On the 5th October we conducted a testing session for our game to determine whether we are heading in the right path, and to gain insight on what aspects need to be improved.  In this report we summarise the results of this session.

General Summary

In general, the responses were rather positive and people enjoyed their time playing the game. While there were some issues and people got stuck, it seems that the freedom that the gameplay mechanics gave the players allowed for them to enjoy experimenting while they were confused.

Notable Issues and Fixes

Connection Control Issues

There were a few issues with the scrolling controls. Several of the testers said that they were having issues with the scroll wheel. A good suggestion was that there should be a separate input using the keyboard so that any users that have issues with the scroll wheel have an alternative option. Q and E would be good choices so over the next few days this functionality will be added.

Connection Selection Tutorial Introduction

In the middle of Level 2, players were required to create two separate connections to progress. Two labels with the numbers 1 and 2 were positioned to try and hint to the player that they needed to press the 1 and 2 keys on their keyboard to switch between connections. This doesn’t appear to have been a clear enough instruction, as most players got stuck here and required being informed how to proceed. Possible fixes would be replacing the number labels in the puzzle with more keyboard looking keys, and also adding some numbers to the UI on the side of the blunderbuss to match.

Connection Selection Clarity

Further on, some players noted that it was difficult for them to remember which connection in the world was currently selected, as they sometimes overrode or cleared the wrong connection. This could be fixed by having different visual effects for the selected and non-selected connections, and possibly some visual cue when they’re outside the players point of view.

Connection Effect Bug

When players selected an object and made a new connection in a specific way the effect on one of the objects would be turned off when it shouldn’t. This is a fairly easy thing to fix as the reason for this bug was that the way connection effects are handled was recently changed but a small part of this wasn’t updated.

Player’s relative momentum on moving objects

On the new models the player would not move correctly and could fall off objects. This is an issue with the code that keeps the player on an object as the scale of the models is a strange value that was not accounted for when the code was initially written. The models will need to be adjusted to make this fix.

Missing Textures

Some of the textures on level 3 were not loading correctly or were missing entirely. There were also some stray objects that the player could interact with that were creating strange visual effects. The objects will need to be looked at to find out why they are acting this way and fixed accordingly

Battery Softlock on Level 3

If the player gets to the second area of level 3 and falls out of bounds they get sent back to the checkpoint but the batteries get sent to their starting position. This creates a situation where the player cannot progress as expected so it is almost impossible to complete the level without using some weird strategies. There will need to be a fix for this as the player gets locked out of progression.

Level Length

The majority of people said that they found that the levels were an appropriate length. There were a few comments about the 3rd level being far longer than the others. Generally it was suggested that either the earlier levels should be made longer or that level 3 should be split up to keep consistency. The length of level 3 was also attributed to a difficulty increase. The levels will need to be assessed for how long they are and we will need to decide how we would like to fix this pacing issue.

(Scale in graph is “too long to too short”)

Player Experience

Intuitiveness

The mechanics were generally regarded as feeling intuitive.  We did notice some confusion that arose particularly in level 3, which was more open-ended in comparison to the other levels, so we will review the footage that we recorded of people playing this part to assess what concepts took too long for players to grasp, and then potentially add UI prompts or visual effects to certain objects to make their intended purpose clearer.

Ease of Controls

The controls were generally regarded as easy to use, and the opening tutorial in particular, which forces players to follow a short sequence with prompts to teach them the controls was very well-received.  However this category also attracted some 1/5 scores, and we have identified some issues with the input as outlined in the previous section that will be addressed.

Visual Style

According to the graph, the level of visual satisfaction was at least satisfactory, with no responses scoring lower than 3, showing that things are going in the right direction. Within the survey, there is also a separate question that asks for individual feedback, and there have been quite a few responses to that question that need to be addressed. One of the players suggested that we should include a visual indicator on the trigger locations, and we had already planned to implement this change in the next update. There have been reports of some of the textures not working properly; our team has a feeling that this may be due to the shader graph not applying the texture appropriately. It was suggested by a few individuals that the dark-toned cyber texture on the first level did not look as nice when placed against the bright backdrop. Nevertheless, this is what we intended for the game’s overarching concept to be, with the first level taking place in broad daylight and the tone gradually becoming more ominous as the player gets closer to the ending. One of the players suggested that we should develop an environment as rich as Black Mesa’s Xen, which is an ideal aesthetic but due to the constraints of the project realising that vision seems beyond our scope.

Ease of puzzle solving

We asked players how difficult on average the puzzles were to solve.  The scale ranges from Easy (1) to Hard (5).  The distribution of answers towards the centre, being neither too easy nor too hard, indicates that the puzzles were at a fair difficulty level.  However, we did observe players having some difficulty in level 3, for example blocks getting stuck on walls. Some players also noted that they found exploits that allowed them to skip very large portions of levels, making them too easy.  By addressing both of these observations the difficulty of the game should be even more balanced towards the centre.

Level Enjoyment

Testers were asked to rate their satisfaction for each of our levels, largely to help the level designer judge areas of improvement. The results showed that levels 1 & 2 had considerably high satisfaction with 44% of testers rating level 1 a 5/5, and 37.5% for level 2.  However this margin fell to only 12.5% for level 3, indicating that it needs more playtesting and design work in comparison to the earlier levels.  

Overall Satisfaction

The game received a positive response from testers overall, which we are pleased with.  This has helped affirm that the game is progressing in the right direction, and using the insights from this testing session we hope to further improve the player experience.