Project C – Journal Entry Postmortem

Updating everything ignored and final reflection on what went right and wrong.

Date: 18.6.2021
Invested Hours: 3 hours 
Activity: Crating postmortem and reflecting on decisions...
Outcome:  Plan was updated, and FinalDesign was uploaded along with this journal entry.

First of all I went back and actualized PLAN, I really hate strict time tracking as it gets annoying to fill, if you do not write it immediately as you do it.  Also it does in my opinion not fit very well into development to track every minute spent, as it prevents from fixing or improving small parts here and there, which I did pretty much all the time during the development.

Anyway, PLAN shows some valuable data. First of all first two weeks were actually spent on designing UI and learning how to work with unity little more in depth. So around half of the time spent on project was doing very little productive work. At the start of this week,  only functionality was Adventurer spawning item and moving item around. 

This brings me to first point I will reflect on: User Interface
In my opinion UNITY tools for UI are trash and archaic. Docking is absolutely dump and it relation to Canvas  gave me just more headaches ...
As I work several years on frontend with WPF and XAML based UI, even through its more complicated and simple task are sometimes pain. They all can be solved without from code. With UNITY you can spent hours moving around objects, just to have everything self destruct itself on any resize... So if I would have to work with UNITY again I would most likely grab something like NoesisGUI, which would bring me comfort of XAML, so I can avoid dealing with offsets and anchors. And I am not even going to mention how UNITY does not support SVG for UI...

In terms of Assets and designing game in editor:

Choosing assets actually took a lot of time (filtered over 5000 assets), and then I defined by hand over 300 items.
This made me realize how dump UNITY is when you change something in code, like enum value. It will nicely remove every place enum was assigned in Editor as it stores value of enum, instead of it name.. There might be some guidelines how to change and refactor code. But in my opinion as software developer it's not tolerable that I can't change something in code without absolutely destroying everything I set in editor. This basically means that If I would decide to work on game more and do some refactoring, or bigger changes I might as well do everything from scratch as UNITY will happily not preserve anything at first opportunity...
This is actually one of the reasons, why I slowly moved more and more into code, so If I decide to change something, it will not blow up...

Gameplay and one I want to reflect most on.
Game is surprisingly more fun then I thought it would be. It's really simple puzzle I guess , maybe  from small part a strategy ?
Gameplay loop is simple to understand as player is trying to prevent adventurers from filling  Loot area. Through it's little more complex as Player needs to quickly decide, which items to sell, store and upgrade. This makes game more dynamic than it looks on first sight. Finally player must reach final rank and complete at least once final quest to achieve victory.

 Even through I did not implement certain mechanics or form of some mechanics changed over the development. I feel like the current game is solid foundation, that could be improved with more polishing and adding some sounds and music. (and fixing UI scaling)

Few hours before finishing I felt, that it will either be too easy or too hard. Through game is not very well balanced and lootboxes serve as better loot generator than adventurers. In early minutes game is slow and get gradually faster as more adventurers spawn. Winning game is possible with right strategy. Currently storing only items of certain rarity and type, while keeping items for repeatable quest series can bring player to achieve victory.  So even if balance is still rough,  game is playable and it shows rough edges only near the end, when player aims for maximizing income.

In the end I am proud that if feels like a game. I learnt a lot and obtained many new reasons for hating Unity.  I could probably provide more in depth about certain decisions, but they were really not that deep, it was generally done because it felt easier to implement or visually looked better than other variant. Gameplay was finished only today, so most decisions were not based around Gameplay. 
UI elements moved mainly because I had problem to make UI look ok on different aspects and in the end I ended up  with UI that feels too small.
Addition of Town  felt natural as I wanted to include more gameplay elements as I was making UI. Town nicely filled that and added more mechanical depth to game, which was previously lacking.