Daniel Crewdson - Dev Journal

Today was the last day we are able to work on Delve.

Me, Sam and Ben came up to the studio and starting looking for bugs and any last minute changes that needed to happen. There was a bug with the player prefs, when the player switched scenes, they would not keep some of their stats, specifically their scrap count. This was pretty simple, for some reason I was setting the scrap count to zero when the player switched scene, I must have forgotten to take this out of the scene change script.

This now means that upgrades and currency are 100% transferable from scene to scene. Which is a nice place to be in. 

Me and Sam decided to try and sort out this jumping bug. We opened the script and started re writing the jump system. We set it so that there player has a bool that will set true if they are on the floor, then that will turn off when the player jumps, but this was occasionally not turning back on. 

It turns out that when the player was stood on tall walls, the recast couldn’t find the floor beneath the player, as it couldn’t reach the middle of the object we were recasting to, and thus, not flagging the bool to true. This was a simple fix that didn’t really require much code change. We just had to make the floors thinner and put a new wall beneath that actual floor ‘tile’.

We still needed the game to actually end when you kill the final boss. So I came up with the idea that, staying true to out initial concept, when you kill the boss, after a certain amount of time, the floor will destroy and you will fall and trigger the credits scene. This really wasn’t too hard to implement at all. I went into the SlimeBoss script, in the method that handles his health, told it to set a game object active when he died. That game object has a timer on it, and an array filled with the floor tiles. After the time is up, the floor tiles delete, the player falls and hits a collider that switched to the credits scene.

And there it is! Delve is finished. It has been a long hard journey, but I am so proud of what we have accomplished. We have had some issues with team members not necessarily pulling their weight, but I guess that is what comes with group work in first year. I am really looking forward to next years project, especially now that I am feeling more comfortable with C#.

This week I got the rest of the gun textures in. Every gun in the game now has its final texture and model. These look really good and we are almost complete in terms of putting in all the final art.

Flame Thrower - 

Grenade Launcher - 

Plasma Pistol - 

Saw Launcher - 

Now that we had these textured models, I was able to make the real weapon pickups. I replaced the boring rectangle model with the weapon models, and gave each pickup its own light source so that they are visible from a distance.

I also got the last three upgrades working in the game, and all locked in with PlayerPrefs,  this meant I could now get the final art in for the upgrade terminal. The lat three upgrades were, hair trigger, this fires an additional projectile for each click. The Spiked Grapple, this causes the grapple hook to deal damage to enemies. The amount dealt is pretty small, but its a good skill-shot finisher for when you run out of ammo on a weapon.

The last upgrade was a last minute idea, but it turned out really well. The shoulder cannon. This fires a homing missile, with a cooldown of 3 seconds. It is ridiculously powerful, so we decided to give it a stupidly high price, the player would have to clear all of level one and most of level two without buying anything to purchase this, but it will make the final boss a lot easier.

Getting all these upgrades working together was quite a challenge, most of them are just editing variables in other scripts to activate a method, but the shoulder cannon was a pain getting it to work with playerprefs. 

I actually took to maya to model the missile that the shoulder cannon would fire. I was interested in how the modelling process went, and I wanted to learn normal maps and uv maps. So at lunch on Thursday, I opened maya and got to work.

I was able to transfer a lot of my skills I learnt in college using Cinema 4D to maya, so this was quite simple to model, but I had never made a normal map / uv map before.

I unwrapped the model and got to work placing the texture on all the segments.  I like the workflow of normal mapping a model, laying it all out in photoshop provides a nice area to work with. I can see myself venturing into modelling a little over summer perhaps for a personal project.

I got the particle effects in for the homing missile explosion and the grenade launcher explosion. These will help make the weapons feel more responsive, as it is a lot nicer seeing an actual particle effect rather than a default unity one.

The pause menu UI was finally finished too! So I got that implemented. I really like the design on this, I love the way it reflects its design on the upgrade terminal with that circle in the top left. This creates a persistence in our world, which is what this game needs.

We had our main pitch on Tuesday! The general feedback was really good, we were told that the visuals have come a long way, and that it is really refreshing to actually see art built up in the game.

One thing I am really proud of is the fact that we had an answer straight away for every question we were asked. I believe this shows that as a team we are really confident when it comes to communicating our product that we have spent so much time on.

One thing that the pitch has made me realise, is how much work we really still need to do. The game is far from being in a finished state, and we are all very aware of that. There are going to be some late nights in the coming weeks!

This week I got the rivet gun model and texture in the game, and it looks really good! I am a massive fan of the work Russell has put into these weapons.

I also finally got round to fixing the grapple hook. I have been putting this off for a little while, as I know it will result in me having to jus re-write it entirely.

Before, the grapple hook system was spread over three scripts, it is now simplified down into one script, and with none of the bugs. It essentially works in the same way, but now the player is able to fire the grapple hook whilst another hook is already in the air. Before hand, this would break the system all together, and you would have to reload the game 

With this, I also got the grapple hook model in place of the little cube we were making do with before. After the implementation of this, I was taken away with how good the grapple hook feels to use. Even if we are cutting it close, I can see this game actually turning out to be something great.

Monday was out last ever sprint plan!

We have a pretty basic goal or just finishing the game. We are going to do our best to not add too much this sprint, as that will just sky rocket our workload. Obviously there are some things that we just have to add, but we can’t be too lenient.

We also made a start on our final pitch for our game, we have decided to use the slideshow we used for our initial pitch as a basis to work off, and change the slide information around to fit what we have now. We have also made a short video, showing off the assets in our first level, as we are finally starting to get the ground clutter in and really fleshing our the environment.

Sam got working on the AI for the enemy swarmer, they now leap down and ambush the player! These work really well in the points that Sam has placed them, but I’m sure one of the designers could go through and point out some good places to these guys to hang from.

Easter  break is over!

Over the easter I have been working on my programming assignment to create an 80′s arcade game of our choice. I chose to recreate 1942, the war plane top down shooter.

I managed to learn a lot of new things in C# in regards to writing maintainable code, and using the single responsibility mind set. I also did my best to make use of interfaces and inheritance. I created an interface that allowed me to apply damage to different enemies the same way, with a modifiable amount of damage being taken, depending on the enemy.

I did what I could to implement a singleton in the project, I started using one in my score manager script, but there is only one level in the game. So there would never be another score manager to even activate the singleton pattern.

This week was really good in terms of progress! 

First off, Sam finished coding the cyborg boss in level 1. It feels so good to finally have a level we can play through and complete! When he had finished that I jumped in and got the weapon dropping working from the boss.

I got to work on coding the last weapon we will put in the game, the Rivet Gun. The rivet gun first powerful projectiles in a very accurate pattern, this is in my opinion going to be the most powerful single target weapon in the game.

I also got some more of the gun models in! The Grenade launcher and Flamethrower models are now in the game and in place of the cubes I was using before.

I love the way these look, and they carry across the corrupted sic-fi theme our game is pushing.

We also finally have all the environmental assets in the first level. This is a zoomed out view of the first level, now that all the textures are in, even from the outside it is looking really good! One thing I would like to note here, the use of unity tags. I have never used these properly before, and I wish I had. They make finding objects in a busy scene so easy. Something I am going to a lot in the future.

We also got the signs in the tutorial level. These signs help explain the controls and teach the player how to pass each obstacle. I sometimes forget that there is quite a lot to learn when it comes to our game, so the tutorial is as essential of a level as the others.

We had our sprint review on Thursday. We were told the progress is looking good overall, but we still need to get more art into the game. Once we have the rest of of the art in the game, it will feel so much more like an actual fleshed out product.

On Tuesday was the Show and Tell day! 

Me and Luke had to be down at the Phoenix Cinema in Falmouth for 8:30am, to prepare for the presentation at 10:45am. I felt that we spoke very well, we managed to communicate our game across clearly, and a video we had prepared helped with that a lot. We had some great feedback and questions from the other members of the Games Academy, which was nice as people seemed very interested in our setting and mechanics.

This week Sam got the Slim boss and Slime enemy coded at a basic level. The slime boss will be the boss in the second level, and will feature an attack that spits out an acid pool on the floor, the pool will then spawn small slime enemies that attack the player.

I got the grenade launcher and the flame thrower coded. These are our next two weapons that the bosses can drop. I also got them working in the weapon switching script, and created weapon pickups for each of them.

I also started putting in some more upgrades into the terminal and working, these were:

Shield cool down/duration 

Longer/Faster dash

Increased health

On Monday we had our third sprint plan!

We really went into detail for this sprint plan, and outlined everything we wanted to complete in some detail:

Design - Another final level - 8 more upgrades - UI design

Art - 1st and 2nd level fully textured & top notch - One more boss model (+ one more enemy) - Rivet Gun, Flame Thrower and Grenade Launcher

Sound - Tracks for each level - All sound effects implemented

Animation - Animation for Robot Boss, Crawler Bot, Drone, Cyborg Boss

Programming - Program the guns above - BOSS AI - UI menu for Upgrade - UI menu for Audio logs

Writing - Boss Dialouge - Upgrade description and flavour text

I am very happy with this sprint goal, and I believe we can achieve everything on this list.

There was a pretty big change to the game this week. We have decided to shift around the levels. What was level 2 is now level 1, as it is more of a top level themed environment, that works as the top level of the facility. It has offices and security rooms, so this just makes sense in terms of progression. And what was level 1 is now level 3.

In terms of environment art, we are looking pretty good! The textures are almost complete and soon we will be able to get the models for the main assets all in the game and textured.

The cyborg boss model has been finished, and Sam has started coding the AI for that fight, which will take place in the new level1. I have got to work on some more upgrades and getting them working in the upgrade terminal. At some point, when all the upgrades have been designed, we will need to sit down and discuss the different scrap prices each upgrade will have.

We made some progress in terms of getting the story into the game this week. In the 2nd level there is now a small box on the table in the security office. When the player walks up to it and picks it up, a text log appears from one of the security guards at the side of the screen. It is really nice to see some story making its way into the world, makes me feel like we are creating something bigger than just a game, we are creating history within that game.

We also got to work on a tutorial level, this will outline the basic mechanics in our game, (dashing, grapple hook, etc)

We also had our sprint review on Thursday. Doug played our game and he really enjoyed it! In his words, ‘This game is brutal!’, I think we may have made the combat a little difficult, but we want players to learn their way around the environment efficiently, and to do this they must make use of ramps to dash up, and grapple points. The enemies fire so fast that you can’t afford to sit still and just fire at them.

We started working on our presentation and video for the show and tell day, I put myself forward to speak alongside Luke, I feel very conformable communicating our game and its surrounding features.

The first few assets are in the first level, and we have great progress on the level 2 whitebox!

We have finally implemented our currency system, scrap. Scrap drops from enemies when they die, and the player will use scrap to purchase upgrades at upgrade terminals. 

Speaking of upgrades, I turned the plasma overcharge from a few weeks back into a purchasable upgrade.

This is everything do with the upgrade terminal script, (Ignore ‘RocketLauncher’, when I first made the saw launcher, I named it rocket launcher as we still didn’t know what weapon it would turn out to be. I have not got round to changing its name), When the player opens the upgrade terminal, all these scripts stop running, stopping the player from being able to move, shoot and look around. 

I also got the weapon drops working, You are now able to pick up the saw launcher after it drops from an enemy. I made some dummy weapon objects to test the random chance of spawning a drop. Again, ignore the names wand and gun4, these will soon be real weapons.

Once the script has chosen a pickup to spawn, it instantiates one at the emeries death location. And there they are, bright and spinning, just like real weapons?...

On Monday we had our first sprint review! 

Some of our most important feedback was on the visuals of our game so far. We were told that even though our game is still in the white boxed state, we should still put the base colours on the walls, floor etc. This will make the game a lot nicer to work on, rather than just staring at white everything. 

We were also told to just put everything have into the game rather than waiting for everything of one thing to be done. By this I mean for example, rather than waiting for all the assets for the spawn room to be finished, put them in as they are completed.

We then had our sprint plan the next day. in which we outlined our goals for this sprint:

Two levels built 

Weapon pickups working

Get a soundtrack in for the first level and boss

Two enemies designed, modelled, textured and implemented

Some animation in the game

2 more weapons modelled, textured and implemented

Fix the grapple hook

I really like the sound of this sprint plan, it revolves mostly around bulking our the game and giving it more depth. Really excited to get back into development and build on what we have.

I decided I wanted to change the way the grapple hook worked. Rather than just aiming at a grapple point, pressing a button and moving towards the centre of it, I wanted the player to actually fire a grappling hook and move towards the hook. I spent a little while getting this working, but it feels so good to watch the hook fly out and latch on to a grapple point.

Essentially it just waits for there to be a game object tagged ‘hook’ in the game, then waits for it to collide with a grapple point. When that happens, the player is moved towards the hook. When the player hits the hook, it is destroyed, grapple ammo is set back to one, and the player can fire another hook.

I managed to get a basic pause menu working. This makes working on our game a hello of a lot easier. It seems like a small factor, but being able to press esc to unlock the mouse and change values in the inspector, rather than holding escape and trying to find the cursor. This already has saved me some time.

This week our two writers, Neil and Connor, finished the basic backstory for our game and world. We also have a Drone model made ready for the AI to be scripted on.

I finished the 2nd weapon in the game, the Saw Launcher.

At the moment it uses the same reload/ammo bar as the plasma pistol, and this is just the basic model, but the functionality of the weapon is present. The saw launcher pretty much does what it says on the tin. If fires spinning saw blades which will deal a large amount of damage.

So, it turns out random level gen is harder than it seemed. We have had to make a pretty hard decision and scrap the idea of random generation, in favour of making premade levels. I believe this will allow us to make more use of the grapple hook and the dash, as we can design sections of levels around these specifically, rather than maybe getting a good random level.

I got started on creating the weapon pickups that the bosses will spawn on death. This was pretty simple, I just made a cube game object into a small rectangle, put a script on it that rotated it around the y axis, and then set its tag as PickUp. I then made it so that when the player walks over it, the pickup object is destroyed. Now I just need to get to work on the second weapon!

One of our designers, Ben, has started white boxing our first level.

This was quite a quiet week in terms of development.

I think most people are just trying to decide where to start. I worked some more on the plasma pistol that we used in the prototype from last week. 

I pretty much recreated the script for the plasma pistol, I also added in the alt fire for it, an overcharged shot, that consumes all 10 ammo in exchange for a bigger, more powerful bolt. I can see this functionality being obtained via the upgrade systems we talked about during the concept stage.

I also got a rough ammo UI element made, that counts down when you shoot. This is nothing more than a bar filled with smaller green bars, one shot removes on bar, one overcharge shot clears them all.

We also got the early model for the plasma pistol this week from our last artist, Russell. I really like the way it looks, and I think it fits the concept for this weapon.

I also managed to get the player dash working. Once Sam had finished the player movement script, I stepped in and got working on the players dash. When the player double taps in a direction, they will dash forwards in that direction a short distance. This actually feels really nice to use, and fits in well with our rogue-like theme.

We had out pith this week!

The tutors seemed to really like our general idea of an FPS/Rogue-Like mix. As it is so unheard of they had few questions as to how our game fits into both, which I feel we answered pretty well.

The biggest criticism we received was about the way we describe our games setting. ‘Waste disposal’ made the tutors think about a rubbish dump, rather than the neo-noir technological facility we had in our heads. I guess it is hard to communicate a visual style to someone who has not been with us through the entire concept  stage. 

We are going to move away from the waste disposal, and more towards an abandoned scientific testing facility. 

On Wednesday we had our first sprint plan. Each member of our team planned out their tasks for this sprint, with a time estimate for each task on trello. We also came up with an overarching sprint goal:

A vertical slice of the game, demonstrating the first level; this will consist of a randomly selected level, with a single functioning boss fight at the end. Assets for one boss, the main character, a basic enemy, the starting weapon and the environment.

I feel this a very achievable goal, I will probably end up making more than just the starting weapon at least working in the game. The randomly generated level might be a struggle, but we will see how it goes.

This week was pretty fun.

We got to work on making a simple prototype, which includes a basic plasma pistol weapon, a grapple hook and a basic version of how we want the UI to look! It will be really nice to show that we are already thinking about the movement and basic combat in our game during our pitch.

Speaking of our pitch, we spent a couple of days in the library this week in one of the group areas, I plugged my laptop into the screen on the wall and we sat around and designed out pitch as a team. It was really nice to have everyones input in there specialist routes and we seem to work really well as a team, I have good feelings about the future. 

One of our other artists, Luke, has been connecting some character styles.

We have decided on a female protagonist. I really like the 3rd and 4th design, the 3rd one looks pretty similar to Isaac in Dead Space which I like, as it really shows off our sic-fi themes. And the 4th one looks almost ninja like, which I also really like, as it shows off the nimbleness of our players movement style, grapple hooks and dashes etc.

Izzy also drew up a quick idea for a boss encounter environment. This style of multiple heights would be good as it can provide the player with creative freedom when using the grappling hook to get around. It would also open up the possibility of designing the boss around gaining a height advantage to kill them.

This is a basic idea for one of our enemies, a swarmer ground type enemy. They will latch onto ceilings and ambush anyone who walks under them. I like the idea of these being a constant threat to the player, forcing them to really soak in their surroundings and watch every high corner.

This was the first week back after Christmas, it actually feels really weird to be back after such a long time off, but I am excited to actually start development on Delve!

Since returning to uni, we have come together as a team and talked about falling as our main travelling mechanism for getting from boss to boss. Rather we can keep the falling after each boss fight, but just as a transition from each boss to the next level. 

As our levels will not be vertical chutes now, the player will now walk through grungy waste disposal tunnels/levels, or boss made routes, fighting smaller enemies, and end up at the next boss.

These are some concept images that our artist Izzy drew, even though we are not going with the falling concept anymore, the visual style and feel of these concepts carries through to our theme.

Izzy also drew how she imagined one of the bosses looking. I love the idea of an old corrupted machine, with wiring hanging our everywhere