Dynasty Survivors: gameplay test 1

The goal of this project is to learn how to make a game using an entity-component-system (ECS) framework, vs object oriented.  I ended up on the bullet heaven genre (ie Vampire Survivors), because ECS frameworks are good at handling many objects at once, and it feels like a fairly compatible paradigm to start learning with.  

The main twist I I wanted to add is the the concept of an army that follows you around.  Theming is ambiguous, but having 100+ units firing on your behalf feels like it could be visually interesting & offer a nice variety of gameplay.  

This current build has most of the basic mechanics around a single follower that has 4 evolution stages (the 4th is highly overpowered, getting to it is what I consider the end of a run).  I like the overall feel of the 90-120 second run.  I have some cleanup to do around the UI and similar, then after that need to figure out what to add to make a run have more variety. Ideas are:

* Add a card-like system for minions you get.  If you're evolving 3+ different minions up to level 4, if split equally we'd be looking at 5 minutes until all 3 have a high tier unit.

* Add pop up UI for one-off boosts.  One-off minions that boost those around themselves or have their own powerful attacks seem feasible.

* Need a variety of enemies.  At this point most of an enemy setup lives in a JSON file... but from what I can tell this genre has dozens of variants per level?

Unpublished approximate build out (didn't realize itch has a nice DevLog feature) to get to this point:

1. Vampire survivors style controls attached to single entity that fires against random spawning entities

2. Added a "charge" mechanic that adds a new follower every x seconds that also fires.

3. Dropped firing mechanic from primary player controlled entity.

4. Added second minion (melee) that was randomly added in place of basic.

5. Added "merge" mechanic to minions on button press.  When merging, they became a static tower.  I liked the feeling of placing down a tower-defense like field.

8. Added concept of increasing difficulty curve (spawned enemies faster).  Spent a lot of time trying to balance to fun.

9.  Dropped "merge" button into automatic.  The binary many entities following vs static tower never felt quite right regardless of how much balancing I did.  In compensation, added 2 more levels of merge.

10.  Spent a lot of time trying to make a 2 minute experience feel good.   Hit the point of feeling "powerful", but due to the linear scaling of the army (vs exponentially more expensive levels in most gams), never hit over 60 seconds.  But the power creep up felt good.

11.  Added in pickups that drop from enemies that cause a minion to spawn.  This naturally slowed the game down and let the player's power vary around a play curve.  Play tested w/ a friend... since I was still sticking around 60-90seconds to max out, they were closer to 2-3 minutes though.

11.  Reworked the curve to contain the idea of a "threat level".  Initially this was "how much total HP should be on field", which yielded a mechanism that spawned more enemies in the faster you killed them.  Which felt like a reward for being efficient.

12.  Added a new enemy type.  Had to rework "threat level" to be a point assigned to enemies.  New enemy type was slow moving but high HP... which meant one very slow target removed 10+ small faster enemies.  Ratio eventually become 1:3, even thought HP was closer to 1:10.

13.  Lot more balancing around pickup radiuses & small tweaks.