view. I can choose slope by squeezing the hexagon together using the transformation / scaling tool. I turned the grid on in Photoshop and came up with a few variants which have clean slopes.

pixel hexagons, in perspective

(There might be errors here. I get confused and make faulty assumptions a lot.)

What's left after this step, is drawing a polygon at a much smaller scale with clean edges which meet up when tiled. I turned out that some sizes were impossible to make, using the edge matchup I had preferred. I ended up with a 44w * 29h tile. It's close enough to the 6w * 4h proportion of my big one. It's viewed at 35.1 degrees, and reasonably equilateral (well, it's in perspective, but you know what I mean).

So, why bother with all this?

It's good to be informed. You have to know the rules to break them, etc. Now if I break something I can know which direction I'm breaking it in.

Also, chances are that since I know the view angle, I can draw buildings and stuff which doesn't feel disjointed with the perspective of the ground. If I feel the need to, I can use a 3D program to help me make complex stuff like castles or skyscrapers.

Completely 'equilateral' hexagons ensures... that all directions are fair... I mean, look fair. The graphics are just a representation of some array of course. It always bothered me that the (text) 'tiles' in rogue-likes are tall, because it makes it difficult to judge distance. My hexagons are wide of course, but we're used to seeing compressed perspective along the ground plane like that. If the perspective and proportions are proper, then it's easier for the brain to... read what's going on, I'd say.

The clean slope is mostly important because it's... clean. The tiling feels clean and edge lines (and maybe selections) will look nice and straight. It also makes the masking job structured and easy.

The art process

I've previously done a few isometrical graphics sets, and developed a few techniques which were useful for this project. You may download the PSDs (right click save):


First I paint at five times the size of the actual graphics. Meaning, I paint/draw at 100% and scale down to 20%. Actually, I have made a macro which scales down the image, sharpens, and fades that sharpen about 50-70%, I don't remember the exact value. I fade the sharpen a bit because the full effect gives a gritty result.

Also, I could downscale the terrain art to an even smaller size if I need to do several zoom levels

I just eyeball the tilt/perpective of the tower and trees here. If I need to, I could consult a box model, because I've calculated the angle of view.

Here's a breakdown of the layers which I often use:

The drawback with this method is that it's cumbersome to make a global change affecting all layers. Right now I use Edit / Transformation / Again if I need to scale a head or so move stuff around.

My fill patterns for this little experiment. I made them pale so the line art on top will show clearly.

The mask and edge layers

Hexagon terrain test

I make two layers, on for the mask and one for the edges. This way I can tweak my large scale image, the terrain art, whenever I need to. Then I can pop the down scaled version in underneath those layers, so I don't have to re-mask everything.

The mask is made by filling an empty layer with my hexagon pattern. I make it transparent so I can see my terrain underneath. Then I paint bucket the empty hexagons with the mask color. The only thing left after that is setting a small pixel brush to 'clear' and erase the mask where there's terrain overlapping, e.g. a tree or mountain sticking up.

Pixel art?

I do no pixel level optimization other than perhaps tidying up some odd pixel here and there. I put these in the mask layer, so I can paste in a new version of my terrain art underneath. Again, the idea is that this way I won't have to redo any work if I need to edit the terrain art.


I feel that a strategy game which relies on grid based movement and abstracted strategy elements should be clear about tile borders. This is why I decided to not do stuff like half-forests or gradients of desert and grass on the same tile. A tile should not be ambiguous about it's strategically relevant content.

Anyways, the edges are just a 1px white edge going around the inside of the hexagon edge. The tile edge graphic could probably be made into a convenient fill pattern, but move-duplicating it works too. I erase the edges where there was stuff sticking up (trees and mountains). I'm thinking that perhaps the edges could be separate graphical objects. This way it could be used for tile highlighting and such.

Beaches and roads

I really don't wanna draw 2^6 edges for beaches, rivers and roads. Thinking about a pie-slice system, an on the fly visual effect (based on adjacency) rather than an actual tile in the array. An overlapping system could also work.


Like stated elsewhere, I don't think the tiles need to blur into each other. I think it's possible to design the map so you minimize clashing. For example, you could place mountains between the desert and grasslands so they serve as a separator. Just putting desert in the middle of the grasslands might look odd of course. So, it comes down to map design.

The reason I bother with the beaches is because it looks nice and I'll have lots of sea-land adjacency. Beach tiles are not something that you'll stand on a lot, and most of them will clearly look wet anyways. They'll basically be 'shallow water' and the exact configuration of the beach doesn't affect that.

Rough ideas

Misc ideas, rough pencil thumbnails.

Concept art by Arne Niklas Jansson, 2006-2010.