Forest Adventure Postmortem

Introduction: Hello, I would like to present a postmortem about a project I worked on named Forest Adventure. This is a modern solo exploration game. The player traverses through multiple levels to travel from a starting campsite to the safety of the Adventurer’s Lodge. From a simple blockmesh, this project evolved into a large, detailed world with lush forests and detailed platform puzzles. I would like to take this time to discuss some of the things that went right and wrong over the course of the development process.

What Went Right

Beginning and ending level transitions – I originally thought that combining the levels for this project was going to prove quite challenging. However, as it turned out, I was able to find a fun way to do each transition. What started out as a blockmesh tunnel leading from the beginning area into my level turned into a rocky opening in the surrounding cliffs leading the player forward. The end transition matched nicely with the idea of leaving the bunker. What used to be simply another blockmesh tunnel evolved into an old tunnel that was built to help bring supplies from the outside world into the bunker much easier, avoiding a trek over the cliffs that surrounded the exit from the bunker.

Living environment – My original blockmesh did not do my level justice. While you could tell what everything was supposed to be, it still seemed so lifeless. Adding a real firepit, a flowing river, and a cliffside waterfall did wonders for bringing the level to life. A further improvement came from the use of foliage tools, allowing me to add lush greenery to the bland level, including patches of greenery on the originally barren cliffs located throughout the level. While I know that it is important to have a strong blockmesh that other designers can understand, especially as far as what various areas are supposed to be, I also learned the joy of enhancing them to create wonderful lands a player can enjoy passing through.

Exposure to Perforce – When I began working on this project, Perforce was a very intimidating beast that I was afraid to tangle with. However, through proper practice and study, it became a good friend to me. There were several times that I was working on the project and had accidentally had the wrong part of the project selected while I worked. Thanks to Perforce, these changes were never permanent until I submitted the content. When doing so, you are asked to check out the content you made changes to, which always allowed me to catch my mistakes. Perforce also makes it extremely easy to get the project version prior to my changes, allowing me to quickly get back to work making the changes with the correct content selected. What was a scary beast became a strong and helpful ally in my journey.

Smooth overall transition from blockmesh to final level – With the exception of a few items mentioned later, transitioning from a blockmesh level to a level with proper assets proved to go very smoothly. The blockmesh level was designed well enough that it was easy to figure out what assets would work best for filling in each area of the level. The cabin and greenhouse became strong features in the level, displaying most of the concepts laid out by the blockmesh design. A vibrant lake-side camping spot showing the possibility of nearby life added mystery and a desire to explore further. What became an abandoned bunker added the desire to know what it was doing there in the middle of the woods.

Playtesting and player feedback – Playtesting and player feedback proved to be extremely helpful during the development process. There was an easy to fix bug that I had overlooked with a treasure chest. The player notified me that the item could be collected without ever touching the chest. This turned out to be a simple mistake of not assigning the item to the chest. There were also several areas pointed out to me that allowed the player to escape from the game world. A combination of appropriate assets and level manipulation quickly fixed these problems. Without the playtesting and feedback, I may never have noticed these issues, proving to me just how important playtesting is in level design.

What Went Wrong

Blueprints had to be reworked - One key aspect of my level was the inclusion of interactive doors that required keys to open. The original blockmesh blueprint that I created for this purpose took a considerable amount of tweaking to get the door to end in the correct location after it was activated. Unfortunately, most of this effort had to be reworked once proper assets replaced the blockmesh door. Switching to the asset caused the doors to open in strange and unacceptable ways. However, having already experienced the struggle once with the original blockmesh blueprint, this was an easy matter to fix. I already knew that I just had to change the rotation values to guide the door into the right position when opened. Thus, the blueprint was able to be corrected and functioned perfectly.

Landscape learning curve – A big part of the final product was the introduction of landscapes into the level. This proved to be quite the time-consuming addition to the level, as I had very little practice with landscapes prior to this project. One of the biggest hurtles was learning why my materials were not being reflected in the landscape. Through trial and error, and some advice, I learned the correct way to add the various materials I wanted to appear in the landscape. Once I learned how to do this, adding foliage, grass, and various other features to the landscape proved to be much easier as my experience with the landscape tools grew.

Incompatible features – One major hurtle I encountered while combining levels together in the project was discovering that not everything in my original level made sense. For example, the end area of my level originally had a village at the bottom of a hilltop path. This was originally meant to be a place of refuge for the character at the end of the level. However, the village made no sense when the next level in the sequence was to be a derelict castle. While most castles typically would have a village near or around them, the flow of the level just did not allow for this, and so that aspect of the level was cut from the final plan.

Valve that no longer served it’s intended purpose – In the blockmesh level, I had created a valve location in my level. This involved having an archway with a tree on each side that fell over when the player passed through, effectively blocking the path back out. Once these meshes were replaced with real tree assets, the valve ceased to work as such. Not only was the player able to pass through the archway with only minimal hinderance, but they were able to freely roam the majority of the open space that had previously been inaccessible due to blockmesh trees and other obstacles. While this did change the overall flow of this section of the level, the end result was much more visually appealing and enjoyable for the player to travel through. While the valve vanished, more immersion was born from this venture.

Level resizing complications – One very noticeable aspect of the conversion from blockmesh to a level with full assets was the change in how much space many areas and features took up. This forced changes in several areas. For example, the large, shipwrecked boat at the beginning of the level became a canoe that had run aground. The bunker, which had originally used massive blockmesh constructs to create it, shrank considerably in size, allowing for a much cleaner appearance. While this was a great end result, it did push me to move the inside of the original structure around to match the new sizing.

Conclusion: Working on the Adventure Project has been a journey full of trial and error. There were many instances where frustration attempted to overcome me, tempting me to cut features that I wanted in my level. However, through perseverance, study, and patience, I was able to successfully overcome these trials. I learned many new skills, including how to use landscaping tools much more effectively and how to manipulate blueprints to get the desired results. I learned how important blockmeshing was, as it made creating the asset-filled level go much smoother than it otherwise would have. Finding appropriate assets for the concept being worked on is much easier if you already have a fleshed-out plan to work with. In the end, a dynamic project full of diverse and enjoyable content for players to travel through grew out of a basic blockmesh design.