Over the past decade, one of the biggest problems in the gaming industry is the explosive development of AAA video game production costs. In 2000, a dozen people could make a great hit PlayStation. Today, the new iteration of every franchise with a console or PC requires hundreds of artists, designers and programmers working for several years. Even free games, such as hidden objects or puzzles, takes many months from the concept to release to the Apple App Store or Google Play.
One of the biggest costs of game production is to create resources. Players want the content of their games to be not only fun, but unique. Models of character, textures, skins, exotic locations, various improvements in the game, all must be carefully adapted, sophisticated, working by artists, which is extremely expensive.
The next problem is scaling. Game Studios must spend many DLC updates and asset packages to remain competitive. This is particularly important today, when monetization trends all switch to shopping in application, such as character skins, accessories, vehicles, emotions and cards. Talent is rare, and one of the solutions is to employ outsourcing studies in games from low salaries, but finding credible cooperation is not an easy task and building long -term cooperation.
That is why studies are always looking for technologies that can help reduce the cost of developing the game. Recent progress in the models of generating neural images, such as Dall-E, Midjourney and Stablediffusion, bring hope that the realization of this dream may not be so far away.
Enter Get3D from Nvidia, the latest entry in this field, announced in the last post Here
This AI model has been trained using only 2D images and is able to generate 3D shapes with high loyalty textures and complex geometric details. The range of objects that can be created is quite wide: vehicles, characters, animals, people, buildings, various external spaces that can be combined into entire cities with their inhabitants. Export formats are suitable for the most popular graphics software, enabling easy import of shapes for 3D renderers and game engines, and can be smoothly built in existing flows in art production.
How close we are the implementation of this and similar models in real game development? “Technology is very impressive and we will certainly see many generated artistic resources in games. However, they will not replace the creative team of human artists, not in the foreseeable future, because AI still requires a well -thought -out contribution to get good results, and careful post -production work to make it coherent and polished”, according to expert opinion from a professional opinion. Absolutistic Game Art Studio. It seems that in many other areas, machine learning applications seem to be very effective assistants of artists and designers, and not a substitute for human creativity and skills.
