Live Chat: Discord.
Get the latest builds: Patreon.
Get Mods here: Reddit.
Public Version for game under development (v4.9.2): Here.
Download for Heroine Rumble is at the Heroine Rumble page.

Saturday, September 30, 2017

Some new Combat Mechanics

One flaw in the combat system that I want to address is the difficulty in fighting many opponents at once. While arguably that it is realistic and it should be difficult, it doesn't make for a fun game play experience. It is certainly do-able by walking backwards and keep swinging - effectively turning a 1 v many into many 1 v 1s, but I want to make it also possible for the player to hold his/her ground. (I am also not going to penalize the backward strategy at the moment)

So I have implemented the following features to help 1 versus many fights:

a) Guards lingers for a bit and stays active for a while when staggered. Suppose if you are guarding top versus 2 opponents, one attacking from right and another from top. If the right swing hit you first, and then the top attack, you will only take damage from the right swing, even tho your character is currently in the staggered animation from the hit coming from right. With the old system, you would have taken damage from both hits.

b) Guard is now usable while staggered. So if a new blow comes from the right while you are staggered, you can quickly guard right and block the attack. However, you may not parry using staggered-guards.

c) Super/Hyper Armor. (Internally known as TP - Tenacity Points). Basically, there is an additional hidden meter that can sometimes prevent your action from being interrupted, while damage applies as usual. This have a dual purpose so that slow, heavy weapons have their place, and to help prevent stun locks.

Whenever a character is staggered, they will gain a bonus to TP based on the severity of previous staggers inflicted on them, making them more and more difficult to stagger in a short time period.

d) Cleaves. (Already announced before but might as well mention it again) Attacks hit multiple opponents with each swing. Weapon and character skill dependent.

e) Stat advantage. (Again, announced before) The Hero category, which the Player is one of, are much stronger individually than the Minions category, but far less numerous.

In the upcoming demo #1, you'll find that the enemies have very little stamina - they can only do a couple attacks and then is forced to back off. Meanwhile, the Player can swing her sword (that is both longer and faster than the Minion's axe) with much more vigor.

Wednesday, September 27, 2017

Bandit Set

Decided to create a new armor set for the enemies. Here is a screenshot.

"You are in the wrong neighborhood, lass."

Monday, September 18, 2017

Outfits and the Plan Ahead

Been working on solving various animation problems arising from independent torso and leg movement. Should be able to wrap it up this week together with the first iteration of combat. But somewhat tired of working with barely presentable in-dev assets so the next step is to (re)introduce outfits and bring the general graphics quality to be on par with Heroine Rumble.

I am a little disappointed with Heroine Rumble's RNG-based outfit selection system. It is certainly better than nothing but leaves a lot to be desired. So instead, the next game will feature a fully functional inventory and equipment system. You will be able to pick and choose individual outfit pieces and wear whichever combo you like.

I plan to have at least the following "slots" for customization:
  • Torso/Top 
  • Arm/Hands 
  • Legging/Bottom 
  • Feet/Boot 
  • Skirt/Dress 
  • + 2-4 accessory/attachment slots for things like wings, hairband, etc. 

This require some changes to the underlying data structure + rendering system so its not a simple "copy and paste" job. But I don't foresee the base system, without the actual user inventory/equipment part, to take very long.

After the base outfit system is done + other touch-ups, I plan to start releasing playable "tech-demo" "not even alpha" "for enthusiast" builds for Patreons. Starting with something very simple, like here are X wandering enemies on the map, and you can fight them, but not much else. And build from there.

The current ETA for first such a build is early-mid October, and then likely monthly builds from there onwards. Again, the first couple builds are going to be very very rough tech demos and they are certainly "for enthusiasts" only.

Sunday, September 10, 2017

Combat, part III

The combat mechanics are almost finished now. There are still quite a few bugs and animation problems that need to be sorted out, but enough core systems are functional now to create an overview of how it works.
So like I mentioned before, the game will be based on Mount and Blade melee combat system. I like to start off by mimicking the functionality of other successful titles. This is because, well, those games probably did somethings right and and there are merits to their setup, even if it's not obvious at first.
First thing first is the camera. The camera is controlled via mouse movement as what you expect in a 3rd person camera game. A set of direction keys (WASD) controls the character movement. The character can run and walk (guard):
The core of the combat system is very simple. There is attack (left mouse button) and block (right mouse button). There are 3 different attack directions and 3 guard directions. To pick the direction of block, swipe the mouse towards that direction and then hitting the block key(right mouse button).
(Attacks works differently than what the following gif shows, it will always come from the same direction of block rather than be controlled by mouse movement)
The purpose of these directions is that blocks only work if you block the correct attack direction:
Some more "advanced" stuff: 
First, you can move during attack/block. 
Secondly, you may cancel an attack via guard before the attack is fully committed.  This can be used to "feint" your opponent but since online multiplayer is a long, long distance away its more of a "oh shit" thing:
Another concept is "parries." If you time the guard to be as close as possible as the time when the attack swing will connect, there is a greater chance of interrupting the swing. The idea is that there is a tradeoff between safety (guard as soon as possible) vs payoff (guard as late as possible, but possibly missing the block)
And putting everything together:
But who am I kidding, the player is so overpowered against the AI that you can simply do this: