YouTube Icon Facebook Icon Twitter Icon RSS Feed Icon

Dev Blog #4: Level Design Workflow and Ideas

26
Apr

My name is Markus Fors. My part in this project is Lead Level Designer and Texture Artist.

I thought I should start from the beginning and explain to you my workflow and ideas about the creation process.

I started off by drawing a very basic layout of the entire level in Photoshop. To get an idea of how it should look like, I searched on the internet for blueprints of real ships. Of course I also had a couple of conversations with our project leader, to really have an understanding of what kind of rooms and how many, that had to be in the game.

Generally I’ve been using my own thoughts and ideas about how everything should look like (as long as you get the feeling of being on a ship at sea). After the finalized layout of the level and the initial preparations, it was time to start working with UDK. Keep in mind when reading the rest of the blog, that it’s based on my own experiences working with level design.

I followed my own blueprint and started creating the level using only simple brush based blocks.

I’ve been creating levels for about four years using Valve’s world editor Hammer, so at this stage I was very familiar with the creation process of the level. At this point not a single model or texture was implemented, apart from the default checkerboard-like texture on all the brush based blocks.

This made the level very, very basic but at the same time this was necessary, so that we later on had an understanding of what kind of models the art creators had to create.

I decided that the rear end of the ship should work as a junction point, connecting all other parts of the ship. Therefore I started with that part and then worked my way until pretty much the entire ship was finished in brushes. Luckily, myself and four other team members went on a small study visit to a real ship during the early stages of this project. For me as a level designer, this was an awesome experience and opportunity to gain ideas.

Finally,when different kind of props started to get finished, like walls, etc. I could start implementing these and replace the brush based blocks with models. This kind of work was very fun but also a bit sad, since I had to start deleting the brushes which I’ve built the ship with.

However one goal with the entire level was to have as few brush based things as possible.

When creating all the rooms and corridors I always try to keep in mind to get everything to be symmetrical, but at the same time to give the player a feeling of things being asymmetrical. By this I mean, a simple corridor could just be walls, a floor and a ceiling but that doesn’t really look all too good. For example, just by slightly decreasing the width of the corridor halfway through, makes a huge impact on the overall look and feeling. Now everything doesn’t feel and look square and boring. Also the use of props really helps to give the player a feeling of the environment being natural (like the pipes in the picture below, going around the corner of the wall).

 

A corridor.

A corridor.

Here’s another example from our game. Notice how the walls divide the room into different sections and makes everything feel less linear.

The kitchen area.

The kitchen area.

During my years of creating all kinds of different levels, whether it’s been with Hammer or UDK, it has come to my understanding that the use of lighting can make a huge impact on the overall feeling of a level or a particular room.

Let’s take the engine room in our game as an example. When it comes to the lighting, I was aiming for a quite dim and brown-ish light. Both to give a spooky feeling and a feeling of the room being dirty and well-used. All the corners of the room are dark, just to intensify the feeling of ”something might be lurking in the darkness”. However, making rooms completely pitch black is a very bad idea, since generally players hate to wander around in completely dark areas (even though they have access to a light source, like a flashlight).

As was the case with our game, I had an excellent opportunity to make use of light sources which act like a guide for the player. The small lights illuminate the path, but don’t actually force the player to follow it. They only act like a hint for him/her, pointing at which way to go.

The engine room.

The engine room.

The choice of lighting can also help your level/levels to feel more dynamic and less repetitive.

I always try to keep a certain kind of color scheme and depending on the room or location, I make small variations. Like with the corridors in our game, they have a blue-ish/white color and in certain places there are quite bright and yellow light sources. Just to make everything to feel a bit more dynamic and to break up the repetitive lighting.

A corridor with stairs and different kinds of lighting.

A corridor with stairs and different kinds of lighting.

When going further down the ship, the lighting changes to a more green-ish color. Both to act like a separator for the different kind of levels of the ship and to give more variation to the overall colors.

Also, the slight change in color indicates that the player has entered a new area of the ship. In other words, you can achieve a lot of things with your level design, just by changing the color of the lighting.

Alongside my level-design job, I’ve also been working with texture creation. My decisions when creating all the different kind of textures were based on whether I thought they looked good or not, and if they fit with the rest of the environment. (Take the above picture as an example.) The floor, walls and ceiling have the same color scheme. The only part that sticks out is the floor, but it still fits with everything else. The bottom spline on the wall works as a smooth transition from floor to wall. Also the white walls act like a more natural color, which works well with the generic corridors.

There is an expression that goes ”less is more”, which has come to be a sort of guide line for me. I always keep those words in mind when designing levels. And I guess it’s something you ”have to learn”, or get a feeling for, so to speak. Because who can say when you’ve placed too many props in a room or made the lights too bright. Well, it’s up to you as a level-designer and there is always something new to learn when it comes to designing levels.

Markus Fors
Lead Level Designer

 

Dev Blog #3: 3D workflow and character art

15
Apr

Well, to start things off, my name Magnus Christensson. I’m one of the co-founders of 12 O’clock Studios and I’m also Lead Art. The art of the game is my main responsibility and thus far for me it has been all about the game character creation.

For an experienced game artist this won’t really contain anything new, but I thought I’d go over my workflow concerning the character creation of our game. To be honest, I, myself, am pretty new to all this, but I’m a quick learner and I have learned a great deal during this production. I have enjoyed this production immensely so far, but we’ve still got a long way to go and I also have a great deal of things to learn that will be all new to me. For example I will have to learn how to use programs like MotionBuilder to be able to handle the motion capture data we will receive. But I will come back to these matters in a later entry.

As for my workflow with the game characters, here we go!

The first step in character creation is often based on the story, at least in the kind of games we are aiming at. Then after a character has taken shape in writing it is time for a concept artist to create a 2D image that represents the character from the story, taking into account the type of setting and traits that have been defined.

This is where I come in during this production. When the concept art of the characters are done it is time for my work to start. To start things of I create a lowpoly representation of the character in Autodesk Maya. This contains the rough outline of the body and the basics of the clothing. What is important to think about in this step is to get the volume about right in size and try to make the faces of wich the character is created about the same size and that these faces only have 4 vertices.

I also do the UV layout to be able to use Mudbox (a sculpting tool which I’ll talk about in more detail later) as a texturing tool, if you are planning to use Mudbox for texturing remember to not do any mirroring in your UV or your texture will automatically be mirrored while you paint it. (This I learned the hard way). For those of you that have never used a 3D tool or anything of the sort, making the UV-layout means that I create an image of the character in 2D, and by cutting along edges in the model i can unwrap the 3D shell so that it lies flat in 2D. This makes it possible for me to paint it and thus give the model a texture.

First steps

First steps

Left: Basic outline and clothing

Middle: UV-shell of the same character, this UV shell was my mistake and shows how NOT to do your UV-layout if you are planning to texture in Mudbox as it is just half the model and then mirrored. Although the image might help you understand what an UV-layout is all about.

Right: All faces only have 4 veritces (corners), and are about the same size in each part of the body.

Next up is exporting the lowpoly from Maya and bringing it into the sculpting tool, in this case Autodesk Mudbox. In Mudbox I start by adding density to the model by using subdivisioning.

For every sudivision I make every face is divided into 4. So by only doing a few subdivisions I soon reach a face density that will be easy to sculpt and will give me the possibility to add alot of detail to the model. For this project I have been a bit over the edge of what I need for the detail in my models – as my computer can handle it well I used up to about 2.5 million tris in a single model. However I usually brought it back down before exporting it to about 800 000 tris. I chose to sculpt the character faces in Mudbox. You can, of course, do a rough face in Maya before exporting, but I prefer to sculpt the face in Mudbox. Although when sculpting the face from scratch in Mudbox remember to give the face some extra density in Maya so that when you subdivide in Mudbox it automatically gets higher density than the rest of the body. This ensures that you are able to give the face the detail you want. When the sculpting is done the highpoly model that I will create my normal map from is finished.

Sculpting finished, Face density of the jacket at bottom.

Sculpting finished, Face density of the jacket at bottom.

Next up is the retopologizing of the model. As the highpoly model I sculpted in Mudbox differs so much from the lowpoly I originally created in Maya I cannot use the lowpoly to apply my normal map. Also, my lowpoly was made for sculpting in Mudbox and by no means to animate in any case. So when creating the new lowpoly I now have to think about topology, which means that my edges in my model have to follow the flow of the muscles of the body. This makes it essential to have some kind of idea of the anatomy of the human body. Also, to be able to use a normal map on a model, the volume of the lowpoly must be as close to the volume of the highpoly as possible, which in my case makes it even more impossible to use my original lowpoly as there is a lot of volume missing on muscles and basic attributes (as the nose and ears are completely missing).

To create this retopology of my highpoly I export an instance of my highpoly. Although I only need the volume and not all the detail for this, I step back one more subdivision before exporting to about 300 000 tris. Then I import this model into a program called 3D-Coat. This program specializes in retopology among other things. I import as a reference mesh and use the Retopo mode in 3D-Coat which lets me create a new lowpoly shell directly on the model. As I mentioned before this is where it is important to take into consideration the topology and anatomy of the character you are creating.

Retopology of my character in 3D-Coat. Edgeflow after the topology of the muscles are essential, especially in the face if you are planning to have your characters talking.

Retopology of my character in 3D-Coat. Edgeflow after the topology of the muscles are essential, especially in the face if you are planning to have your characters talking.

When the retopology is done in 3D-Coat I export the shell into Maya, here I do some small adjustments. I add seperate eyes as these need to be able to be animated so the character can look around. I also create the inner parts of the mouth and add teeth, gums and a tongue. After doing this I do the UV-layout for the whole model and combine it all into one object. After this it is time to go back to the highpoly from Mudbox and by using a tool called xNormal I use the highpoly model as reference and the new retopologized character as the target for the baking of the normal map. Then xNormals automatically generates a normal map onto the UV-layout of the retopologized model.

When this is done I can actually apply the normal map I generated from my highpoly onto this retopology in Maya.

The result of my normal map applied onto the retopologized character.

The result of my normal map applied onto the retopologized character.

My work with the model is almost done. The last part of the journey is the texturing of the character. I used Mudbox to layout basic colouring and projected some texture onto the model and the rest was done in Adobe Photoshop. After the texturing of the character was done there was still one last step for me to do and that was the hair, which I have made plane based. That means the hair standing out from the head is simple 3D planes textured with alpha in Photoshop.

The character textured and with hair applied.

The character textured and with hair applied.

This concludes my workflow with the characters of our game thus far. After these steps a rigger is making a skeleton inside of my 3D model and then uses that skeleton as base for skinning the model. (Connecting the 3D shell to the bone structure and making them deform properly while animating, considering muscles and anatomy once again.)

After that it’s time for animation, and in our case that will include motion capture and other neat stuff which I will be blogging about later.

Making the complicated simple, that is creativity”

Magnus Christensson
Lead Art

Dev Blog #2: Music Composition Workflow

8
Apr

Allow me to introduce myself! I’m Fredrik Häthén (known as @SakimotoFanboy on Twitter), and I’m handling the music in The Horror at MS Aurora. I’m also Lead Sound Designer, which basically means that I get to boss my friend and colleague Calle Leppäjoki (known as @DBHGamer on Twitter) around all day. It’s a difficult task, but someone has to do it.

I thought an inside look into the music composition for this game would be fun for you to read, so that’s what I’m writing about this week.

This is, quite possibly, the most difficult and simultaneously the most rewarding project I have worked on to this date.

Now, I’m not saying it’s hard to write music for Aurora. On the contrary, I find that the music for this game sort of writes itself, as long as I keep an ear open for the different ideas that flow through my brain. I have established a central theme and a general thematic idea for the soundtrack (which I will discuss in more detail in a later devlog), and I know which instruments will primarily be used for the different parts. What do I mean, then? I’m merely saying that it’s not easy to write music as the project is slowly forming beside me.

Howard Shore pretty much did this with his well-known Lord of the Rings film scores. Essentially, the films were being cut and re-cut every day, and he had to write more music or edit his existing material to fit the action on-screen. That is sort of what I’m doing, too; seeing as this game is being produced right now, and I’ve been involved in some plot ideas and discussions along the way as well, it simply isn’t possible for me to come in at post-production and score a (mostly) finished product. I’ve had to adapt several times to changes, edits and shifts – and it’s been extremely enjoyable thus far!

In this week’s devlog, I thought I would mention the changes in workflow I’ve been through over the course of Aurora‘s production thus far.

I started out much like I always do when I approach a game that needs scoring: I looked through the available material (i.e. the script, concept artwork and inspirational footage) to get a feel for the game’s atmosphere. Seeing as I knew pretty much what the story was about, I could bypass that and go directly for the actual feeling of the game. Sadly, there wasn’t a whole lot to go on. I started out by composing “MS Aurora Weighs Anchor” (the main theme of the game) where I created a basic piano motif heavily inspired by traditional Swedish music. After that, I began writing an orchestral score for real musicians, seeing as having a live orchestra perform the music was the original plan.

Score page from "Aurora Weighs Anchor"

Work in progress score page from "Aurora Weighs Anchor".

The whole live orchestra deal got scrapped pretty quickly, though, seeing as there was no room in the budget for it. Now I faced another problem, which was the fact that I had to create the music entirely through computer instruments. I adopted another way of working, where I would write an orchestral score first and then “re-orchestrate” it for a computer orchestra. That didn’t work very well for me and slowed my workflow down to a crawl, so I began writing brief motifs and ideas in my sequencer and subsequently send what I’d written to Victor.

These pieces did not always match Victor’s vision, however, so I decided that creating a “track list” based on the game’s plot and then writing music for those “tracks” would be a good idea. I began to work, but it soon turned out to be way more difficult than I had initially thought it would be; writing music for set “track titles” was, if anything, more limiting than my previous method of simply trying out short ideas.

I then started to do something I’ve never done before. I read through the game’s script and picked up key areas that inspired me. I then turned to my sequencer and wrote short tracks based on that inspiration. Said tracks would receive a short, descriptive name such as “Talking 01″ or “Sorrow 02″. Suddenly I was able to crank out at least one track a day, and Victor enjoyed what he was hearing. So did I!

A work in progress in Cubase 5

A work in progress in Cubase 5.

I’m now in the process of simply getting things done, speed-composing and getting tracks written every day to fill up a list of cues that may or may not make it into the game in the end. In any case, I have learned a lot about how I work and the many ways I can find inspiration. It’s been a lot of fun thus far, and I think the fun’s only just beginning.

Yours in music
Fredrik Häthén
Music Composer & Lead Sound Designer

Dev Blog #1: Dawn of the Episode-based Cinematic Horror-short

1
Apr

Hello world!

A cheesy way to start – but somehow appropriate. I’m 12OclockVictor, and I’m here to tell you a bit more about who we are, and more importantly, what we’re up to (chances are that you’re already aware of our current work in progress, The Horror at MS Aurora).

Who are we?

As stated above, I’m 12OclockVictor, one of five co-founders of 12 O’clock Studios where I’m currently working as producer.
Never heard of us before?
Don’t feel bad. 12 O’clock Studios is a relatively new kid on the block… in fact, we’re currently working on our very first title. We’re going into the business with a bang though! But more on that later.

What are we trying to do?

To be honest, “Episode-based Cinematic Horror-short” – while a very attractive amount of buzzwords for selling the concept to investors – might be a bit much for the average person to decipher. In fact – even I can’t explain it easily in one sentence (unless I completely forgo punctuation) and I’m producing the thing!

Therefore, I’m going to break it down into segments:

The fist and most obvious bit is “horror”.

Here at 12 O’clock Studios, we love horror, and we’re expecting the vast majority of our titles to fall under this genre. We will, however, be separating from the large bulk of existing horror games – the kind where the “scare” comes from monsters jumping up behind crates, and if you shoot enough of them you save the world. No such things here, no sir. At 12 O’clock, we look to the old masters: Poe, Lovecraft and King. Horror that’s lurking in the shadows – patient and inevitable. You won’t be saving the world – but if you’re lucky, you might be able to save yourself…

But how will we achieve this?

The keyword here is cinematic. When I said that we’re entering the industry with a bang, I wasn’t kidding – already in our first production, we’re going all the way with state of the art motion capture animations, brought to us by a high-profile specialist we will hopefully be able to name in the near future.

Animation, of course, is only half the story. Gameplay and story are the other two pillars on which our premier title will rest. But I’m rambling. There’s a ton of interesting topics, but only so many words. (And let’s face it. The average attention span is only that long…)

So let’s move on quickly.

We’re not just calling it horror. We’re calling it “horror-short”. Why? What is that? Basically, The Horror at MS Aurora will be a rather short game compared to some of its competitors, weighing in at a modest ~2 hour game-time. I can picture half of you frowning about now.

But this is why it’s good for you:

1. Shorter game does not equal lower quality. (Rather the opposite!)

2. Shorter game means shorter production time. (That’s more games for you!)

3. Shorter production time equals lower prices!

That’s right: The Horror at MS Aurora won’t just be an amazing and immersive experience – it’ll also be quite an affordable one!

That only leaves the question of episodes.

Yes, there will be episodes. Does that mean The Horror at MS Aurora II – the Horror Strikes Back?

The answer is: “No. No, it does not”.

Rather, our hope – our vision – is to continue producing and releasing titles of a similar format (Cinematic Horror-shorts) as completely stand-alone stories, based on the same concept: short, immersive, high quality stories at a low price.

It’s a rather new concept – as far as we know, it’s never been done before – but we’re hopeful that you will like it. That is, after all, why we’re doing this – to entertain you.

So what can you expect to see here in the future?

As you no doubt have noticed, even though I’ve given you a wall of text, I’ve only touched the surface of many things – all of which will be covered in future Dev Blogs. So whether I’ve managed to pique your interest, or you’re simply feeling that things can only get better from here on – keep an eye on this space! (as well as here, here and here, of course.)

Be afraid. Be very afraid…

Victor Dahlqvist
Producer

Twitter