Play-testing

Heartland, my project 4 “Home” game has given me the best opportunity for Playtesting and provided valuable information both about the game and how to go about playtesting in the future. The biggest lesson I unfortunately learned during playtesting Heartland, was that without a build up by the deadline, you wouldn’t get to participate in the playtest. This was a frustrating and disappointing experience. The fact that we were unable to provide a build for the second playtest was such a horrible feeling, letting down both my mentors and my peers. It also meant we missed out on the analytics which would have really benefited the game knowing how many time people talked to the other characters or even just a heatmap of the paths they took.

Although we missed that playtest our final playtest was incredibly successful and gave us clear direction on what to polish. For example on of our questions asked players how they felt after playing the game and one playtester wrote “Bittersweet”, this was the exact feeling we were aiming for and to see that written was incredible. To give some context to that as well since we had missed the previous playtest and our early build was incredibly different from the one we then tested there was no outliers or bias towards the feelings produced in the game as noone had prior knowledge of what the game was ot what was going to happen. Another question we used was asking the players if they understood the narrative. Although our narrative hadn’t been fully nailed down at that point it was fantastic to see people grasping the overall concepts we were trying to portray.

As with any good playtest aswell we always asked why. Why the players felt the way they did, why players understood what to do and so on. This helped us craft our player feedback in our game around these answers, for example people knew what to do because the cursor changed on where they could and couldn't walk and on what they could and couldn’t interact with, placing a character right next to the farmer to that the player would instinctively mouse over them and see this change became a very subtle tutorial for the player which would not have been possible without breaking down and studying our playtest results.

All of this actual usable playtesting knowledge didn’t come from just anywhere. I spent time researching professional level playtesting techniques which helped us conduct the sessions we were able to. I found Dan Felders “Design 101” blog on Playtesting and its breakdown of the different stages of the development cycle intertwined with playtesting incredibly beneficial. For example he states that “In the earliest stages of a game’s design, you want to quickly test if your core concept is fun. This is often easy to do even in an incomplete form.” This resonated with me upon reflection of our first playtest. We had an isometric camera and you could move the farmer around two distinct large areas filled with objects. Only some of these objects were interactable and that was it. Going into it I was really happy with what we had created, I felt like we had a beautifully built level and our systems were well on track. But as some other research into playtesting showed me, playing your own game after looking at it all week makes you look over any blemishes. Firstly people thought our character was insanely slow and combining that with a expansive level made it tedious just to play it. They also noted that nothing actually told you what you could and couldn’t click on which made it unsatisfactory to actually click around the map. This was frustrating but when reflecting it was incredibly valuable feedback which we completely phased out of the game. The slow movement speed was pretty much always an issue but we gave the farmer a limp so the player would understand why he was slow and still spent times speeding him up just trying to make sure it wasn’t unnatural. We also completely changed the camera and map layout making it one area and the only objects in the level were ones the player could interact with, to further convey this we added a cursor swap on mouseover which would show the player how they could interact with things differently, for example clicking when the foot cursor was up it would move the player and the speech bubble would allow the player to talk to the NPCs.

One aspect of our play-testing that I believe needed improvement was what Jeanne Collins wrote about in her 1977 piece on in house play-testing, as she puts it “project legacies”. She states that by “project legacy, I mean the harmful folklore used as justification for not solving an often-cited problem.”

This was clear in our project when I tried running a playtest at my house with my family and had to explain that even though we were halfway through our development cycle we still hadn’t added the functionality to press escape to quit. Two lines of code that can be copied and pasted into the game still hadn’t been done. I took me all of one minute to slap it into the game. It took me longer to create an incredibly passive aggressive hack and plan task.

Although our playtests for Heartland weren’t smooth sailing, the overall outcome and information we gleaned from what we did test was incredibly valuable. Researching the different development cycle stages of playtesting were important to get the most beneficial information from the tests we did partake in, which in our final test gave us amazing results and gave us the drive to polish the game and be able to stand proud next to it at our final showcase.