Closer look at WildLands

I'm going to begin 2023 with a new process to document better and share my experiences with smaller projects, I hope that with this I can grow to better understand and make choices that help me grow in the right direction.

Personally, when it comes to writing anything or expressing myself I struggle greatly and so to keep me from rambling on I'll layout my points for discussion. As a side note; while I have had my fair share of experience with chat GPT, this post will be (and probably for the worse) written wholey by myself.

1. General feedback for Wildlands

Post-motems can be a strange look back on journey you have taken within a certain timespan, and when you put some much time and effort into something, you generally want others to see what you wanted the project to be. A vision in your mind is hard to manifest into something digestable for others within a short time frame.

When it comes to feedback there were comments about the atmosphere in Wildlands provided. The word "world" (used loosely in this respect) felt alive and was supported with great audio & visual goodness. However a lot of bugs still plagued the game and consistent issues such as animation freezing and doors not opening. Even wood piles could not be picked up without being locked into place. There were also valid points of the game being to slow when having to traverse & search the environment. Secrets were littered sparingly though generally people that took the time to find them enjoyed them and would have liked to see more secrets scattered to reward the time taken traversing.

Overall when it comes to praise, the glory goes to presentation which has (and always been the case) carried the game into a respectable 20th position out of 115 submissions posted, that places Wildlands in the 23rd percentile bracket for overall quality. Wildlands was a new experience for me in a lot of ways and while it may not have lived up to my expectation in its percieved results, I am quite proud of what I managed to achieve within 72h set limit.

When it comes to analytical data I get all giddy and excited, I like to pride myself on organisation, keeping track of statistics that I generally care about, as you can see below from previous gamejams:

I find great pleasure in looking at statistics to better improve myself and see how the data reflects the results and while it may not really drive my final decisions. Score*I feel is the most consistent stat in isolation that confirms if I improved generally (there are multitudes of things I consider, and I believe 1 or a handful of stats will never paint a real picture of "improvement") With this in mind, statistically Wildlands failed in a majority of ways and I will cover below what I feel could have been changed to improve player experiences.

2. What I feel was a success

When it comes to defining success, one must really ask: "What's important to you?" - and that really is the question. For me success could be defined as finacial or personal growth and in this case, growth is always the way.

If I learn or do *anything* new in a gamejam that I haven't done before, I believe I have grown and the jam was a success and worth while of my time, because of this, despite my placement in statistical data, I can always come out of the tunnel at the end and know I learned something new.

WildLands hit a perfect tone for atmosphere, something of which I feel I have captured extremely well, the setting is dark * moody, feels chilling & bitter. With the excellent audio sfx selling the moment. It's in the moment that I feel my "mojo" shines, my "thing". Whatever it is, I can make an environment pop with small a small amount of time. This was the first time I used Unity's Terrain tools and took full advantage of some free PBR resources to shape a playarea for our protagonist to walk around in. Initially the map was huge and after several playthroughs I brought it down and then even more so throughout the course of playtesting, eventually to the last day where I realised it was more of a walking simulator and placed additional fences to bring in the playarea and add to the enviroment, this helped greatly.

Since the focus of the gameplay was to light fires and activate totems, I knew I wanted to make these the key focus and bring this to the front of the players attention. Using Cinemachine, for the first time ever I used transitions to move from one key point of view to another, and with that same technology I gave fireplaces a meaning of emphasis with their own camera views. This added an additional level of polish and atmosphere building I didn't initially expect but paid of well in the long run.

With one project to the next, I have found my own unique style that I enjoy developing in and that would be the world of 'paper-like' genre. There are many great games out there that use this style and it comes with its own set of challenges and problems to solves. I have had few people speculate that I use Spline which is an excellent tool for creating a skeletan bone system for 2D Animations with layers & deformation. I have looked into spline and also its cheaper counter part "DragonBones" and while both are fantastic tools, for gamejames its hard to recommend as the pipeline is too tedious for a solo developer.

Despite the extra additional advantages those programs bring, It does not allow me to work quickly in an interative state. For the longest time I have been using Unity 2D Animation package with additinal built packages for 2D Sprite features. Using Unitys own version of bones and skeletal animation I can rapid create animations in engine and use smoothly from one state to another however it does not come without its disadvantages which we will cover below.

3. What went wrong?

After a week of working on your project, the honey moon phase passes and you get to take a look at your project without the rose tinted glasses for what it truly is and what it is percieved as, for no one can quite love your own work as much as yourself.

Reflecting back at my submission I came to realise that it lacks rewarding gameplay (..as like many of my other games) and it is a nasty pattern that keeps rearing its ugly head at me after all my projects. Players are forced to walk in 360 degrees "searching for needles in a hay stack" (proverbally speaking) and br punished if they do not find wood in time, by slowing their progression down even further. It's quite a blunt outlook on my "baby", submission that I am still oddly proud of. Like previously stated, I have learnt many new things from this project, so its okay that I can see my downfalls and work to improve on them next time.

Unfortuntely the gameplay isn't the only thing that truly lacks. After all that slow walking and patience testing searching wasn't enough, you could be one of the unlucky players to run in to the myriad of issues plagued by "crunch-coding" in which technical debt is built up from squeezing too much in one go. Sadly this leaves a sour taste in many players mouth and ruins the experience. Soft Locks, Hard locks, Unresponsive and Freezing animations. Sadly WildLands has all of those for the unfortunate ones that experience them. The issues I believe have been from the way I setup my characters using mecanim transitions to move from one animation to the next as smooth as possible. When done right it looks extremely smooth and impressive! When it goes wrong though you end up with something like this:

When you add a lot of animation transition, there are conditions that need to be met for each animation to play. I believe my management and poor implementation of these conditions caused conflications and stopped the player from performing a new action. I wont go into the nitty grittey but for the future I will be looking to make statemachines a core of all my player states and animations.

Also as a last though for this section, my interpretation of the limitation may have been too weakly implemented. Initially I had planned on every fire you light adds to an overall score, this score would decay over time natrually and you would need to keep fires topped up with firewood to stop that. Above specified scores new tiers and titles would be unlocked which would be necessary to hold when speaking to NPC's to unlock additional rewards. Unfortuntely, over development stages, the initial idea got lost to time and totems were brought in as a goal with your score being used to open locked notes around the map. These locked notes would require a certain amount of your score to be opened, but I set the score so low many didn't even notice they were spending it. The idea and concept was ultimately very weak and not fleshed out with sufficient feedback to be a good interpretation of the limitation.

4. Things I had learnt or done differently.

When I set out to create Wildlands I had these simple thoughts:"Stay warm, collect resources in the cold, keep the heat going".

I understood from day 1 that it wouldn't be enough to have just these simple words describe a great game, mechanically there would need to be more. Fitting in "more" would be the troubling part as I have a bad knack for prioritising gameplay over a polished visual experience. I have known this for the longest time too, I actively go into each Minijam with hopes to actually provide a decent gameplay experience.

I have a lot of growing when it comes to understanding core principals that make for "good" gameplay. Pillars to rely on, checkmarks, a list, key features, the formula. I am still searching for that "sauce" to make a game fun to play. If you have any experience on how you improved in this department please share your experiences.

Despite not learning how to successfully develop a fun gameplay loop this time, I am happy with my further exposure to cinemachine. Using a dolly cam for the intro was extremely satisfying and while I only animated the position of the camera on the dolly and dactivated the camera towards the end of the track, the results look good. In the future I would have liked to try more 'cinematic' animation with moving cameras and characters similar to something like this:

Ultimately I need work on developing concepts and core gameplay values and that will take some time.

5. Useful Tips & tricks

If I can end on any good notes it would be my advice for achieving such look as my own. Universal Render pipeline was used for all the nice features it brings like post processing effects, camera stacking, screen space ambient occlusion and decal (which I used for camp fire radius's and other additional markings on the ground over uneven terrain.

Speaking of Terrain, the built in unity terrain tool will save you hours of work if you get used to the toolset. Trees were something I placed manually but there is a good tool for placing a lot of trees if needed.

Normal maps on 2D sprites looks great! If you can incorporate normal maps with your 2D images then I would highly advise it, when working with atmospheric lighting, flat 2D sprites will look like they belong when they interact with lights in a 3D world. Personally I love the RIM lighting look on 2D sprites that make them pop from the background.

The main player is quite complex in that it casts a shadow from light sources, this is not something build in my default currently and you have to do some initially setting up. There are multiple ways of going about this for different edge case uses but ill explain the ones I used the most:

• Using 3D planes with 2D textures.
  This method was the first one I did to achieve 2D looking characters that cast shadows. Create a 3D plane object, attach a new lit/unlit URP material on to the 3D plane. Not on the material attach the 2D texture. For me I seperated all limbs and body parts, so you might end up with 6 or more 3D planes. Set the rendering mode to cutout. This will allow light to cast shadows based on your alpha values and cutout values. This will also avoid z-sorting issues so long as you dont use Transpaceny as the rendering mode.
  
  When animating you will need to setup all the 3D plane objects as childen in a heirachy that allows you to move one or many limbs with the rotation tool. This takes a little bit of setup but works okay. It's not advised for everyone as the animation can be harder to achieve. This is what my results looked like with this method:

  



• 2D Sprite shapes
  Using 2D sprites and 2D animation skeletal animation can lend to some excellent results with animation but you still need to apply a material with rendering mode set to cutout. This allows for better animations using Skeletal bones, and allows for sprite library swap out of assets so you can reuse animations for multiple characters (I have yet to do this myself.) ..