Behind the scenes #3 – Release

Last time we gave you some insight on how we handle feedback. Today we want to share some more delicate details with you: The time right around the release is without any doubt the most stressful and tense part of making a game but at the same time the most thrilling and rewarding. We would like to tell you what we experienced and what knowledge we gained.

After Train Fever reached the prototype state in 2012, a successful crowdfunding campaign took place between September 2012 and March 2013 which enabled us to plan for finishing the development. After the campaign reached its goal of 250.000 Euros on March 20, 2013, the release was scheduled for May 2014.

gambitiousThe successful crowdfunding campaign on Gambitious funded the game’s development.

As (almost) usual in game development, it became clear that a lot of features will take longer to implement than expected. The planned feature list had grown and – opposite to the original idea – now included industrial goods, bigger maps and far more vehicles. The release date had to be pushed back a couple of months to September 2014. The two publishers paid 20.000 Euros each as an advance to cover the costs for the additional development time.

After the majority of the features found their way into the game, the beta test was finally scheduled for July 2014 with less than two months left until the release. Having only so little time for beta testing was not the only thing showing that we really ran out of time. For instance preparing the 64-bit version was probably on the tightest schedule: Testing only started five days prior to the release, luckily we got everything done in time.

But not everything was done prior to release. Although it was planned to be delayed, the Mac version was released one week after the game came out on Steam. In general a lot of polishing had to be shifted past the release and many things had to be left as they were.

Next to all the last minute tasks we also needed to take care of public relations. Even we did not prioritize it, preparing the homepage, make a trailer for the game, sending out steam keys and giving interviews to press consumed a lot of time.

In the end we were forced to publish an unpolished product. Many people asked why the release was not pushed back further to leave more time for development and bug fixes. The answer is quite simple: We had a limited budget from the crowdfunding campaign. After already rescheduling once to September it would have been just too expensive to extend the development again.

Despite all obstacles we still successfully released the game to the public. However, scheduling the release time turned out to be another challenge. Steam releases are usually within the offices hours of Valves´ headquarters between 9am and 5pm PDT (pacific time), which translates to 6pm and 2am CET (central European time). This is why for Europeans most games are unlocked on Steam sometime in the evening.

This was not an ideal situation: In addition to Steam (and later some other platforms), Train Fever is also sold as a retail version containing a Steam key. With stores opening in the morning, people buying the game on the release day might want to play it right away, so unlocking the game in the evening was not an option. Unlocking the game before stores open right after midnight would have been too early, as the potential buyers from Europe – our main target group – could miss the game on Steam. A possible hype generated by a new release could get lost during the night.

Fortunately we were able to persuade a Valve employee to stay in the office until past midnight. After having some coffees, he unlocked the game at 10am CET.

store_addThe Train Fever trailer running on all the TVs in an electronics store after release day.

Releasing digitally and retail at the same time is difficult enough from an organizational standpoint. What furthermore complicated the situation was that there were two publishers involved with different interests. Naturally everybody wanted to change the conditions according to their best interest. The retail publisher wanted to have a lower price than digital, the digital publisher wanted a discount for launch and both were not happy with each other’s decision.

We did our best to mediate and in the end, even there were notable differences in prices and discounts across the various distribution channels and regions, everything went smoothly. Still it took quite some effort to explain the situation to the customers.

Not only was the night before the release short for that certain Valve employee, also here at Urban Games people did not find much sleep. After putting the last touches on the game everybody left the office at around 6am in the morning, only to be back two and half hour later awaiting the release.

“The coolest thing about the release day is that you can stop by a games shop and buy your own game!“
~Urban, CTO

In the office it was quite a tense situation. This day would decide if the work of several years was a success or not. Finally, when the sales figures where rising and the first reviews came in – Train Fever scored an unexpected and staggering 82% at GameStar – everybody was relieved. At the end of the day Train Fever even made it into the top seller list on steam.

But the work was far from done. As mentioned in the last blog post, we automatically receive crash logs. With so many people playing the game at the same time our server just could not handle the load and gave up – nobody expected that to happen. With so much going on and everybody already being tired, in that moment it just made us laugh. A reboot later we were ready to receive the crash logs again and started working on the first bug fixes.

Everybody would have needed some time off the days after release, but encouraged by the success we kept on going.

With more and more people buying the game the amount of feedback and support request increased. Also the press showed a lot of interest in the game, asking for interviews. Still being energized from the successful launch we shared too much details about our plans and mentioned what we would like to do in our next game. With a yet unfinished product it was misinterpreted by some players. They assumed we would already be working on the next game, which in fact did not apply for another year.

Not sharing what comes next is just one of the things we learned from making our first game. Developing games is time consuming and costly. A classical beginner’s mistake is scheduling too little time for polishing after all features are implemented. It is better to cut down on features and leave more time to flesh them out.

Making everybody happy – the developers, the publishers and most important the players – is very difficult and with many people involved communication and coordination can get complicated. Deciding on the right price and having every customer treated equally, with digital vs. retail, currency exchange rates and opposing interests involved, is next to impossible.

team_orgThe original development team: (ltr) Manu, Basil, Urban, Coach.

It was our first game, so it is to be expected that not everything was done according to the book. But since then over a year has passed and we have put a lot of work into Train Fever fixing bugs, improving the performance and adding new features.

This reflects in very positive feedback from the community and the sales figures of the game: Train Fever sold over 100.000 copies, 10.000 of them from pre-orders, 40.000 only within the first month, and is considered a success.

Overall the joy it brings to release a game outweighs all the stress and exhaustion, but it is still hard work and a lot can be learned from it. Even that some mistakes have to be taken with humor.

We will utilize all the experience we gained in our next game. You are for sure eager to hear about it! We promise to share the first details with you as soon as possible! Until then stay tuned and wait for our next blog post, where we will give you some insight on sales figures and talk about the different distribution channels.

Behind the scenes #2 – Feedback

In the last blog entry we talked about modding and the importance of the community’s contribution for the success of Train Fever. This time we will give you some insight into how we handle feedback and suggestions. We will illustrate in what way it influences the development of the game.

A considerable part of the feedback reaches us directly via email. We keep a close eye on our forums and the Steam discussions page. In addition we participate in the forums run by our dedicated community. Gameplay videos, Steam reviews and tests from game magazines are other valuable sources. Furthermore, reports are generated when the game crashes. Players can opt to let these reports be sent automatically to us.

feedbackWe receive feedback from a lot of diverse sources.

Reading through all the comments is a time-consuming process, but it is crucial to improve the game according to the players’ opinions. We accumulate all the feedback and divide the collected data into various categories, including new features, improvements or bug fixes.

Bug fixes have the highest priority. The automatic sending of crash reports allows for finding and isolating bugs in real time, and we can react to problems very quickly. A comprehensive stock of diverse hardware helps us to test Train Fever and reproduce errors or problems. The systems range from consumer Laptops up to devices equipped with the latest processors and graphic cards. We run various versions of Windows, Mac OS and Linux to cover the most common hard- and software combinations.

After sorting the bugs by severity and estimating the time and effort to resolve them, we work on a solution with very high priority. If it is a critical bug, it will be released immediately as a hotfix, otherwise we decide what we will include in the next game update along with the fixes.

bugsA screenshot of our real-time crash report database; a crucial tool to minimize the number of bugs and crashes.

As opposed to bug fixes, handling feedback in regard to new features is a more difficult process. In addition to our own (huge) feature wish list, players send us lots of feature suggestions. Hardcore players are important, as they deliver most of the feedback, but we are always happy to see everybody being involved. If so, we get a more representative picture, and it is easier to estimate if a certain feature is really requested by the masses.

“The more feedback we get, the better we can tare if it is necessary and a welcome feature. “
~Manu, 2D/3D Artist

New features come with obstacles we have to overcome. Players often share great ideas with us, but they can not specify the exact way it should be implemented. With new mechanics, we always face the possibility of them having an unfavorable impact on the gameplay. Rebalancing the whole game might be needed.

An adaption of the data structure, which determines how games states are saved and loaded, can be necessary. A lot of effort has to be put into assuring old save games load correctly. We also need to consider the big amount of mods available – they still have to work after an update.

One thing we learned is that releasing a patch which includes any kind of restriction compared to the previous version is always a bad idea. When we introduced the so-called main connections – a feature intended to prevent cheating – many players were not happy with this change. Finally, we made another patch which added the option to disable the new main connection feature.

“Players naturally get used to the as-is situation and might prefer it over the new version.”
~Urban, CTO

Certain improvements can represent a big challenge. We have released several performance patches, yet we are still not completely satisfied. Due to the complex simulation Train Fever is very demanding, especially in regard to pathfinding and town growth in the late game.

Assigning resources to bug fixing and improving the game sometimes leaves us less time for new features than we would like to have. A long list of features was waiting since the release, but we had to delay them, like the waypoint system for trains or the vehicle filter.

waypointsWaypoints were one of the most requested features and have been introduced in build 4619.

A little known fact about Train Fever is that some fundamental mechanics are based on external feedback. Initially, when the crowd-funding campaign for Train Fever started in 2012, the game design only contained passenger transport. However, the backers made it very clear that they want to transport cargo, too.

“The more players, the more ideas, the more creativity.”
~Basil, CEO

Since recently, we are working on what will come after Train Fever, and the feedback we get is very useful to prepare for that. Some work can be used for both the current and the upcoming game, which safes a lot of time and avoids splitting resources. In a couple of cases we concluded that it is not possible to adapt the current game without risking to trigger one of the mentioned problems. We direct these ideas to our next project.

Feedback is very important to us. In that context let us say thank you for your comments on our last entry and showing us your interest. We have a lot of topics we still want to cover. The same applies as for new game features: Let us know what you want us to talk about! It helps to determine what you are interested in. So feel free to leave a comment!

Next time we will focus on the release of Train Fever. It was a very delicate and exciting process. We would like to share what we experienced.

Behind the scenes #1: Modding

Last week we announced that we start to write a regular blog series and talk about various topics. We want to begin with the most important part of Train Fever: The community and how they contribute to the game. Train Fever and its development profit in a lot of ways from its players. On the one hand, we get a lot of feedback and suggestions on how to improve the game (the subject which we will discuss with the next blog post). On the other hand, and this is the topic which we focus on today, thanks to modding players can directly modify and add content to the game.

speed_trainKT4D and KT4Dm Berlin streetcars. Image courtesy of community member Stepke.

Realizing that not everybody knows about mods and modding, what it means and what it can do, we would like to briefly talk about its meaning and history:

Despite the definition of modding being diverse, related to video games it is commonly referred to as “modifying (a video game) by changing its behavior or content and also adding to it or even making a completely new game based upon another one”. An important part of modding is that it is not done by the developers of the game but by the players, so called modders.

“Making games moddable gives players a lot of flexibility. They can create their own world without having to give up on the games functionality.”
~Stephan, Senior 2D Artist

The history of modding goes way back to the 1980s and has a rich history with a lot of prominent games, in the early years mostly within the first person shooter genre. They did not only allow for modding but actively supported it by making it easy for players to access core parts of the game.

In the genre of transport simulations well known representatives are games like Locomotion, Railroad Tycoon 3 or Cities in Motion, where countless bits of new content were created by players since the release, and these classics still have active modding communities today.

Non-commercial games created as open source like OpenTTD and Simutrans are special cases of modding, where not only everybody can add content to the game but even further, due to the nature of open source, obtain and change the source code.

There are a lot of benefits to modding: Mods allow the community to create new content for a game which everybody can enjoy, keeping it alive for a long time. Players can change the game to their needs from making it a complete sandbox to ramping up the difficulty, add their own buildings and vehicles as well as change how the game looks. Players like to actively contribute and add to a game not only by playing it, but by creating content.

speed_trainReconstruction of Liège-Guillemins railway station. Image courtesy of community member Meister-Zogi.

Allowing players to add to the game and at the same time offer payed DLCs can contradict each other. Traditional game development prefers to generate more profit by selling bits of content as DLC. On the other hand extending the games life spawn by allowing everybody to add to the game can be more favourable for the developer in the long run, so we decided not to charge for any additional content we create.

With that in mind, when the development of Train Fever started back in 2012, making the game easy for people to modify became one of the primary goals.

We tried to make as many parts of the game as open as possible for the players to modify, but there are still limitations and not everything in the game can be modded. Exposed game functionalities can have critical impacts on the game´s behavior, allowing mods in the worst case to make the game impossible to play or break certain aspects of the game logic.

“Not everything can be modded. For something to be moddable, the relevant functionality must be extracted from the program code and be made accessible via configuration files or script functions.”
~Urban, CTO

There is also a tradeoff between simplicity and powerfulness. Interfaces for mods should be generic enough to allow for extensive modding but easy enough for everybody to use. Therefore a lot of thought went into planning what functions should be made accessible in which way.

Train Fever primarily uses the Lua (Portuguese for “Moon”) scripting language. The decision to go with Lua was based on the fact that Lua is a free, fast and powerful, yet simple tool, and is one of the leading scripting languages used in making games. Lua is easy to learn and understand due to its solid reference material and needs nothing more than a text editor to edit scripts.

As early as in the closed beta we saw the first mods being made for the game, introducing new vehicles and even new industries, like the Cargo Mod from one of our beta testers Gwinda. Since then hundreds of great pieces of new content have been created.

“Modding brings up some great ideas. There are some mods that utilize the functionalities we have provided and use them in ways we have not foreseen.”
~Manu, 2D/3D Artist

Since the release, beside many stunningly designed trains, trucks, buses, trams, stations and depots, we have seen modders creating mods that utilized functions of the game in an unforeseen and creative way, like the broad selection of decorative items using waypoints and the invisible locomotives to help make Train Fever feel even more realistic. On top of that, a lot of new textures, shaders and various script mods change the way Train Fever looks and feels.

The community also helped building the free USA DLC, and a total conversion to bring Train Fever into a nordic setting is currently in the making by some dedicated Train Fever fans and players.

speed_trainScreenshot from the upcoming fan made nordic total conversion mod. Image courtesy of community member mediziner.

Despite getting a mod into the game is as easy as downloading scripts and models, and putting the files into the right place, it can take some practice to get used to. But also here help is right around the corner in the form of Mod Managers, also made by people from the community, making it possible to install and activate mods by the matter of a few simple mouse clicks.

We hope this post aroused your interest in modding Train Fever and if you are already using or even creating mods, let us know in the comments what your favorite one is! Even though we pointed out some examples, we could not decide on a single one.

Next time we will talk about the various and extensive feedback we get, in which way we handle it and how we decide what we implement into the game. If you have any comments or even suggestions what we should write about, we will be happy to read your comments!

Links and resources:

Mods and tutorials can be found on various fan sites:
- http://www.train-fever.net/ (English and German)
- http://transport-games.ru (Russian)

Mod Mangers can help installing and maintaining mods:
- Train Fever Game Manager (TFGM)
- Train Fever Mod Manager (TFMM)

Links to the mods mentioned above:
- Cargo Mod by community member Gwinda
- Tatra KT4D by community member medizinier
- Liège-Guillemins Railway Station by community member Meister-Zogi
- The Nordic DLC community project

Train Fever – behind the scenes

As the development of Train Fever is continuing, there is a lot going on at Urban Games. We would like to share with our community what we are currently working on and also meet your interest in behind the scenes information about Train Fever.

So, in a series of blog posts, we want to take a look back at a thrilling, diverse and informative time, share our thoughts and experiences with you, discuss what we learned and give you some insight into topics you usually don´t hear developers talking about. Of course, we will also say a few words about the next game update.

speed_trainPart of our office in Schaffhausen with artists Manu, Coach and Stephan

The development of Train Fever turned out to be very dynamic with constant need for adapting. We want to highlight some examples on how we integrated difficult features and why sometimes something looked simple but turned out to be a big challenge.

Despite our company is a small team and started only as a two man business in 2013, since then from investors to publishers quite some people were involved in making the game happen. We want to show that releasing a game is a lot of effort and involves a lot of various processes.

Train Fever is the first game we released and it was exciting to go through this important and delicate process. How reviews and Steam sales can influence sale figures was very interesting for us to monitor and will provide another topic to reflect on.

As modding and the community are two of the main cornerstones of Train Fever´s success, we will talk about user created content, what we learned from the community and how we handle suggestions and wishes towards the game and what was done based on the immense amount of feedback from the Train Fever players.
So stay tuned until next week when we share our first blog entry about one of the many topics mentioned above.

If you want to get informed when we post something new on our website subscribe to our newsletter and we will keep you in the loop!