The demand for a decent AI

During my ongoing literature review I often discover interesting facts about things I’ve never thought about. Sometimes I can connect these facts with my own observations: The result is mostly a completely new idea why things are as they are. Maybe these ideas are new to you, too. Therefore I’ll share my new science based knowledge with you!

This week: This time, I think about the huge impact of the AI on the user experience and discuss some of the major issues an AI can have.

In single-player computer games, enemies and competitors are controlled by an artificial intelligence (AI). Depending on the type of the game, the AI executes a certain strategy and tries to flank or outperform the player in order to ultimately put the player’s skills to a test. For this purpose, the strength of the AI is determined by the chosen difficulty level as beginners are more likely to make mistakes than expert players. In the end, the AI greatly affects the user experience as a weak AI does not challenge the players resulting in boredome whereas an overpowered AI is often regarded as unfair resulting in anxiety.

This makes the AI to one of the most complex elements of a computer game as it has to recognize the player’s actions and react to them accordingly. In addition, the AI needs a realistic situational awareness in order to provide the players with an authentic challenge and feedback. For instance, the AI should recognize if the player takes out an enemy who was talking to a different enemy. Just letting the untouched enemy walk away feels unrealistic. The same thing applies to a situation when the AI is able to recognize a hidden player over a long distance. It feels unnatural if enemies suddenly start to attack the player just because a certain threshold was crossed.

Assetto Corsa

Lastly, it is very important that the AI recognizes and respects a player. This is especially important when the AI’s purpose is to create a field of competitors instead of enemies. In a competitive environment, the player expects a gentleman-like behavior that demonstrates a certain strengths wihout being disrespectful to the player. The best example for a poor implementation of a competitive AI is the AI used in Assetto Corsa. The virtual race drivers have a poor recognition of the player’s position on the race track and, as a result of this, they constantly run into the player’s car or even push the player off the racetrack. In a competitve envrionment, this behavior seems to be disrespectful thus resulting in a negative user experience. A good competitve AI would have a well adjusted performance as well as the function to recognize and respect a player’s presence.

In sum, the AI creates challenges for the players by recognizing a player’s actions and reacting to them. In order to have the desired effect, the AI has to have a human-like behavior that is not too overpowered and not too weak. In addition, the AI has to respect the player in order to reduce frustration. In the end, the AI has a huge impact on the user experience and can influence the success of a game.

A small virtual racing community

During my ongoing literature review I often discover interesting facts about things I’ve never thought about. Sometimes I can connect these facts with my own observations: The result is mostly a completely new idea why things are as they are. Maybe these ideas are new to you, too. Therefore I’ll share my new science based knowledge with you!

This week: This time, I think about internet based communities and present some insights on the amount of data that is being generated by a small community.

Internet based communities are not a new phenomenon. Since the early days of the internet, users started to form communities and dicussion groups in order to share information with others. Naturally, communities also evolved around computer games as players like to share their experiences and results with others, ask for help or discuss certain elements of a game.

Furthermore, some computer games allow players to form communities directly in the virtual worlds as players can join others via the internet in order to play the game cooperatively or competitively. In addition, MMORPGs like World of Warcraft allow players to join guilds and thus become a permanent part of a community that mainly exists within the boundaries of the game itself.

Finally, the emergence of YouTube created another special form of community as users can upload videos and share them with the whole world. Interested viewers can discuss the videos, become a subscriber and ultimately form a small community around the content creator.

DiRT Rally

Recently, I became a part of a small community–The Conelanders League–consisting of YouTubers and their fans who enjoy playing the racing simulation DiRT Rally. DiRT Rally provides users with the option to create own leagues and to invite friends in order to competitively race against others. The Conelanders League was initiated by ConeDodger240 in order to allow fans of KurtJMac and himself to come together and experience an exciting community event.

However, rallying takes place in form of a staged race on public or private roads. The participants consecutively race on those stages and try to complete them as fast as possible. In this case, players do not need to come together at a certain date in order to participate in a DiRT Rally league. Instead, the game just provides them with the stages they have to complete and subsequently uploads the results to a database so that the players can compare their runs.

This approach takes away the chance to see the runs of others, however. Hence, participants started to record their runs and upload them to YouTube. In addition, a small database was created to keep track of the overall standings. Personally, I am mostly impressed by the size of the community and the amount of data that has been generated so far. Moreover, the Conelanders League has also united different YouTubers and players who began to support each other.

By now, at the end of the fourth event (one event had the length of a week and consisted of 3 to 4 stages) of the Conelanders League, YouTube found „about 1800“ videos that are related to „Conelanders“ of which „about 1000“ have a length of more than 20 minutes. In addition, the standings database lists 82 participants coming from 20 different nations.

In the end, I really enjoy being part of this small community that connects the world. Moreover, it is fantastic to watch the runs of others as it resembles following a real world sports event.

Guilty until proven innocent

During my ongoing literature review I often discover interesting facts about things I’ve never thought about. Sometimes I can connect these facts with my own observations: The result is mostly a completely new idea why things are as they are. Maybe these ideas are new to you, too. Therefore I’ll share my new science based knowledge with you!

This week: This time, I think about YouTube’s Content ID system that should protect owners of copyright-protected material, but causes several issues on the side of the content creators.

In 2007, the video sharing platform YouTube implemented the Content Identification (Contend ID) algorithm to protect owners of copyright-protected material. Copyright is a legal right that grants the creator of an original work exclusive rights for its use and distribution. For this purpose, the Content ID system provides owners of copyright-protected material with a tool to upload a sample of their work. Subsequently, the algorithm automatically scans uploaded videos for matching content and issues Content ID claims, if matching content has been found. If a video has a Content ID claim on it, then it is up to the owner to decide whether or not the user can reuse the material. In most of the cases, the owner allows the reuse of the material in exchange for putting ads on the video. However, the owner can also mute or even block the video in order to prohibit the reuse of it. In the end, the Content ID system ensures that owners of copyright-protected material get paid for their work.

Unfortunately, the Content ID system is flawed as it frequently misidentifies videos, issues false Content ID claims and works under a “guilty until proven innocent” system. This is especially problematic, as a Content ID claim is not just a simple warning. Instead, it automatically diverts ad revenue from the content creator to whoever has uploaded the sample to the Content ID database. In case of a false Content ID claim, the user can dispute it which starts a lengthy process as the copyright owner has 30 days to respond. During this time, the user can not monetize the video and thus gets no revenue from the ads running on it.

The flawed Content ID algorithm is, however, not the only problem. Once a company has access to the Content ID system, it can upload samples of material they do not own. This led to the problem, that some companies started to exploit the sytem by uploading false samples in order to claim content they do not own.

In the end, the automated Content ID system is needed to protect owners of copyright-protected material as hundreds of hours of video are uploaded every minute to YouTube. However, the “guilty until proven innocent” system and the method to automatically divert ad revenue to a third party based on a single sample in a database is not a good way to treat the content creators who made YouTube to what it is today. Unfortunately, as long as YouTube has no strong competitor, content creators can not do much to fight this system.

Attention to detail

During my ongoing literature review I often discover interesting facts about things I’ve never thought about. Sometimes I can connect these facts with my own observations: The result is mostly a completely new idea why things are as they are. Maybe these ideas are new to you, too. Therefore I’ll share my new science based knowledge with you!

This week: This time, I think about the way how RimWorld simulates a vivid world without overwhelming the player with too many options.

Over the course of the last weeks, I played a lot the computer game RimWorld. RimWorld describes itself as a colony simulator that tells the story of a few people who survived the explosion of their spaceship and stranded on a distant planet. The goal of the game is to build some infrastructure, gather resources, craft gear, defend the colony and research technologies in order to finally construct a new spaceship to leave the planet.

A RimWorld colony.

The player controls the game from an aerial perspective and can give indirect orders to manage the colony. For instance, the player can build a new house by placing a blueprint of the walls and the door. Subsequently, the colonists will gather the required resources and start constructing the building. However, the player can also give some direct orders to the colonists in order to press ahead with a certain task. Furthermore, being able to give direct orders is important in order to defeat some aggressors that occasionally attack the player’s colony. During a battle, the player can draft colonists and control them the same way as in a real-time strategy game in order to defend the colony.

However, the game’s most interesting element is the attention to detail that creates a realistic world without overwhelming the player with too many options. For instance, the temperatures change throughout a year and can drop below freezing during the winter. The temperatures have an effect on the growth rate of crops a player can grow in order to produce food for the colonists. The plants will decay once the temperature drops below freezing, thus requiring the player to fill their storage rooms in advance or build greenhouses to grow crops during the winter time.

Even more impressive is the simulation of a colonist who is represented by four main categories: gear, character, needs and health. The gear section allows the player to equip the colonists with weapons and apparel that protects them against the elements and during a fight. A colonist’s character provides the player with a short backstory about the colonist and an overview over the colonist’s capabilities expressed in twelve different skills. Furthermore, a colonist has needs, such as food, rest and joy, that affect the general mood. When the mood is too low, the colonist can suffer under a mental breakdown. Finally, a colonist is represented by its health which is expressed in twelve different values such as consciousness, sight, manipulation and metabolism. In addition, a colonist can have injuries in different body parts that will negatively affect some of the general health values and sometimes even the performance in executing a particular task.

A colonist in RimWorld.

In the end, this attention to detail results in a very realistic and complex simulation of the life in a colony. Despite the complexity of the world, the general gameplay is not too complex and does not overwhelm the players as their main goal is to manage the colony by expanding the infrastructure or improving the defenses, and, as a result of this, the game can successfully be played without getting lost in the micromanagement of each colonist. However, vested players can also spend a lot of time managing every element of the game.

In conclusion, RimWorld successfully demonstrates how a complex world can be simulated without overwhelming the player. The key is to implement fitting variables that contribute to the concept of the game without making them too powerful. By doing so, the game receives a higher substantiality without requiring the player to take care of every single option. Nevertheless, vested players can start to care about every single value and achieve a better performance in their gameplay.

2015

During my ongoing literature review I often discover interesting facts about things I’ve never thought about. Sometimes I can connect these facts with my own observations: The result is mostly a completely new idea why things are as they are. Maybe these ideas are new to you, too. Therefore I’ll share my new science based knowledge with you!

This week: This time, I think about a few of the highlights of the year 2015.

The year 2015 is over and so it is time to take a look back at a few highlights and events of the last year. In general, 2015 was unusually warm and another indication that global warming is real. Extreme weather phenomena, shrinking glaciers and a rising sea level are results of the climate change and a first insight into our near future. In December, the UNFCCC adopted the Paris Agreement–an agreement to address the climate change by reducing emissions–and showed the whole world that we have to act now [1]. Hopefully, the 55 countries that account for at least 55% of global emissions will sign the agreement during the year 2016 so that it enters into force.

My personal highlight of 2015 was my second journey to Iceland [2]. It was absolutely amazing to explore the unique nature of Iceland and to experience geothermal activities, huge waterfalls, as well as the impressive highlands. Standing in the middle of this spectacular environment and being exposed to the elements rewarded me with an overwhelming feeling of freedom. It always amazes me how certain landscapes and environments can provide me with a feeling that is similar to the feeling of being at home.

The year 2015 had also some highlights in terms of computer games. For instance, I finally got my hands on a force feedback racing wheel which greatly increases the immersion of racing games [3]. Driving a virtual race car using this steering wheel feels almost as if I would be sitting in a real car. Furthermore, I really enjoyed being able to selectively control the arms of B.U.D. in Grow Home [4] and to learn new facts about the Iñupiat people by playing Never Alone [5]. In general, I was able to observe and discuss many motivational and educational effects of computer games throughout the year 2015.

Lastly, I made some good progress towards my Ph.D. as I did an important survey to examine the educational effects of game mechanics that encode a particular knowledge in a serious game. In addition, I was working on the improvement of the serious game to enhance its educational effects during the second half of the year. The improved version of it could have the potential to generate a lot of more data that can provide insights into the method of encoding knowledge in a computer game and show the positive effects of using computer games as educational enviroments.

Despite those amazing highlights, the year 2015 also had a lot of downsides and bad events. In the end, 2015 was ‚ok‘ on my side, but it was not really a great year. Luckily, the new year has begun and so there is a new chance to experience a great year.

Happy New Year!

I wish all of you a great 2016!

The transformation of a content creator to an institution

During my ongoing literature review I often discover interesting facts about things I’ve never thought about. Sometimes I can connect these facts with my own observations: The result is mostly a completely new idea why things are as they are. Maybe these ideas are new to you, too. Therefore I’ll share my new science based knowledge with you!

This week: This time, I think about the phenomenon that a creator of Let’s Play videos is regarded as a representative of a computer game.

Throughout the last years of watching and creating Let’s Play (LP) videos, I observed certain types of viewers showing a particular behavior across all YouTube channels I am subscribed to. The first group consists of viewers who play the game on their own and compare their gameplay with the content creator’s gameplay without taking into account that the content creator has a completely different approach. The second group consists of viewers who just started playing the game and regard the content creator as someone who can assist them with their initial struggles of successfully playing the game.

Creating LPs is a special form of playing a computer game as the player records and commentates the gameplay. In general, the main goal of content creators is to share their gameplay with others and to entertain their audience with an exciting commentary. The viewers, however, occasionally start to regard the content creator as a central fixed point for the way how a computer game is meant to be played. This especially becomes obvious when viewers start to compare their own performance with the YouTuber’s performance. Although it might be entertaining to compare the own accomplishments with a YouTuber, it neglects the fact that the content creator has a completely different approach to play the game than most of the viewers.

On average, a content creator releases one or two videos about a certain game per week. Each of those videos has an average length of about 30 minutes and hence the content creator accumulates a total gameplay time of one hour per week. The viewer, on the other hand, mostly has a completely different approach and plays the game for a much longer time per week, thus resulting in a much higher experience in playing the game.

The second group has a completely different view on the content creator and assumes that the content creator has internalized the knowledge encoded in the game throughout the process of creating LPs over a period of many weeks. In the end, they even start to ask the content creator questions how to successfully play the game. Although the content creator might know how to solve the viewer’s problem, the YouTuber is often achieving things in the game by trying different things, thus learning to play it by playing it. In addition, due to the fact that creating LPs is a very fragmented way of playing a game, content creators often have to relearn the controls of the game every time they start to record a new episode as they have played the game only for a couple of hours during a month.

In the end, both groups see the content creator as a reference point for their own performance and their own goals. The content creator ceases to be just another gamer or an entertainer. Instead, the content creator becomes to an institution for the game. This transformation provides the viewers with a positive feedback as they can compare their own performance with a central fixed point. Finally, they can get inspired by the LPs to train their skills and they can get positive feedback by comparing their outcomes with the content creator’s achievements.

The skewed perception of YouTube

During my ongoing literature review I often discover interesting facts about things I’ve never thought about. Sometimes I can connect these facts with my own observations: The result is mostly a completely new idea why things are as they are. Maybe these ideas are new to you, too. Therefore I’ll share my new science based knowledge with you!

This week: This time, I think about the issues that are created by the assumption that the YouTube subscriber count is a measurement for quality.

Amongst other things, YouTube is famous for the huge variety of different content creators who publish videos related to computer games. The majority of those YouTubers create so-called Let’s Play videos that feature their gameplay which is enhanced with their commentary.

One of the most common ways to measure the impact factor of the content creators is to take a look at their subscriber count. Large YouTubers have more than 1 million, medium sized YouTube channels more than 100.000 and small content creators less than 100.000 subscribers. Although those large YouTubers reach a high amount of viewers with their videos, they are just a small amount of the total content creators on YouTube. The bulk of the videos that are created and uploaded to YouTube comes from the smaller channels.

Unfortunately, this method of categorizing YouTube channels led to the conclusion that the subscriber count is a measurement for quality and that only large and medium YouTube channels feature entertaining and good content. This, however, neglects three main factors that played a crucial role in reaching those high numbers. Most of those bigger YouTubers have started during the early days of Let’s Plays. At this point, the market was not saturated and it was much easier to gain the attention of viewers who subsequently formed the initial basis for a medium or large YouTube channel. Today, the market is saturated as many other content creators have begun to release their own videos and thus it is really hard to gain the attention of new viewers.

In addition, YouTube’s search function increases this effect as it ranks the search results by the amount of views and likes. In other words, YouTubers who already had a good initial basis of frequent viewers had a huge advantage over YouTubers who started during the last years as their videos are always shown on top of the search results.

The third factor is based on the skewed perception that only those larger YouTubers create good content and, as a result of this, game developers and other relevant players started decorate themselves with the content of large YouTubers. For instance, it is much more likely that a medium sized YouTuber gets a retweet on Twitter than a small content creator with only a few subscribers. This behavior creates an additional brickwall for smaller channels as getting announced by other relevant players is currently one of the best ways to gain new viewers.

This behavior can be even seen as a huge marketing mistake as smaller content creators can create content that is superior to the content of larger YouTube channels. In the end, game developers miss an ideal opportunity to advertize their products as they do not recognize the value of smaller channels. Furthermore, announcing the content of smaller channels would also help those smaller channels to grow and subsequently to reach even more users. This, in return, would help the game developers as more people would then be aware of their products.

In general, just taking a look at the subscriber count is a huge mistake these days as the whole YouTube market has changed within the last few years and huge subscriber counts are no longer (and never were) a measurement for quality. In addition, larger channels only remain large as the majority of the viewers does not recognize that there are more interesting alternatives available due to the flawed system. YouTube, game developers and other relevant players need to become aware of the opportunity that leveraging smaller channels can result in a higher return of investment than continuing with the current system of favoring larger channels only.

Epic scale games and a limited amount of time

During my ongoing literature review I often discover interesting facts about things I’ve never thought about. Sometimes I can connect these facts with my own observations: The result is mostly a completely new idea why things are as they are. Maybe these ideas are new to you, too. Therefore I’ll share my new science based knowledge with you!

This week: This time, I think about the different lengths of computer games and one of the most problematic challenges of playing epic scale games.

One of my biggest complains about story based computer games is that they are often beaten after about twelve hours of gameplay. According to my Steam library, I have completed Max Payne 3 in 13, Mirror’s Edge in 5, BioShock Infinite in 14 and Tomb Raider: Underworld in 13 hours. Those times, however, are not completely accurate as Steam starts to measure the time as soon as I load a game and does not recognize when I pause the game in order to do something else. Although those games were enjoyable and entertaining, it is disappointing when they are over as soon as I got used to the virtual world. In addition, it does not feel right when I am able to save the world within such a short amount of time as it does not provide me with the feeling of being part in something that is of an epic scale.

Some games, however, are different and provide me with enough content to play the game for more than 30 hours. For instance, I completed Deus Ex: Human Revolution after 36 and each game of the Mass Effect triology after more than 32 hours of gameplay. Progressing through those games provided me with a completely different experience as I started to develop a feeling for those virtual worlds. In the end, I made so many different experiences during the gameplay that it really felt as if I have achieved an important victory and that I was part of something epic. Moreover, in contrast to the very short games, those games last for about a month in real time and thus they start to become a part of my daily life.

Finally, there are also a few games that last for more than 100 hours and evolve around an epic story. Those epic games mostly are open world games that provide a main story line but allow the player to diverge from it in order to follow the story of a subplot. For instance, I finished the main story of Skyrim after more than 120 hours of gameplay. I think I could have been much faster, if I would have not explored the entire world to search for amazing places. Being able to explore and rescue such a huge virtual world rewards the players with many positive experiences and the feeling that they have made the difference and saved the life of many people.

Those epic games benefit from the fact that a player is able to travel between distant places in real time which allows them to develop a different feeling for the whole virtual world. During their journey, players can even meet inhabitants of the virtual worlds who approach them and ask for help. This provides the players with the feeling that the whole world is living and recognizing them, thus contributing to the immersion of the virtual world.

The only downside of those epic games is due to the fact that new players must develop a basic understanding of the virtual world during the initial hours of the gameplay in order to be able to understand the story of the game. This, however, can be difficult if the player takes longish breaks between the playing sessions and is not able to complete the initial hours of the gameplay within a few days. Personally, I made this experience twice as I tried to play Fallout: Las Vegas and The Witcher. After having played both games for two to three hours, I had to take a longish break and I was never able to find my way back into those virtual worlds. The next time I will try to play such an epic game, I have to make sure that I can play it for several consecutive hours in a row to develop a basic understanding of it.

The joy of improving

During my ongoing literature review I often discover interesting facts about things I’ve never thought about. Sometimes I can connect these facts with my own observations: The result is mostly a completely new idea why things are as they are. Maybe these ideas are new to you, too. Therefore I’ll share my new science based knowledge with you!

This week: This time, I think about the emotional rewards of playing DiRT Rally.

Currently, DiRT Rally is one of my favorite racing games as it allows me to experience the excitement off-road racing and to put my driving skills to a test in the disciplines of rally, hillclimb, as well as rallycross. Additionally, the game realistically simulates the behavior of the rally cars on different surface types. For instance, it is much easier to keep a car under control when driving at high speeds on dry asphalt than on a gravel road. Racing along those rally stages provides me with new challenges as rallying requires a completely different style of driving than racing at a regular race circuit.

Racing in DiRT Rally.

In general, playing a realistic racing game mostly results in a strong flow experience as the racing itself provides drivers with clear goals and a constant feedback about the own progress. The most obvious goal is to beat the own personal best lap time on a race track and/or to drive a perfect lap. This goal is even recursive as each time a driver achieves a new fastest lap, the goal is renewed and the player can start again to beat the own lap time. In addition, the player is constantly informed about the own performance as the game displays the current lap time and thus the player is able to observe the own progress.

As a result, players can get into the state of flow which constantly motivates them to continue and to exhaust their skills in order to overcome the next challenge. The lap time in itself becomes unimportant to the players as they do not crave for the achievement. Instead, they derive their motivation from the emotional rewards of completing a challenge after having exhausted the own skills.

This is also true for DiRT Rally, as the main goal for the players is to complete a rally stage as fast as possible. In addition, the driving experience of a rally car in itself provides the players with a new challenge and allows the application of special techniques that are used in rallying. For instance, the players can learn and master rally maneuvers, such as a handbrake turn or a Scandinavian flick. The behavior of the virtual rally car then constantly provides the players with feedback about their learning progress.

This is also the main reason why DiRT Rally is currently one of my favorite games. It provides me with a great amount of new challenges, constant feedback and lots of emotional rewards.

Force Feedback

During my ongoing literature review I often discover interesting facts about things I’ve never thought about. Sometimes I can connect these facts with my own observations: The result is mostly a completely new idea why things are as they are. Maybe these ideas are new to you, too. Therefore I’ll share my new science based knowledge with you!

This week: This time, I think about my first time experience of using a force feedback device.

Realistic racing simulations belong to my favorite computer games as they have a very strong immersive effect and are an ideal environment to observe my personal improvement over time. On the one hand, the strong immersion is a result of the visual representation of the virtual racing as racing simulations allow me to enjoy the gameplay from the driver’s perspective, thus giving me the feeling of actually driving the car. On the other hand, the strong immersion is also a result of the realistic physics-engine that simulates the behavior of the car correctly and provides me with feedback about the car’s performance and the surface condition.

Furthermore, the realistic physics-engine helps me to observe my personal improvement as it provides me with an immediate feedback about my driving performance. Over time, I develop a feeling for a race car which enables me to drive it at its limits without loosing control over it. In the end, the feeling provided by the physics-engine allows me to recognize when I start to master a particular race car. In addition, the racing itself provides me with lots of feedback about my improvement as I get constantly informed about my lap time and thus I can visualize my progress.

DiRT Rally

Recently, developing a feeling for a race car has reached a completely new dimension for me. Until the end of last week, I was playing racing games with a very basic, no force feedback racing wheel that featured only 280° of rotation. Now, I have upgraded to a 900° of rotation force feedback racing wheel that really helps me to develop a feeling for a race car. Although the positive effects of force feedback are already well known, I was blown away as I played a racing game for the very first time using the new racing wheel as this was also my first time experience of the current generation of force feedback. Using it resulted in a new dimension of immersion as the force feedback provides me with an additional layer of feedback about the status of my car. Additionally, based on this feedback, it is much easier to develop a feeling for a particular race car and to control it at its very limits.

Finally, I was really amazed as I played for the very first time the off-road racing simulation DiRT Rally with my new racing wheel as it provided me with an accurate feedback about the surface condition. In particular, my first race took place on a gravel road and the emulated vibrations of my racing wheel resulted the same feeling I experienced while I was driving along gravel roads in Iceland.

In the end, driving a virtual race car now feels almost as if I would be driving a real car. Well, except for the fact that I unfortunately do not feel the acceleration on my body. Still, I can not wait for the next ride along a virtual race track!