AURO: The Design Challenges of “Attacks”

We’ve been in a state of closed alpha for almost six months now, and in that time I have been really lucky to have the time I needed to nail down some gameplay stuff for AURO.  By the way: I have also posted this on our Kickstarter page, here.

The tactical skills, being a cornerstone of the gameplay, have always worked in the design, as a concept.  However, one of the big questions was:  What about attacks?  How should they work?  In the spirit of trying to take nothing for granted, I have posed the question, “maybe the player shouldn’t be able to attack at all, and just rely on disciplines.”  It was pretty clear that that wouldn’t work out.  The player needs some very easy, super-accessible way of defending himself that doesn’t require using a special ability.

On Health and Damage

The obvious, standard answer to what “attacks” should do is deal damage.  However, there’s an issue with that, regarding AURO.  Actually, I think it’s generally a sign of weak game design generally if attacking, something you’re doing extremely frequently in a game, has no effect except for “reducing the number of hit points the target has”, in a game where the goal is to kill enemies.  The reason that this is too direct.  It’s as though in Tetris, instead of having to get lines to get points, you could always just press “up” or something on a piece and trade it out for some points.

Directly reducing enemy hit points is not tactical, it’s like grabbing victory-points right out of the bag.  Games are interesting when you have to take an indirect route to the goal.

So, it was decided early on that  “health” wouldn’t really be a thing in AURO, at least not in the normal way.  Almost all monsters have one hit point, with stronger monsters having two.  Auro (the player) does have about 10 health points – but this was a deliberate decision.  We tried a ton of different numbers for how much health Auro should have:

18 Health:  This was too high, because players didn’t feel tension.

1 Health:  Definitely too low – takes the ambiguity out of decision-making.  Most decisions are clearly “checkmate” situations that you definitely don’t want to move into.  Will make the game feel like more of a puzzle than a game;  we want players to be able to “play from the hip” a little bit, making up for small mistakes with big ingenuity.

3 Health:  Still probably too low.  I want players to be able to sacrifice a few health points for the sake of pulling off a big combo.  Also, having this few hit points still doesn’t allow the Fire tree (which is based on self-damaging) to work.

Currently, we’re shooting for 10 HP, although that could change.  I think somewhere between 6 and 12 health will be what we settle on – that seems a good balance between the ambiguity and the tension that we want.

So what should attacks do?

We’ve tried so many approaches already for what attacks should do.  It’s been pretty clear since day one that what we’ve been doing for attacks is not right.  We’ve experimented with random critical hits, various amounts of damage, damage that’s conditional based on if the actor is against a wall or not, and various amounts of knockback (1, 2, random between 0 and 2, etc).

A couple of weeks ago, I was playing with some of the abilities, and it occurred to me.  Now, after sitting on the idea for a week or two, I am not only feeling really certain about this direction, but I almost feel silly for not having thought of it before.

Just as movement in AURO doesn’t kill or damage, but instead is about repositioning, attacks too should be the same way.  Attacking an actor should knock them back 2 tiles – “heavy” actors (noticably large ones) will be knocked back only one tile.  Also, you can use this to knock actors off of the side of the map, where they land in the water and are killed with a big splash (that’s right:  none of the monsters in AURO know how to swim).

In game design circles, this technique is referred to as "Thwacko Sploosho".

Thwacko Sploosho

Don’t worry, it won’t actually be called that.  But the idea that you can knock actors off the stage is really exciting to me.  My favorite level in Mario 64 is Hot Top Volcano, where you fight those bully guys who try to push you off the stage into the lava.  My favorite fighting game is Super Smash Brothers.  I’ve always been really intrigued by the addition of “ring-outs” in stuff like Virtua Fighter.  I think the reason that I’ve always liked ring-outs in almost every game is that it turns your position into a more dynamic and more tied-into-the-gameplay “health bar”.

Now, you might be concerned about balance (i.e., it’s too easy to just run through knocking everyone out of the level).  There are several elements I have planned to balance this with:

Some, perhaps as much as 50% of every randomly generated map will have “guardrails”, making it impossible to knock an actor out through that tile

Generally our levels will be a bit wider than was previously planned, meaning you’ll be less likely to just happen to be in a situation where monsters are killed by your attacks

Some monsters will only get knocked back for 1, and some won’t get knocked back at all

If there’s any kind of obstruction behind an actor, it will stop him (ice blocks, other monsters, etc)

Of course, this is all new ground we’re covering, so we might have to change other stuff in order to make this work.  There are also some other really interesting ramifications, particularly for the ICE and ELUDE skill trees.

For example, GALE can be used to get a HUGE combo now, if you line a bunch of monsters against a wall (which is possible through smart use of ICE WALLs and ICE FLOES) and then use it.  FLOE itself is also really useful, since you can channel actors into it and then just send them careening to their deaths over a long distance.  With ELUDE, you can use traps and jump to jump out of the level if you want.  Yes, AURO can fall off of the level.  Not with movement (for your playing convenience), but he can if he slides off a floe or by jumping.  If he falls off the level, though, he simply loses 1 or 2 HP and is brought to the start of the level.  However, in a bad situation, this can actually be worth doing!

So anyway, we hope you enjoyed this article, and please tell us if you have any thoughts/ideas/questions/concerns about the changes to Auro’s attacks.


keithburgun • 06/07/2012

Previous Post

Next Post


  1. embarrassing 06/07/2012 - 10:31 pm Reply

    Sounds like it could be fun! Are you going to keep the tiles/map dull/plain looking, or will they eventually get some nice looking graphics (like the sprites)?

    • keithburgun 06/07/2012 - 11:02 pm Reply

      Oh no, we’re definitely going to have awesome tilesets that will make you go “OH MAN !!!” We’ll have a preview of them up soon.

  2. cuc 06/08/2012 - 1:11 am Reply

    >Some, perhaps as much as 50% of every randomly generated map will have “guardrails”, making it impossible to knock an actor out through that tile

    Would it work if the guard rails on some maps have openings? Either randomized openings, or there are a fixed few types of guard rail patterns (each with different chance of appearing), from “no guard rail” to “small openings on each side” to “no opening”.

    Then again, the entire map generation algorithm would have to take this into account.

    • keithburgun 06/08/2012 - 9:46 am Reply

      Oh, so like a “door” that opens and closes every X amount of turns?

      • cuc 06/08/2012 - 12:10 pm Reply

        I was only talking about guardrails having “doors” – either random (which would make them part of random map generator), or there are a few possible patterns, few enough that the player can easily memorize and expect.But doors that open at certain time intervals? Sounds cool!

  3. Lavaflyer 06/08/2012 - 9:21 am Reply

    It could be some work but the above comment would make more sense than having some maps entirely closed off to falling off

  4. Kdansky 06/11/2012 - 4:00 am Reply

    After pondering it for a while, I am sceptical on a single issue: It makes monster AI incredibly visible. If monsters very aggressively avoid borders, the game might become hard and frustrating. If they don’t, the monsters just look stupid, and require a ton of suspension of disbelief. I’ve seen this is current versions already: Sometimes a monster has two options to close distance, and one of those is by far inferior (such as a burning flame), but they seem to choose at random. Some things are very obvious, like flames or traps, others are not, such as blocking off another monster’s path, severely hampering their own faction, but not themselves. The second kind is easy to forgive, and the player feels clever about tricking monsters into interfering with each other. The first kind just feels silly: Run backwards long enough, and at some point the monster will make a really bad decision (such as walking through fire). Any thoughts?

  5. keithburgun 06/11/2012 - 8:29 am Reply

    >require a ton of suspension of disbelief.

    Not sure what you mean here. Actually, I’m not sure what you mean, overall. I don’t see why the knock-back issue makes monster AI more visible, or why that’s a problem if it does.

    • Kdansky 06/12/2012 - 3:20 am Reply

      Imagine a corridor three spaces wide, and both Auro and the enemy at one end. If the monster is clever, it will never bother to leave the central line. There is no reason ever to do so, because you can attack every space from there, and you are vastly safer than on an edge tile. Now assume Auro is on an edge tile next to the monster in the center. Auro can now walk backwards safely (keeping at the edge). Now the monster has a choice: Either enter Auro’s former tile, or stay in the center and *also* close the distance into melee range. If it does the first, it’s a stupid monster. If it does the latter, it might be impossible to actually win. If it does both on a random base, the game feels like you just wait until you get a lucky roll for the AI.

  6. Homepage
  7. sitsols 08/15/2012 - 3:44 am Reply

    I would like to see a demo for graphical (dynamic) titles.

    • keithburgun 08/15/2012 - 3:49 am Reply

      Can you explain what you mean?

Leave a Reply

Your email address will not be published / Required fields are marked *