Best of luck with GreenLight!
I will be getting my hands on this as soon as i can.
Thank you both for your greenlight votes! The game’s now entered the top 100 and is currently listed at #92, so we’ll see…
Dev blog update:
One of the nifty features of Airships is that you can create your own coat of arms. You can do this for your single-player empires, selecting a symbol (a charge) to give you a particular bonus. And you can design and register a coat of arms for your unique use in multiplayer. The system hasn’t really changed since the initial release with the exception of a few added charges, but with dev 6 - the update of prettiness - I want to improve on what’s there.
Currently, a coat of arms is defined by selecting a layout, a charge, and two or three colours. This leaves some coats of arms looking rather bare, though. In a quartered coat, only the top left quarter contains anything, whereas it would be much nicer and more realistic if all quarters could contain charges.
In the new system, each layout has between one and four slots that can be filled with different charges. I’m also adding a whole bunch of minor charges such as stars and wings that don’t have a bonus effect in the single-player game and can be added freely. You will be able to combine multiple major charges (say a scales and a lion) on your arms, too. In single-player, the most prominent major charge (the one closest to the top left) will be the one that provides the bonus.
This opens up a whole lot of options for much nicer-looking arms:
Your already-registered arms will stay unaffected by this change, of course, though you might want to look into upgrading them.
I’ve been hard at work on Airships dev 6, and have now fixed upon a release date for it, and for Airships on Steam: February 25. So unless anything goes majorly wrong in the next two weeks, that’s when you get the fancy new version of the game.
There’s still a fair list of things I want to improve before the release, mostly on the visual side. Right now, I’m working on making the strategic map screen nicer. The idea with that screen has always been to make it look like a desk you’re looking at, plotting your empire’s path. To make it look less bare, there’s going to be some pieces of paper at angles half-beneath the map, and probably some coffee stains.
Of course, the pieces of paper are an irresistible opportunity to do some worldbuilding:
Can you decipher the scrawled handwriting to discover the terrible secret of dragon-rearing?
An important part of any imperial presence is maintaining positive relations with the natives.
To be fair, airsailors will eat pretty much anything that’s not even more hardtack.
That new coat of arms mechanism adds so much more freedom to strategizing and decor! Well done indeed! But quick question about the dragons: will they be neutral and in battles randomly, or are they more like bossfights?
I’m glad this project still seems to be going strong. Looks very promising still.
It’s out! Airships version six, bringing new lighting, new heraldry, new guns, and weather!
It’s out on Steam too, which means this finely crafted link now brings you to the Steam store page.
So what’s next?
First and foremost, with new players coming in, there’s sure to be new bugs discovered. So my first priority is likely going to be putting out some bugfix patches. After that, I want to spend some time on game performance. The new fancy graphics take their toll on the graphics card, and while you can switch them off, it would be much nicer if everyone could enjoy them.
Improving performance means restructuring how airships are drawn from the ground up, which is why I didn’t want to risk it in the week going up to version 6. The core problem is that graphics card really hate it when you do lots of small draw calls, and currently that’s exactly what’s happening. You might think that a 2D game shouldn’t tax the GPU that much, but when you have hundreds of airsailors on-screen and make a separate draw call for each of their limbs - it’s quite bad.
The solution is to organize drawing into layers. Instead of drawing one ship, one crewmember, after the other, the game will draw all the back walls, then all the modules, then all the gun barrel, then all the crew limbs, and so on. Each of these layers can be a single draw call, collapsing hundreds of them into one.
OK, so that’s the immediate plan, but what about new features? What’s coming between now and the final release? Well, the development plan I laid out months ago is still more or less correct, though which feature comes in which version is subject to change. There are five major additions I have in mind:
Better troops
The boarding stuff is cool and works really well, but it took such a long time that I stopped working on it as soon as I had something to show. Which means that there’s some… odd missing bits: Crew stranded on the ground just stand there and are unable to move anywhere. Fixing that would pave the way for having ground troops fighting amongst one another and attacking buildings and ships. Most importantly, they can then be trod on by…
Landships
Okay, yes, the name of the game is “Airships”. But landships, war stompers, steam mechs: totally cool, right? And they’re basically the same as a ship or a building, they just have a different method of propulsion. The slightly hard bit here is getting their limbs and tracks to interact properly with the ground.
A more detailed strategic game
As it stands, the strategic mode is really very simple. It could be a lot more interesting. I’m still working on figuring how best to do this. There’s a lot I could add, but it’s important to pick additions that make the gameplay deeper rather than just more complicated. More detail on this in a later post.
Diesel!
I used to insist that Airships was totally a “Steam/Dieselpunk” game. This will finally become sort of true when diesel engines are introduced. These will be much more powerful and lightweight, and won’t require manual supply of fuel, but they’ll also be expensive and quite prone to catching fire in a bad way. The reference time period for Airships is about 1860 to 1940, so we’ll see a bunch of slightly more modern things cropping up.
Monsters!
Last but not least, the game needs monsters. There’s actually dragons in the game now, but only ever seen in the distance, flying past at high speed. Dragons, Turtledoves, Air Krakens, Sentient Floatweed, Giant Spiders, Mechanical Ducks! Some of them you must fight, others might be useful to your own side.
Stay tuned.
Wow! OP is a great artist, and this game is amazing! Why isn’t he getting slapped with medals?
Seems like theres a load of ideas that can be thrown into a game of this sort. Great job on the whole thing!
Having covered the kind of stompy war machines I want to make in the last blog post, let’s get into how to implement them.
Mostly, landships and airships can be treated the same. Buildings are technically a kind of immobile airship, so landships aren’t much of a stretch. The difference is that a landship is held up and propelled by tracks or legs, rather than through Suspendium and propellers.
The question is how the ship navigates changing terrain. What happens when there’s an upwards step? A real vehicle tilts when it drives up or down a slope, but the game engine can’t support rotation of ships without massive complication.
On a slope, I want to keep the body of the ship level, but raise it up gradually so it can adjust to the new terrain height. To do this, I add a suspension, a set of springs that raises the landship off the ground. The ship won’t bump against minor changes in terrain, and its vertical position gets smoothly adjusted. The springs aren’t real physical objects, so they won’t bump against the side of a hill. The tracks or legs the ship uses will get drawn on top.
This is what happens when the ship encounters a step upwards: As the springs move across the ground, the first one encounters the new height. It becomes more compressed and the additional force starts shifting the body of the landship upwards. As more springs move over the new area, the ship keeps on rising. Of course, if the hill is too steep, or the ship too heavy for the springs to push it up much, it will get stuck. But that’s fine, that’s realistic.
It also works when going down, though again the ship can get stuck if it doesn’t have enough clearance.
So let’s start implementing these springs, which will be attached to new tracks/legs modules. I start by creating a Spring class and putting in code so it can find the closest bit of ground. That gives me the length of the spring, and from that the upwards force the spring exerts, which can be easily plugged into the physics engine. Create a test module and put it on a building - and it sproings up!
Great! The next step is to actually introduce landships as a type of construction, beyond ships and buildings. Turns out that lots of things key off “is it a ship or a building”, which is too specific. What matters is “can it move” and “does it fly”, which nicely separates airships, landships and buildings in how they need to be treated.
Next time: Creating some basic landship modules that are able to both hold up and propel the ships.
Hey, you are getting famous 
I’ve found very small article about your game in todays 20min (a local free ‘news paper’):
http://m.imgur.com/TjxNBZG.png
Sorry, but I could not make any better picture with my phone.
Good stuff! You live in Zurich too, then? I wonder if we know each other…
I’m still working on Airships 6.3, and it’s shaping up to be a very pretty update. Most recently, I’ve made particles and explosions look much better and am now replacing the simple low-res backgrounds with more detailed ones. Also, Airships has an official wiki now!
Up until now, particles in Airships have been very simple: they were all coloured squares of various sizes, which works well for small things like sparks, but is a bit bland for smoke and explosions. In particular, I really wanted the smoke to be lit up dramatically by the flashes and fire of battle.
Getting particles hooked up to the lighting engine wasn’t that hard - all the shaders were already in place - but it took me a while to find a good look for the smoke. I first tried using a smooth look for the particles, and while it looked quite nice, it didn’t fit the look of the rest of the game.
Eventually, I ended up with square-ish particles with square-ish lighting maps, which I do think is an improvement:
Then I came across an article about how explosions are handled in Vlambeer’s Nuclear Throne, including a very nice-looking explosion animation:
Nuclear Throne explosion by Paul Veer. Reproduced with permission.
Looking at it in detail, you can see that after the initial flash, the explosion consists of a number of particles that don’t move all that much but steadily shrink and darken. I produced a similar effect in Airships by spawning a central light-emitting flash particle and a number of smoke particles surrounding it:
And the same effect but on a smaller scale for cannon blasts:
Having settled on a style for this kind of thing, I also re-added clouds, again using the lighting system for them:
As I wrote previously, the point of 6.3 is to really get the game looking pretty in a bid for mainstream success. This left one major sore point: the low-res and rather uninspired forest and mountain backgrounds. At this point, they were pretty much the only thing not using the new colour palette or lighting system, so obviously they had to go. This is what they’re getting replaced with:
A much larger set of modular backgrounds: hills, forests, towns, towers, windmills, and so on. This is a lot of pixeling, and it’s kept me pretty busy over the last few days, as each image needs to be first drawn and then given a bump map for the lighting to work. But the end result should be really quite good.
That’s all for now, except for this wiki thing: There’s now an official Airships wiki hosted on Gamepedia. Right now, content is still pretty sparse, but I’ve put in a bunch of entries. Head over there and contribute, and one day it should be an excellent resource for the game.
And join me next time, when I’ll be writing about the new modules getting added in 6.3.
Airships 6.3 adds lots of new decals to apply to your ships. The decal system’s been around for a long time, but after the initial set of options for putting your coat of arms onto your ships, it got neglected for a while. As part of the push for increased prettyness, I added a whole bunch more, mostly inspired by real-life aerial decorations.
Eyes and teeth: These were in fashion for decorating fighter planes in WW2 and ancient Greek triremes.
Gold leaf, because sometimes you want your airships to look fancy. None of the decals cost any money, which is a bit weird in the case of gold leaf, but roll with it.
Roundels, which turn out to be a fascinating side-branch of heraldry / vexillology (the study of flags). Much as with flags, there are a number of classic styles of roundels, which many countries follow, some variant styles that are popular, and some truly weird stuff where the designers clearly didn’t get the memo of “maybe make your war plane unambiguously identifiable”.
In flags, the most classic layout is probably the tricolor, a field evenly divided into three, horizontally or vertically. It’s a design associated with European nations. The roundel equivalent is a roundel made of three concentric circles of a solid color - pretty much the tricolor in polar coordinates!
European roundels: France, UK, Germany. All insignia images are public domain and sourced from Wikipedia.
Not that there’s a perfect match: the UK uses a classic tricolor roundel design but doesn’t have a tricolor flag, whereas Germany uses a kind of worrying throwback design instead of their tricolor.
Variations: Bangladesh, Equatorial Guinea, Switzerland
Fairly popular variations include different numbers of concentric circles. Bangladesh has two, Equatorial Guinea has four. Another popular option is putting a symbol into the center of the roundel. Switzerland’s got it pretty easy there.
Bad ideas: Belize, Syria, Yemen
And then there’s Belize, who just use their overcomplicated flag, and Syria, who picked a design near-indistinguishable from Yemen’s. Good job!
In Airships, I’ve stuck to two basic designs: a bi/tricolor roundel with colours extracted from your coat of arms, and one containing your main heraldic charge.
So, what’s next? I spent much of last week at A MAZE Berlin, showing around my game, and returned with my head full of ideas for improving the way it feels. Thing is, I may be a decent programmer and creator of game mechanics, but I have to observe people playing the game to notice all the rough edges in what I make. And other developers are way better at figuring out what makes a game feel fun. So now I have a long list of small things I can do to make the game subtly feel better, and it’s a list I urgently want to get to.
This means that rather than going straight to version 7 with its landships, there’s going to be a v6.4 and 6.5 incorporating these improvements. I know you’re all looking forward to your war stompers and spider mechs, but smoother UI and more satisfying screams and explosions are going to be a pretty nice thing to tide you over until they arrive.
No, I’m not adding a paintball mode to the game.
One of the major bits of feedback I got from showing my game at AMAZE Berlin last month was that there’s a lot of potential for more spectacle in the game: more satisfying crashes and explosions and things whizzing past. We like large amounts of kinetic energy, don’t we?
So Airships 6.4 concentrates on making that happen. Airships now crumble majestically instead of evaporating in a single flash. Bullets and cannon balls leave trails in the air, which both looks cool and helps communicate what’s going on.
With these updates, it became clear that there’s one major missing piece to the spectacles: module destruction looks really boring. The module just winks out of existence and leaves behind a small number of boring brown particles. This needs fixing. Modules should break apart into parts and shrapnel, spinning and arcing away. To make this work properly, I’ve created a new layer in the sprite sheet that splits each module sprite into the regions it should break up into.
This information is then going to be fed to a program that combines the original sprite sheet and this layer into a new sheet of shrapnel and a bunch of data that ties the shrapnel to each module appearance. Then, when a module is destroyed, the module still winks out of existence, but at the same time, the shrapnel bits are placed in their starting positions and given a velocity and spin, which should give the appearance of the module breaking up dramatically…
In other news, if you’re in Switzerland or Germany: I’ll be exhibiting Airships at Fantasy Basel on May 14 - 16. More details. (Airships isn’t listed there yet, but it will be present!)
Your game looks awesome, I love the art style.
I used to devour “Making Magic”, the game design column written by Mark Rosewater, then lead designer of Magic: The Gathering. I never got into playing Magic itself, but I found the inside view of game design and balancing interesting nevertheless. Probably the most famous column is Timmy, Johnny and Spike, where he introduces three archetypal player profiles that the game caters to.
Airships isn’t much like M:tG in its details, but there are certain similarities: it’s a 1v1 game that is played in two distinct stages of preparation and combat. In Magic, the preparation is the deck building, whereas in Airships it’s the ship design. Victory is secured by some combination of good preparation and being able to use your assets effectively. With that in mind, I think that Timmy, Johnny and Spike are applicable to my game. I’m going to first explain what each of these player types is about, and then how I’m working to accommodate the interests of each of them.
Player profiles form a loose categorization system for different motivations and styles of play. Not everyone falls neatly into one category, and beyond the Timmy/Johnny/Spike system, there’s a whole bunch of other categorizations that draw the lines in different places or are applicable to different kinds of game. TVTropes has a decent overview.
Timmy
This player likes things big, and dramatic. Sure, winning is nice, but spectacular wins are so much more satisfying than subtle ones. This is the guy who wants to take you out with the really dramatic finishing move, or crush you with the giant battleship.
Johnny
To Johnny, playing the game is a form of self-expression. What matters is cleverness and innovation. Sure, winning is nice, but winning in a clever way is so much more satisfying than doing it in an obvious fashion. He will try out the obtuse combos and weird tactics that take you by surprise.
Spike
Spike wants to win. He doesn’t care how he does it. Tactics don’t have to be clever, or exciting, they just have to work. He’s the one who will happily kill you by using a perfectly executed punch 57 times in a row.
(Note of course that these archetypes apply to all genders - their names are male, and so I’m using “he” throughout for grammar reasons.)
What does Airships currently do to cater to each of those archetypes, and what future improvements can I make?
Timmy
Big airships, big weapons, flamethrowers. Ramming! The most recent update, 6.4, was very much one for Timmy, adding more visual spectacle and visceral destruction. There’s still a bunch of things I want to try out for increased spectacle:
- Temporarily panning the camera for particularly large explosions.- Adding crew portraits to the shouts so you can see their faces, lit up by fire, screaming dramatically.- Better and more varied audio for all things smashy and explodey.
Johnny
The ship editor is Johnny’s domain, of course. What makes a game like Airships work long-term is a wide variety of viable strategies: high-altitude snipers, rammers, boarders, swarms of grenade ships, etc. What matters for Johnny is adding new options that expand the space of interesting things to do.
This is trickier than it sounds, because each added option needs to be situationally dominant rather than globally dominant or inferior. If I add a new gun that’s better in all situations than two existing weapons, I’ve actually shrunk the space of interesting designs. And if I add an option that’s just not very good, well, it just clutters up the game’s design.
So what’s definitely planned for Johnny, apart from a continual search for interesting new weapon concepts?
- Landships: Not as flexible as airships, but cheaper and a bit more durable.- Most likely, small auxillary craft. This is a concept that’s been worked out in some detail by people on the forum, and I think it’s a good one.- Diesel engines: More powerful, easier to operate, but more flammable and expensive.
Spike
Winning is what matters to Spike, and especially winning against other people. Airships has had online multiplayer for a long time now, but we haven’t reached a critical mass of players where you can play a match when you want to. The multiplayer calendar sees the occasional flicker of activity, but it’s kind of cumbersome. This may get better over time as more people discover the game, and old players revisit it at the time of major releases, but it’s clear that Spike needs some love, in the form of:
- Tournaments! Players are self-organizing one at the moment, but it’s something I want to build support into the game for.- Relatedly, a ranking system where players can compete for the top spot by winning multiplayer fights.- Practical improvements to the multiplayer calendar like being told further in advance that you have an upcoming match, and the ability to opt in to being told that there’s an unmatched slot available in the near future.
All of these improvements are going to come in over time in the next three versions of the game or so. As you can see, I have a vast quantity of things I want to do with the game to make it more fun for players, as well as finishing up the list of large game features. So stay tuned, let me know your feedback and ideas, and fight me in multiplayer some time!
I’ve been trying out a new thing: streaming the development of Airships on twitch.tv[/url]. So far, I’m really enjoying it for a whole bunch of reasons: It’s nice to interact with people. It’s helpful to think out loud. And, uh, I can’t procrastinate while I’m streaming. I can’t exactly check my Twitter account every few minutes in front of everyone watching.
I’m still very much learning the ropes on how to stream stuff - figuring out how to set up things to that I can interact with the twitch chat while working, getting better at pressing the right buttons in the streaming software, and so on. I also haven’t settled on a standard time to stream yet. Probably, the ideal point in time would be evening in the US, but that’s the middle of the night for me - so it’s going to be either US daytime or late night.
Today, I’m going to do a dev stream at 18:00 CET, which is 12:00 EST and 09:00 PST. So if you’re in Europe and want to see some late afternoon streaming, or in the US and see some morning gamedev, do tune in. If you’re lucky, one of the cats will be present as a guest star.
Probably, I will eventually settle on a regular time slot, but it’s still early days. Also, twitch makes it easy to upload the stream recordings to YouTube, which is what I’ve been doing. So even if you can’t see the live broadcast, you can watch the streams later.
Anyway, what’s been going on on these streams dev-wise? Lots of work on legs and leg movement for landships.
We’ve got four-legged ones:
Two-legged ones:
And the gait is starting to look more realistic:
Once all the leg stuff is fully operational, I’ll move on to tracks - which will likely be much easier - and then, better support for troops. So tune in if you like, or follow development on YouTube or through these dev blogs, which will of course continue.
I haven’t noticed this before, it looks awesome though! The art style is great. I’ll definitely try to catch one of your streams sometime!
The challenge with landships is that they require a much more detailed interaction between physical objects. Until now, the only interactions that mattered were collisions as well as buildings and grounded ships sitting motionless on the ground. Walking, on the other hand, is a pretty complex interplay of movements and forces, and so it’s taken me a few weeks to adjust things properly. A partial list of considerations:
- Landships need to move their legs, placing their feet in suitable spots that are both well-placed to hold up the ship where it’s going.
- They should only be able to move if their feet are actually touching the ground.
- They shouldn’t be able to balance on a set of feet that’s way off-center.
- They should correctly estimate whether they can safely traverse terrain, and emergency-stop only when needed.
- Feet should crush foliage and bushes, and in cases of heavy impact, even other ships, buildings and soil.
As I previously mentioned, the fact that landships can’t tilt means that they have quite a bit of difficulty navigating slopes. I’ve generally compensated for this by making their legs and tracks pretty huge, but that in turn has led to some weird behaviour like giant war stompers gracefully tip-toeing across the treetops. Finally, though, it’s coming together.
Beyond leg physics, I’ve also been improving the particle system, allowing particles to stick to ships and other places as decals. Explosions, impacts, mud and blood splatters can now all dirty up your ship. Did I say blood splatters? Oh yes, trampling enemy troops is now a thing that works:
Take that, tiny Air Grenadier!
Another kind of particle I’ve added is snow, and to make that more complete, there’s now snowy landscapes:
So what’s next for dev 7? With legs pretty much working now, it’s time to pay some attention to tracked vehicles. Finally, I want to instill troops with some basic ideas about navigating the ground. This lets them board buildings by simply dropping to the ground and running in, abandon crashed ships, and reinforce their comrades elsewhere.
If they don’t get squashed beforehand, that is.
So I’m just about finished with work on tracked landships. Compared to ones with legs, they’re faster, sturdier, but less able to deal with rugged terrain.
In terms of game physics, they’re held up by a set of springs much like legged ships - essentially a suspension. Wheels and tracks are drawn on top of this. The wheels are simple to draw, but the tracks are made up of lots of individual links. They alternate between big links with surfaces that touch the ground and slightly smaller ones. You can see what this looks like in real life on this image of a Japanese tank I used for reference.
Now the Hard Bit was calculating the shape along which the links are drawn. The track needs to run along wheels and between them. It boiled down the track being composed of two strictly alternating types of sections.
Curves following a wheel between two angles.
Straight lines that connect two wheels.
Curved sections can be of length zero, which happens when three wheels in a row are perfectly in line. Finally, there needs to be some special casing for when three wheels in a row form a concave shape, as the tangential lines connecting wheel 1-2 and 2-3 would cross in that case.
Once I’d coded that all up, I tried it out with debug graphics, then spent some time fiddling with things so that the rotation of the wheels gets correctly fed into the offset of the link drawing. Or to put it simply: making the tracks move!
The results look pretty good, I think.
Now as you may know, the heraldry system in the single-player campaign lets you select a heraldic charge (a symbol) for your side’s coat of arms (your flag or “logo”) that gives you some bonus. For example, the scales give you extra income from cities, while the lion improves your troops’ abilities at boarding combat. With such a major new feature, there of course have to be some new bonuses. There are going to be three. I will unveil the first one now, the second in the next blog post - and the third one when v7 comes out.
Boar’s Head Erased +20% HP for tracked landships
So what’s next? I’m just working on making the crew smarter so they can move across the ground, between ships and floating rocks, enabling them to do things like transfer between ships. Beyond that, there’s about two dozen bug fixes and then it’s time for dev7.