A group of young German developers, who previously worked on Goodgame Studios and Gameforge have joined forces to build an amazing open-world project Lost Ember. It’s a complex exploration game, build with UE4. In our interview the team discussed production, mechanics and other important elements of the game development process.
We are from all over Germany, basically, and met 2014 in university in Hamburg in a Game Development Master’s program. All of us worked in different companies in the gaming and media industry before, mainly in German ones like Goodgame Studios or Gameforge. After working together on a few smaller projects during our Master’s program for some time, we saw that we worked really well together and decided to found Mooneye Studios.
We started with four guys. Max as our concept artist, Matthias as 3D artist, Pascal as a programmer and me as programmer and technical artist. And very soon we got our narrative designer Florens on board.
I think the key is (as always) communication. All of us can take criticism very professionally, although we all are quite perfectionist and love what we do. If there is something that doesn’t feel right, even if you worked on it for weeks, it has to change. That is bitter of course, but in the end everyone knows that it’s the best for the project. This happens quite a lot, actually. Especially in the early months when we were still finding the right mechanics for Lost Ember. But we managed to not take it personally and keep being motivated every time and put all our effort in searching the right way to do it.
Lost Ember started almost two years ago as a rough vision of a game that lets you lean back and dive into a world full of possibilities. In the early months we developed multiple prototypes and changed the core mechanics pretty frequently. We were trying to cut everything that distracted from our main idea of a relaxing and still fun game that lets you enjoy our world at your own pace. We got rid of too complex puzzles or enemies that could be frustrating and only hindered you from getting where you wanted to go and see what you wanted to see. Now there are only the different animals that are really fun to play around with, a companion that can show you old memories of a fallen world and a story to uncover.
The world is already pretty big and we’re still not even close to the size it will have in the end. Although we’re aiming at a playtime of about 5 hours for the main story, you can definitely walk, fly, or swim around in an area for a few hours and still find things you haven’t seen before. Especially because you see the world from very different perspectives. If you run through a forest as the wolf, there may be lots of things you will miss and only see clearly if you fly above the tree tops with a parrot or dig around under small bushes as a mole.
To do that with a reasonable performance, the world can’t be all loaded at the same time. We rather created different zones as you said and only keep the closest and most important ones loaded while the ones further away may be completely hidden or replaced by simpler versions that look the same from the distance. A high mountain with springs, forests and lots of rocks may just be a painting if you’re too far away. Unreal already offers the functionality of simple Level Streaming that we use for this purpose.
That’s definitely the main feature, yes, and we basically built the whole game around it. We wanted the players to be able to use the different perspectives to get a more complete image of the world with parts you probably missed as your main character. It obviously was not easy to decide which animals we wanted to implement and which ideas had to be dropped. But our resources are very limited and we wanted a wide range of different and interesting characters, so we had to drop some of our ideas.
As a main character we pretty soon landed on the wolf. We wanted a main perspective and movement capabilities not too different from our own as humans and, well, wolves are pretty awesome. After the more obvious next choices like birds or fishes, we also wanted something a little more interesting that the players might not be very used to. For example our mole that turned into more of an armadillo-mole and seems to be very popular among our test players or the mountain goats that let you climb steep cliffs. There are a few more that we’ll keep a surprise for now, but I can promise they are pretty amazing as well.
Unreal definitely offers a lot of things out of the box. You can pretty easily setup different characters with their own set of skills, controls or camera settings. That came in handy for switching between very different characters like you do in Lost Ember. But we also had to add or modify a lot of things to suit our needs. For example smooth camera transitions from one moving character to another or systems for moving and animating quadruped or winged characters. Most of the more complex Unreal parts are designed for humanoid characters and we don’t have too much of those. But fortunately Unreal’s Blueprint system (a visual scripting system that enables you to do lots of stuff directly in the editor without opening a code editor, for those not familiar with it) lets you prototype new features pretty easily and fast. Most of the time our workflow consists of trying different things in Blueprints and then, when we have a proof of concept, translating it into code where we can fine-tune and optimize it.
Stylized forms, but high quality rendering. That’s basically the key for our style in Lost Ember. We chose this to make sure that we can get to kids and adults equally and have an easy way to transport different emotions or story details in different parts of the game even though there might not be any kind of narration or dialogue.
The mood of the current moment defines what kind of form we use. For example we use sharp edged forms for aggressive or frightening situations and rounded and clean ones in moments of joy. The story plays a big role in this as well and tells us from the beginning to the end how the forms have to change.
Building Virtual Spaces
Unreal does take a bit of the work off our hands here as well, as you can “paint” different assets like foliage onto the ground pretty easily. But that only gets you so far. The biggest part of the work in Lost Ember is definitely set dressing. We want to create a very unique and interesting world and for that have to work on every small detail just as long or sometimes even longer than on big landscapes. The production of the assets itself is a relatively easy part and doesn’t take too much time (most of the time). The difficult thing is the composition as every scene is Lost Ember is carefully designed.
We developed a workflow that works great for us: at first we have a rough concept art of the area that is then built in 3D and brought into the engine. As soon as that starts to look as it should, we take screenshots and use these for more detailed concept arts where we just paint all the remaining details over it. These overpainted screenshots are again built in 3D and the scene is complete.
Yes, every plant and tree is handcrafted. We tested different tools like SpeedTree in the beginning, but returned to modeling the trees by hand for different reasons. We wanted to be able to control every little detail of the models as they are sometimes very stylized and those tools would have required pretty much time to get into and aren’t free, so we decided to stick with what we know works here.
To make that work a little easier, our 3D artist Matthias created lots of modular branch- or leaf-sets that he can use to build a wide variety of trees very fast.
In Unreal we then change the materials of, for example, some bushes and achieve very different looking plants easily. Sometimes just a small change of color makes a big difference and lets the world seem more varied.
As we know from all good filmmakers, color is one of the key elements for the mood and should be used for story progress. To guide the player and make sure he sees everything he should see clearly (and other things maybe not so clearly) we use light and color to influence where the player looks at first.
In Lost Ember you run through a varying world and we wanted to intensify that not only through a lot of different assets, but also through a lot of colors.
One of our goals is to get a shot that you could use as a wallpaper basically every few seconds. And so far it seems to work pretty well.
At the moment we’re aiming at some time at the end of 2017, maybe early 2018. But if we’re able to raise enough money to hire new people, we might be faster.
The one thing that helps us the most at this point is probably just spreading the word and get everyone who might be interested in a game like Lost Ember to know about it! We’re a small team and don’t have big marketing tools or anything, so we really appreciate it how our small community is supporting us already. To make that a little more rewarding for our fans we just started a little challenge a few days ago: we want to get to 5000 newsletter subscribers at our homepage and only then start our Kickstarter campaign with a Lost Ember tour through Europe during which we want to visit youtubers and press all over Europe and maybe even meet some fans and play our demo with them. The person who can recruit the most subscribers who will then actually back our Kickstarter will get a 300$ Lost Ember set when the game is released. Every other subscriber will go into the draw to win the same set as well.
And yes, as part of some of the Kickstarter rewards you can get beta access a few months before the game is released and help us find the last bugs and improve the overall quality of the game.