YEAR 2 - BA2b - Evaluation

Both Programming and Designing a game whilst Managing a team of 5 people for 6 weeks has been tough. There have certainly been times when frustration was high with people not doing what I'd asked and there were times where I didn't do things when I said I'd have them done for. I feel like the game itself is good though. I do think mechanically this game is lacking in areas and I take responsibility for that but the idea itself is very novel and was unexplored territory for me. I've learned a lot from developing this game. Firstly as the manager I need to keep a better track of what everyone's supposed to be doing. I did experiment with Trello and I probably should've stuck with it throughout but that's a lesson learned for next time. We did make good use of the Slack group I set up though. As a designer I think the concept was a good choice because it gave everyone on the team a potentially infinite amount of work to do so no one should've felt like they had nothing that they could be doing but it also didn't need TOO much to actually function. So the scope was flexible. One problem with the design is that it required a lot of foundation code in place before I could really call it 'playable' and by that time we were far through the project. So player testing was limited. We honestly didn't even get the point where testing would have been particularly useful because the systems never got to the point where they interacted with eachother enough and I had enough on my plate to work on throughout. Perhaps the game was overscoped slightly. Me and some other team members have agreed to carry on working on this game after hand-in because there's still more work we want to do on it and it gets exponentially better every time we add something new.

As well as what's previously mentioned I modeled, textured and created 2D artwork for the game where it needed doing. I feel like I distributed work evenly throughout the team who, also did work of their own volition. The ideas, documentation, assets and concept art I received that I didn't ask for were some of the most useful and contributed greatly to the project.

I did a lot of work for the team during the cross course collaboration. I was a big part of any discussions we had and I developed 2 interactive demonstrations working with other members in the team. I stayed back late in university working so as not to hold up development at any point during development.

YEAR 2 - BA2b - Behind the scenes

I did a lot of little things to the scene to make it feel as rich as it does that probably won't get noticed by a lot of people. This is what the landscape looks like from above. There are huge white blocks below the ground in particular areas to control the intensity of the sun at different times of day.

The arrangement of all the trees on the horizons took ages to get right and only look correct from the angle the player will see the game. This was all done through trial and error.

Another trick I put in the game is to have the sun do an 8 shape in the sky rather than a normal circle. Imagine the cross in the middle of the 8 is above the players head and the rest of the 8 curves around the earth. This allow the sun to light the house from all angles whilst it's in the sky but more importantly, it lets the player see the SUN RISE AND THE SUN SET on the same side of the screen.

https://youtu.be/PhM7xZJlPj8 - Watch this video. You can see that the sun spends a lot longer in the sky making the days longer. You can also see the sun rise and set on the same side of the screen.

The way I fixed the issue with the dolls jumping around too much when you try to move them is with an invisible wall that sits inside the house. This keeps the cast ray hit point at a fairly consistent distance from when you initially press down meaning there aren't as many sharp jumps in distance making for smoother doll movement.

YEAR 2 - BA2b - JukeBox

The Game having a jukebox is something that was discussed from the start but it's something I've not really been able to implement until now because I didn't have the foundations in place to allow it to affect the dolls in the way I wanted it to. We discussed having the UI change into a playlist with a choice of music when the player clicks on the jukebox and the player was going to be able change the music here. They would also unlock new music by doing different things in the game (but time).

So the way the jukebox works in the game currently is it just switches to the next song in the sequence when clicked on where 'no song' is in the sequence.

the jukebox acts as a way for the player to change the music during the game and they can use it to control how the dolls are feeling when they're in the basement. This means that as the player progresses, gains more knowledge and unlocks more songs, they have more control over how the dolls behave and they can use this to enact specific scenarios. This should allow the player to acquire a greater feeling of accomplishment the further they get through the game.

https://www.youtube.com/watch?v=81fjoAsoj3E - Follow this link to see the jukebox working

YEAR 2 - BA2b - Ghosts, Gravestones, Facial Expressions and Rare Hats

https://youtu.be/9DNcQQS90YY - Go to this link

So I've just implemented souls into the game. When a doll dies their soul becomes a ghost and flies away. Occasionally if a doll is particularly immoral, their ghost sticks around and haunts the other house mates (Not implemented yet). I also plan to allow dolls to capture souls and reincarnate dead dolls or sacrifice them to Lucifer in return for rewards but I don't have time to implement this before the deadline.

Players can place dead dolls in the back garden to bury them and a gravestone will appear in their place.
There's a picture of a dog in the kitchen looking out towards a gravestone. I'm using environmental story telling to imply that the dog is buried there. I'm also using the placement of the gravestone to imply to the player that they have the ability to bury dolls there.

I've also made expression to give the dolls. The plan was to have these expressions quite extensive and unique to each doll but we ran out of time. Currently all the expressions do is show how happy the doll is.

I made all these doll faces. I tried out a few in game but I chose the ones I did because they were the most expressive. I decided the one with the open mouth was too goofy. It is important that the faces are expressive because they are difficult to see from how far zoomed out the camera is. In game they are still hard to see and when I carry on with the game after the project is over I'll make sure they're easier to see.


Still though, I feel like the doll faces bring a lot to the game. It's easier to sympathize with the doll faces. I found myself feeling bad when I made the dolls sad. I think they would have been even more powerful if they each had their own individual face types.

In order to make the dolls more individual I asked Ben to model me some hats to give the dolls.

I made it so the dolls have a rare chance of starting with the hats. This is because I wanted the hats to be somewhat of a rarity. I think it's exciting when you happen across a rare occurrence in a game and it's the kind of thing that players share with their friends and over social media which is obviously good for the game.

YEAR 2 - BA2b - New Background and Dialogue Box

Most of the assets are in the game now and It's looking really nice. We were discussing whether to have something behind the UI so it didn't blend into the background too much and I trialed a few things. We figured that we'd risk blocking out too much of the background by putting a box behind it so it was decided we'd do something with the mountain.

I tried emulating a 2D UI box by placing one physical one in the level and it looked pretty cool but a bit out of place. In the end we decided just to have the mountain rise incredibly high so as to surround the UI.

Also, ever since implementing the day/night cycle the lighting on the house has been FAR to light during the night, but I'm happy to say I finally found a solution.

I placed trees going up the mountain very carefully so as to follow the very edge of the horizon. This was to give the game more character and to make the mountain feel less slapped on. It was important that the area just to the right of the house was left free because that's where the sun sets and I wanted players to see that. Suddenly it feels like the dolls are living in a fleshed out, breathing world.

I've been experimenting with a dialogue box that describes what dolls think of eachother or just generally giving more clarity to what's going on in the house.

Unfortunately the compression on this gif is very high but what's happening is, as the dolls are eating above a certain food level, the dialogue box is telling the player which doll is too full.

I would've extended this concept to more things in the game. It could've had it's own quirks and give hints when not much is happening. I couldn't get it to work how I wanted though. ALMOST but not quite so I've scrapped it for now until I figure out how to get it working how I want. (Basically if there was more than one thing to talk about at once it would always prioritize the earlier in the array first and I wanted it to choose a thing to talk about randomly) (I could swear my code should work in theory! But maybe there are still thing I don't know about how UE4 handles arrays).

This recorded this gif to show cannibalism since I didn't actually show it in my last post

YEAR 2 - BA2b - Day/Night Cycle, New UI, Sex and Mingle, Cannibalism

Since the last post a lot has changed.
- Firstly, we now have the doll assets in the game with clothing that changes depending on the situation.
- I made a day/night cycle that currently is just aesthetic but I'm looking to sync it with how hungry/ tired the dolls are and with specif event too.
- Dolls now lay down when they're placed on a bed and if two dolls like each other enough they'll have sex (which raises their happiness drastically but lowers their hygiene drastically), they also start to like the other doll even more. This would be an unlimited resource of happiness however they won't sleep together if one of the dolls is unhygienic and they'll start to dislike eachother if they spend to long together. Since water is a limited resource, so is sex.

Initially I made it so that the dolls could have sex even if they didn't like eachother and that would make them dislike eachother even more. However, it felt wrong and pretty controversial to leave that in the game so I changed the code to make it so the dolls just lay next to eachother sharing the bed. I do think that controversy isn't inherently a bad thing but it's an area you have to approach with caution and I don't feel like I can give it the caution it deserves with the time we've got remaining.

The dolls get little hearts above their heads if they're going at it. They also remove their clothes. They also remove their clothes when they're alone in their room normally.

Side notes - I made the table to their right and I'm pretty proud about how it looks.
- The bed has a 50% chance of spawning as a different bed made by a different member of the team.


I actually trialed having a hand drawn 2D background that would animate with the time of day and I tested it out with an image in game. It looked okay but getting it looking good was a big responsibility and no one on the team had a particularly strong background in animation.

Next, I made a huge day/night circle that would spin around based on the time of day. This looked good but not incredible. With a dedicated artist working on it and putting little markers around the edges with the time on I think this could have worked well.

What we actually went with works best I think. I was doubtful of how good it would look initially but using UE4's inbuilt landscape editor (and a keen artistic eye) has turned out really nicely for us.

I think it needs some more work but the way the shadows roll across the hills as the sun moves overhead looks really good and it allows us to do loads in the background for the events (zombie uprising, thunderstorm, plane crash, etc...).

You can also see from this image the new UI Connor made for us. I think it helps to show how the dolls are feeling at a glance. A new variable that I've added is HP too. Currently this only goes down if a doll eats part of another doll in return for HP (cannibalism) (The eaten dolls fear will increase and their likeness for the doll that ate them will severely decrease) (The doll that ate the other doll will lose morality and this increases their chance of doing immoral things in the future). Cannibalism only occurs if a doll is very hungry and in the room with another doll. Lower happiness and higher fear contributes to a hidden variable called 'Irrationalism' and this + low morality increases the chance of cannibalism.

https://www.youtube.com/watch?v=QHTCUcsFSbY - Follow this link to watch a video of the game in action in it's current state.

YEAR 2 - BA2b - New Assets and Intereactions

The ground needed modeling eventually and since everyone else was busy making assets to go inside the house I figured I'd make it. It kept the poly-count fairly low and the colours simple. I think the simple style looks nice and fits with the theme of the game. The idea is that we'd model a few basic assets that could then be placed on top and around the ground/ garden to make it look and feel more interesting and genuine. I took inspiration form Little Big Planet since the handmade style is similar and the way the ground is kind of a cross section is the same.

The flower is one of the assets I was going to replicate to make the ground feel more alive. In game I added a code  that would randomize its colour on spawn. I also made this as an example of what I was looking for from the team in terms of assets and style.

The two number under the UI represent each dolls 'likeness' for each other doll. You can see in the gif how Doll 2's likeness for Doll 3 is going down because they're spending time in the room together when Doll 3 is unhygienic.  Also, Dolls can take showers now to increase their hygene.

Dolls likeness for one another will affect certain actions they take. I will implement that next.

YEAR 2 - BA2b - Materials, UI and variables

I've been unable to work directly on the game over the holidays because I wasn't able to transport my computer home but since I've been back I've made a fair few updates. I've changed how the dolls move so instead of trying to get exactly where the cursor is, they move towards that place and move slower the closer they are to the correct place. This means they don't feel as snappy but they don't fly off as much when you let go of the left click. The problem still persists that the dolls sometimes mistake what room they're in though by not realizing when they enter a new trigger box so I need to fix that.

I've also added a few variables to test out how they change under certain circumstances. The gif at the top of the page shows how the total food decreases as a doll is eating and you can see how the dolls in the attic and basement increase in fear. They also have hidden variables that are randomised at the start which increase the rate at which they become scared and whether they enjoy time alone or with company etc . . .

A lot of the UI should go in the final build as I'm looking to express emotions through how the dolls act instead and I'd like it to be ambiguous how the dolls actually feel a lot of the time. This should lead to some mystery for the player in figuring out why a certain doll might be acting a certain way. the ambiguity should make some of the dolls actions feel slightly random (but still within reason) and therefor the player can make up stories and scenarios for what they are seeing happen.

YEAR 2 - BA2b - Darkest Dungeon and Negative Effects on Characters

Darkest Dungeon is a Roguelike which has some relevance to Doll House because of the way it incorporates multiple characters psychology into the core gameplay. I watched Mark Browns Analysis of the game and he states that "Darkest Dungeon ultimately makes you the bad guy but it doesn't tell you that in the premise or as a twist in the story. You explore your morality organically simply by engaging with the mechanics on offer.". Doll House doesn't want the player to be the 'bad guy' but I do want players to be able to explore immorality should they please and I'm supporting that by allowing murder, cannibalism, thievery, sacrifices to the devil etc as mechanics that the player can incorporate into their stories should they please.

Darkest Dungeon makes the player feel the results of negative effects on the characters by giving the player lower chances of success and therefor leading to more negative effects on the characters. This 'snowball' effect is something I've been thinking about for Doll House. I feel like I can make the mechanics more interesting and more realistic by making one lead into another, which is why the interplay between variables (emotions) is so important. Seeing our characters fall into desperation and potentially go against their beliefs will make the dolls in doll house much more life-like than characters in other games even though they follow their own slightly skewed sense of logic. Looking at the effects of Stress, Sadness, Jealousy, etc is something that many games don't tackle so I think it would be something at least worth trying.

YEAR 2 - BA2b - Multiple Dolls and Initial Asset Tests

With some extended discussion and deliberation I've managed to figure out how to efficiently model my blueprint to minimize the amount of code repetition. Essentially all the dolls are duplicates of eachother and they all have the same instruction EXCEPT each doll has a unique identifier that means that when they sinc their variables (every 5 seconds) with a global actor it saves them as individual variables within itself. eg. D1_currentRoom, D2_currentRoom, D3_CurrentRoom. Then when the dolls want to reference variables within other dolls they just have to check what the variable in the global actor is.
Anyway, this took a surprisingly long time to figure out and implement but I like it so far. It should make programming much quicker and therefor testing and making changes quicker.

Also, I made a super low-poly bed to see how it would look in game (since the camera is far out I told the team that they don't need to make particularly high-poly assets as a lot of detail would be lost).
This isn't my best effort but more of a test to see how low poly we could go whilst still looking good. It's safe to say that 13 polygons is too little.




I went slightly higher this time for some bedroom assets and concluded that about 100 polygons was about the minimum you could go without sacrificing quality but I'm letting the team make assets as high poly as they like as long as they're willing to UV map them.

However, given that ideally we need about 100 assets in the game I recommended that they keep the poly count as low as possible.

YEAR 2 - BA2b - Movement and Room Understanding

Currently I've managed allow the player to move a doll around the house by clicking and dragging as you can see in this gif.

I've placed trigger boxes in each of the rooms that the doll references to figure out which room it's in and it seems to work okay but the doll jumps around a lot when the player moves the cursor. I think this is because it essentially set the dolls Y and Z axis to whatever the trace from player to wall hits. ie. when there are bumps in the walls it causes the doll to 'jump' to different positions as it's constantly trying to the correct position. I might need to include some kind of minimum movement threshold or limit the dolls maximum movement speed.

I've also set homework assets for the team to create this week since we're going to need to populate the house soon and I need to program interactions between the dolls and the objects in the house.

One problem I knew this project would have from the beginning is repetition of code. UE4 makes it really awkward to access variables from within other assets and even for the most basic interactions to occur the dolls are going to need at least 5 different main variables each but this number will probably increase drastically. I don't want to have to copy and paste bits of code from one blueprint to another especially since I want the game to allow for more than 3 dolls in the house at once potentially.

YEAR 2 - BA2b - Design Document

The first thing I did once we decided on the idea was to write up a design document for the team to refer to and to make sure everyone understood the concept. As you can see, I included a development schedule and an overview before describing certain areas in detail.

I'll include a copy of the design document when I hand in.

YEAR 2 - BA2b - brainstorm

We all decided almost immediately that (3) was far too open and so any ideas that fell into this category were not right for the project. (1) seemed like a healthy challenge but any ideas we came up for it seemed too small for the size of the team and the length of the project. The two ideas that really stuck out to us were both in category (2):

[IDEA1] Cat Ship (working title) (if FTL was an RTS)
 - A side on 1v1 RTS set on pirate ships controlled by cats. The player must invest in various cat based weapons, defences and production as they go with the ultimate goal of destroying the opponents ship. The ship itself is stationary but the player can invest in different upgrades (the different rooms of the ship) and this determines what they will develop. Clicking on rooms in the opponents ship tells your weapons to focus that room. You can use this to stuns the growth of your opponents plans and/or earn you gold for more upgrades.

The game could include multiple unlockable costumes and ship skins with the potential to be monetized with optional micro-transactions.

[Good Points]
- Very marketable (everyone likes pirate cats)
- RTS is starting to gain popularity on ios (Clash of Clans) but has yet to fully engage the casual audience so it has the potential to fill a gap in the market.
- A good ratio of 2D to 3D with a healthy dose of design and programming (perfect for our team composition)

[Challenges/ Problems]
- The scope might be a bit small
- Similar theme/feel to a lot of other casual games (might blend in too much)

We also had plans for a (worms-like turn based version of the game)


[IDEA2] Doll House (subversive 'The Sims')
- A story telling game for lets players and fans of The Sims and subversive play (The idea originated from reading chapter 2 of Mary Flanagans book 'CRITICAL PLAY: Radical Games Design' and wondering what would happen if someone made a virtual doll house about subversive play)

A side view of a doll house with a 7 day/night cycle where the player has the ability to move dolls from room to room by clicking and dragging. the dolls each have their own randomly generated stats and desires and this determined how they might react to any given circumstance (taking into account what room they're in, who they're with, their mood, wants etc). The layout of the house may have some element of RNG too . It is up to the player to decide where to move the dolls but what the dolls get up to is of their own decision and the unexpected is to be expected. Over the course of 7 days in the house random events can occur including ghosts, mole infestations, zombie uprisings, squatters, plane crashes, alternate dimensions and more normal things like daily deliveries of packages. Whether the dolls end up happy, dead or something else is up to you (kind-of).

the idea is that if the player wants to they can try to keep these characters alive and learn about who they might be whilst surviving weird events and creating a new story each time OR they can subvert the play and try to get the characters into ridiculous situations where they are forced to murder/eat/attack/cheat on eachother or other entities in the house.

Also, after each playthrough the player can unlock new potential room types, character traits and objects.

The game could be monetized with expansions such as:
Medieval Edition  - torture dungeons, famine, witches, cults
Fantasy Edition - various fantasy races, dragons, questing

[Good Points]
- Very Marketable
- The kind of game that lets players would love to play given the huge amount of potential interactions and creativity they can bring to it.
- Never been done before (as far as I'm aware)
- High replay value
- Potentially cover real world issues and is quite shocking and weird. Should get lots of publicity on social media.
- Lots of fans of The Sims > No good Sims games released recently
- Good quantity of work to keep us busy for the whole 6 weeks
- Good ratio of work. The game should keep all 5 of us busy.
- Large scale but flexible. All we really need to make this good are the core systems and a few variables and interactions but it's almost infinitely scale-able and the more we do, the more open ended the game will feel.

[Challenges/Problems]
- Potentially boring players if they lose the feeling of being autonomous or nothing interesting happens
+ Possible fix would be to include achievements for various events the player can try to cause, lots of RNG, daily (ingame) events and packages with new objects the dolls can use
- Potentially very complicated programming and prone to bugs given the RNG and that it's not the kind of game I've ever made before.

YEAR 2 - BA2b - 1 Button Ruleset Clarification Classification

For the next 6 weeks myself, Connor, Tegen, Geena and Ben will be collaborating on a game with the theme '1 Button Game'. It's a rather vague mechanical theme rather than a contextual one so brainstorming was a little slow at first but it did give us the freedom to work on any pet projects we may have had in mind. Before the brainstorming session, myself and Connor discussed ways we might go about minimizing chaos (since last time we all worked together for the 4 day game jam the idea generation stage got fairly hectic). We concluded that meeting up in person this time would be beneficial, also I think we were more relaxed given the longer project length.

Before we met up I created a classification we could use to categorize our ideas and to outline the limits of our brief. This was to help organize our ideas and to more effectively analyze them.
The categories are:

(1) Strict - Literally 1 single button as the only input. Uses:

inputs - button press, button hold, button release

different contexts - eg. which part of the screen the character is at when the input is inputted.

Analogue input - how far down the button is pushed could determine of the character for example

(2) Mid - Allow non-button inputs too such as analogue sticks, mouse control and touch screen.

Possible custom controls like sliders, twists (like Bop-It)

Possibility for on rails shooter, point 'n' click, racing (where mouse_x = turn_rate) etc...

(3) Not-strict - Where the 'single button' refers to the 'action button' and other buttons can be used for other things.

(4) Other:
Subvert the theme - All button except for 1
physical buttons in the game that you can move around as a character
Single button with loads of different ways of interacting with it.

Continued in next post...

YEAR 2 - BA2b - Ambivolence Can Collaboration

For our 1 week collaboration project with the Graphic Design and Design for Publishing students we were given the task to re-purpose the object (Tin Can) with an emotion in mind (Ambivalence). The first thing we did was to discuss ideas for our object and a concept for a ballot box caught our attention. The tin cans would become containers of information and the public would place a ring pull in the can for the cause they felt the most strongly about. It was suggested that the question be something people are often ambivalent about and I suggested that the ballot be rigged to make a statement about how all the answers to the question are actually the same but reworded. I made us all a slack group and we used it to discuss ideas further and share files and research between one another.

The next day when it was pitched that rigging could be achieved through random chance via a hidden pachinko machine. The pins in the machine would be metaphors for everyday things that could sway your decision. The surprise is when the ball ends up in a can that you didn't expect and we reveal the influences that changed the outcome. A few of us started working on a playable mockup that would be shown on screen during our presentation. I did everything in UE4 (The general layout and programming), Matthew made the assets and Natalie made the 2D art.

By Thursday the mockup fully working but given that we still had 3 days we brainstormed some more ideas and Ross came up with the idea that the can could be a container of blank dice. You write outcomes on the dice and they make decisions for you to solve your ambivalence. Me and Matthew made another mockup using maya and UE4. I animated the can turning and the dice rolling out onto the game board. This was shown during our presentation.

A gif version of the animation we made

We made sure to include the audience where possible in our presentation where possible to make it more engaging and to test our product out. I think it went well. We got a good response to the presentation and we worked together as a team very well. I've kept in contact with some of the team members and we are planning on collaborating again at some point.