testing… 1, 2, 4 testing.

So a second round of play testing was run for our Brief 2 project. By this time the games design, premise and aesthetic had completely changed as the original concept didn’t seem like it was gong to meet the requirements. After this tear down  process we were a little behind and for this session had only the core functionality of our game and lacked a re-playable game loop. This obviously limited some of the feedback we could receive about our game but in spite of this we gained some useful insight via face to face discussion with testers after they had finished playing. testament to this we had a full page of notes so I will cover a couple of the major things we took away from the 2nd testing session.


Small tweaky things that we got used to.

AI detection ranges. Our AI were a little short sighted allowing the player to walk straight up to them. This felt unnatural and lessened their level of threat.

Player sprite is off center. The players sprite shoots bullets from the gun on the sprite rather than from dead center, combined with the slight off center sprite setup this made aiming quite difficult as the expectation of the bullet trajectory was not met. This lead us to adding a ‘laser sight’ that projected from the players gun which gave the player a clear expectation of where bullets go.

So much HP. Health pickups restore the players health but only up to a maximum. We saw multiple times that players would walk over pickup while on full health, then walk back over it multiple times wondering why they couldn’t take it. This was a two fold problem in that the diagetic and UI elements for health weren’t clear and that there was no feedback for when you were full. One fix implemented was a small ‘error’ tone that plays if you are maxed out when walking over a health pickup.

Anime bullet mechanics. Simply put our bullets could collide with each other. Although this could be interesting to work with we had to adjust the trigger settings on bullets to maintain our desired game feel.


Less obvious things, the vibe of it all

A great discussion that was had with one of our lecturers was one about game feel. This discussion could have gone on far longer but this challenged us to be far more deliberate in the small details of our input devices and how this tied in to the mechanics of our game.

During the test our player can auto-fire bu holding down the trigger at a set fire rate. The question was posed as to why it was setup like this. Our game had a survivalist feel in some regard but our mechanic allowed the player to run and gun which feels powerful. But alternatively what if we had allowed only semi automatic firing, what if there was a clip size, reload time and how far does the trigger have to be pressed on the controller to shoot? We didn’t have all the answers immediately but these raised questions for ourselves in terms of design goals and to make decisions about our desired player experience. For example restricting bullet fire rates and only allowing semi automatic fire can make it feel like ‘every bullet counts’ changing the way the player approaches encounters and the way we balance enemies.

It was noted that the environment we had constructed from an off the shelf asset pack had a great feel and aesthetic to it and that we should utilize it as much as possible to enhance the current experience. This was a hard one to digest and decide how to act on this feedback. The environment was reminiscent of an industrial district from the 40’s or thereabouts with many tall buildings that looked great from the top down perspective. One way we somehow used these assets to our advantage was in the lighting. We added spotlights to the streetlamps throughout the level to highlight the critical path and disabled the directed light to enhance the aesthetic. On top of this spotlights were added to the car models for their headlamps and were placed through out the level and pointed in the general direction that we intended for the player to traverse. This was immensely important due to the low lighting of the level. Players in later tests reported that they defaulted to following the car headlights, so big success here).


Final thoughts.

It’s obvious, but be aware of when you are going to play test or schedule them accordingly to milestones so that you can get the most out of it. The more functionality and feedback that is in your current build the better, even if audio and visuals are just placeholders get them in! Something a little less obvious is to prepare your play test environment. Ensure that the area is not cluttered, especially in our classroom environments, remove extra chairs, personal belongings. If possible adjust lighting, test area proximity and anything else applicable to make sure your testers are comfortable and aren’t being distracted by anything that might taint their experience. This sets a professional tone and can help in setting expectations that your only testing  and not shipping the final product.




Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s