Category Gamedev

The Future of Video Games

A very interesting article that I have read over at mashable.com that talked about how close to reality the new games are getting to.

Some types of video games could be considered as interactive storybooks, and the ones that are not yet there, are slowly integrating some narrative features. Instead of words, they give you a visual representation of what the universe and everything inside it looks like directly from the creator’s mind. Some, even might argue that video games hinder our imagination but that is a very conflictive though, that as a widespread range of answers that would make everyone being right and wrong at the same time..

In the last decade, the visual aspects of video games have improved dramatically. For the simple reason: Technology has managed to catch up with us and suddenly, the images have become impressive.

 

The Uncanney valley.

First we have to answer to a question,  what is the uncanney valley?  It’s a hypothetical relationship of an object in a 3D/2D computer generated asset and how close it might look to a real-life object. The difference between the simulated reality and our perception of reality is the uncanny valley.  As an example, take the many human robots that are out there. They look human, but once they move or talk, the alarm goes off in our head.

We feel uncomfortable because we want to perceive it as something real, but the movements, the interactions, and in some cases, the features, do not feel very «human». Even something like a prosthetic hand falls into the strange valley because the way it looks or moves does not form well with the rest of your body.

 

Somehow, when it comes to video games, it doesn’t feel that way. Even when we know the characters we play aren’t real, we can relate. For example, in the first game of The Last of Us, when the main character lost his daughter, it was a moment of tears, because it felt so real. You feel the sadness, the pain and the anger of a human being who was created on a computer.

So, if that’s the case, why not go beyond the mysterious valley and go directly to look like a human? It’s not that simple. Humans have complex facial features. Our skin stretches, pores expand, muscles move when we talk, smile, cry and laugh.
Simulating all of this in a video game would require an immense amount of processing power and would make most games unplayable even on the most powerful gaming computers out there.

Video game development software ability to create photorealistic worlds.

When Sony held the PlayStation 5 launch event, it showed a technical demonstration on the state-of-the-art console, which was performed entirely on the upcoming Unreal Engine 5. It was a fantastic showcase of the future possibilities of  games development tools and hence of the possible paths for game developers in the chase of modelling reality.

Photorealism is what a lot of the global audience of videogames isl looking for, because, as it happened with the evolution of art, while we have not dominated the realism, we will not be able to break its rules as to find new paths for creative exploration. Certainly, there will be a lot of developers and games that will go down the path of realism and photorealism, but in personal case I prefer to think that it will not be mainstream, because if we look at what happened to art, we don´t see hyperrealist art around us, although we are able to make it.

 

Lighting is a key elemento of realism

Light is everything. Not only do we need it to live, but it also plays a monumental role in creating the mood and tone of a game. It is not something that can just be turned on, and it illuminates the game world according to real life.
You have to tell the game engine how the light will act on surfaces and objects. This was one of the significant themes that can break the immersion. Ray-Tracing is an example of how to achieve real-world lighting.
Even Minecraft can look like a totally different game with Ray-Tracing.

 

 

Raytracing

Instead of the developers telling the game world which areas should be brighter than others, Ray-Tracing, to put it simply, creates light as part of an object in the world. This means that light is pixels that bounce off and are refracted onto surfaces by reading whether the object is rough or bright.

Take, for example, your bedroom. If you turn the lights off, the only source of light would be from your window. Now look at the corners of your room, do you realize that it is darker than the other parts of your room?
That’s because light is made up of rays, and there are a finite number of times light can bounce around your room before being absorbed by objects. You would get some illumination, but it would be difficult for this light to reach the corners.

The image on the bottom is without Ray-Tracing. The top  image is with Ray-Tracing.

 

Now imagine translating this into a video game. You walk into a rundown house, and the attic is slightly exposed. Because the light comes in through the window, anything above it or the small cracks will not be illuminated.

The developers have the opportunity to play with these elements, perhaps including a monster there and make your eyes glow red.

Ray-Tracing has the potential to change the way we play video games. Suddenly, shadows or darker areas are a threat. Ray-Tracing may not show its full potential on current NVIDIA RTX cards, but seeing it in action, makes my imagination soar.
Smarter artificial intelligence makes it more than just a video game.

The future of reality is Artificial Intelligence too

We all know the A.I. gets better every day. Take OpenAI Five, for example, an A.I. who was taught to play Dota 2, and even managed to beat some of the best players in the world. He was able to formulate a strategy based on how human players played and beat them. He was even asked to play 42,729 games with random players online, and only lost 4,075 games. This should only be taken for the opportunity to represent, because as all of us that have designed behaviours in games, we now, that players don´t want to play against the perfect enemy, but only an enemy that seems to be perfect, but that has weaknesses that they are able to find as they face it off. If the level of difficulty is to high, then, we have the opposite problem where 80& of players never play more than 20% of the game due to its complexity.

But, it could in some manner be taken to the advantage of adapting to the level of skills of the player, and more casual players will have a less intensive complexity, while very good players will have to have to face a much more challenging and intelligente enemy.

All of this is very interesting from a point of view of the gamer, but we have also to think in the drawback of all these elements, from a gamed development point of view, and that is related with the amount of art and programming that is required to create all these realistic worlds full of real living creatures. That is a massive amount of elements to be created that will required very large teams working over long period of time. Plus, the amount of work that will have to be done to test it and make sure that everything is at the acceptable level of quality will be also massive. Which leds us to the affirmation that was done a couple of weeks ago that games production is becoming unsustainable due to the amounts of resources and funding needed to develop it. ( if we keep doing the things in the same way)

The only possibility is to keep this realistic path, is to take the advantage of AI not only as an element of gameplay, but as a production assistant, that will allow us to create infinite worlds, or infinite variations of realitic creatures based on a set of parameters specified to the system. With this feature, I think that we will be able to progress through this realistic road, if not it will certainly become very complex and hard.