Voice Line Entry with Syd: 2h
Total Hours: 2h
For the majority of this week I was at my parents’ house watching their dog while they went to visit my grandparents. Unfortunately, this meant that I wasn’t at my computer and was basically out of commission for the week.
I did manage to do a couple hours of data entry with Sydney, where we ran my script to get all of the voice lines into the database, and then started going through the game and manually updating conversations. We got that about halfway done before we decided to call it a night, as I have a few more scripts and functions I’d like to make before we continue. There are a couple conversations and hints that are accessible in-game now, and we should be getting the rest of them into the game within the next week. Luckily, the script added the conversations to the database successfully, so that majorly cut down on the manual work we have to do.
I wasn’t really around last week, so I didn’t get a lot done. I should be around all of this week, and will be able to finish up putting the conversations into the game, among other things as needed.
Bug fixing as needed
Finish adding conversations to game with Syd
Create Database Input Script: 7h30m
Teach Syd how to create input file: 30m
Fix Opening Scene Selection Bug: 15m
Various Minor Fixes: 1h30m
Total Hours: 9h45m
I was able to create a powershell script to automate putting the dialogue with the appropriate mappings into our database. The script takes in a JSON file that contains all the conversations with the audio files and outputs a JSON file that shows the starting prompt ID of each conversation (which we need to “unlock” each conversation as the player progresses through the game). This means that we don’t have to worry about spending tons of time manually inputting our dialogue into the database. We can create the JSON file and update it as needed, and run it as needed to keep the build updated. I had to spend a little time teaching Sydney how to create the input file, and I added one conversation to it with her so she could see how it worked.
My only assigned task this week was to fix the opening scene selection bug, where the right hand trigger wouldn’t work to select the option to start the game. This was simply an incorrectly assigned variable, which was instead assigned to the left hand trigger. I reassigned it to the right hand trigger, and it works as expected both in editor and in a build.
While I was in the scene I also made some minor changes and fixed a few things that I had noticed were accidentally pushed incorrectly. Primarily, I updated the opening dialogue scene in the hallway so the player can move around while the characters are talking, and if they solve the door puzzle it simply cuts off their dialogue. There are new tutorial images placed in the hallway, and a grabbable object, so the player can use that time to get the hang of VR controls instead of being stuck in place.
The one negative from this week is that I am the only person who knows (and has their computer set up) to run my new powershell script. This shouldn’t be a problem, since the majority of the work now is creating the input JSON file, which many people are capable of doing.
Most likely bug fixes
Hook up New Room Model: 3h15m
End Game Logic: 1h15m
Unity Bug Fix: 1h30m
Database Entry: 2h
This week I made good progress on our backlog of tweaks and bug fixes, along with adding in some finalized art assets and dialogue.
First, I hooked up the new room model that Sydney created. This took a bit longer than I originally thought it would because an obscure change broke the moving walls. Thankfully, I finally realized what was happening (after some rubber duck debugging with Erin), and we were able to get everything set up again.
I added in some logic to our end game scenario so that we can tell what character the player accused and what evidence they used in the ending scene. This will affect what voice line plays. In this video, there are different sounds played depending on the character that was accused.
There was a bug in the lightweight render pipeline version we were using that affected our canvases (and thus, our conversations and tablets) in builds. Originally, we were going to write a workaround for it, but I realized that the bug was fixed in a newer version of the lwrp. So, we upgraded our entire project in order to use the newer version of lwrp. It took a while for both Erin and I, but luckily it fixed the canvas issue we were having in our builds and also our corpse rubberbanding around the room.
Finally, I did some database entry with Sydney. We have all of Mavis’ voice lines recorded and cut, so we added his conversations into the database. Since the player’s voice lines haven’t been recorded yet, you can’t really have the finalized conversations with Mavis in game. However, this should cut down on the work we have to do involving entering the rest of the conversations into the database in the upcoming weeks.
The database entry this week was a little tedious, and more time consuming than I’d like. If I have time in the upcoming week I’d like to write a script that can read in a file in a more human-readable format and put it into the database for us, to make our lives easier.
Bug fixing as needed
Database entry as needed
I didn’t do any work on the project over break besides some minor branch maintenance and helping Erin migrate a model.
However, here’s a brief rundown of what I did do over break:
- Slept a lot
- Got a tattoo
- Started and finished Dead Space
- Made macarons with Sydney successfully for the first time
Next week I’m mostly on general debugging, but I know that I will specifically be taking a look at the piano cylinders and the lever to try to refine or update those interactions.
Unity Update: 30m
Conversation Update: 3h
Conversation Bug: 15m
Play Test: 2h
Total Hours: 5h45m
The recent Unity security patch forced us to upgrade our game to a version with a patch available. I spent a little time researching the changes between our previous version and the one we wanted to update to. I was able to update our project with no issues, but I ran through a bunch of tests afterward just to be sure.
The main thing I worked on this week is an update to our conversation mechanic. Previously, we could only have one conversation thread active for each NPC at a time. I changed it so that we can add and remove multiple active conversation branches, which can be accessed at the start of a conversation with any NPC. The video first shows a quick run through of what happens when a new conversation is added. If the conversation has begun, it will not be removed until it is finished. The video also shows a second run through where the extra conversation is active, but not started, and it is removed at a set point in the game. The extra conversation does not have any voice lines, it is just text shown as a proof of concept.
While I was working on the conversation update, I found a bug. If you started talking to a NPC, but teleported away from them while they were still talking, the next dialogue options would still be shown even though you were no longer in an active conversation with them. Luckily, the fix ended up being rather simple, and didn’t take too long to implement.
Finally, while Erin and I were down in Charleston with the pep band, we held a small play test. We got some informative feedback from it, especially having to do with our sound and general UI.
I wasn’t able to spend a ton of time on the project this week since I was in Charleston with the pep band. Fortunately, I didn’t have a ton to do, but I generally like to spend more time with the project.
Bug Fixing as Needed
Updating Puzzles as needed