![[Img #74683]](https://thelatestnews.world/wp-content/uploads/2024/12/The-main-mistakes-to-avoid-when-betting-on-electronic-sports-150x150.jpg)
First of all, it must be taken into account that G-Assist is a project that has been presented as a technical demonstration. That does not mean that it does not offer us a vision of what the integration of AI assistants in games will be like, a function that will arrive sooner or later. “Project G-Assist aims to put knowledge about the game and the system within players’ reach,” explains Nvidia. “Just ask a question about the game or system performance and the assistant will provide contextualized answers.”
G-Assist combines the player’s voice or text with what the game shows on the screen. This information is fed into AI vision models that provide context awareness and understanding for the large language model (LLM), which in turn is connected to a source of knowledge about the game such as a wiki. The result of the LLM is a personalized response, text or voice from the AI, based on what happens in the title. Developers can customize vision and language models, but ultimately the quality of this response will depend on information from the Internet.
Nvidia has shown how G-Assist works with ARK: Survival Ascended, the open-world survival game from Studio Wildcard. The AI assistant is activated by pressing a key or a specific phrase, and can help players answer questions they often find themselves searching on the Internet (completing quests, crafting items, how to beat difficult bosses…). For example, at the beginning of the game it can suggest the best weapon and where to find the materials needed to make it. In addition, thanks to the AI vision model, the assistant understands what is happening in the game and can give advice on whether to avoid some type of enemy or how to tame a particular beast.
By being context-aware, G-Assist has the ability to personalize recommendations based on the user’s experience.
Here are some of the examples that Nvidia gives based on gender. In a role-playing game the wizard could point out the location of secret objects and indicate which part of the story can be skipped; in a real-time strategy one it could indicate the recommended construction order or the best resource management strategies; while in an FPS it can recommend the best equipment. “These are just a few of the many possibilities,” the company says. Remember again that G-Assist uses information available on the Internet and the quality of its answers will depend on what it finds.
On the other hand we have G-Assist as an assistant that adjusts the performance of the games. In this sense, the AI assistant seeks to determine the best configuration based on the equipment’s hardware and to demonstrate how Nvidia has used Cyberpunk 2077. With a question G-Assist evaluates the configuration and performance of the system and adjusts it in the application. Nvidia for an optimal experience. As? The settings determined by the AI assistant are based on testing thousands of hardware configurations.
In addition to setting up the basic visual experience, G-Assist can do many other things, including applying a overclock sure, check that you are taking advantage of the monitor’s resolution, its refresh rate or other capabilities such as HDR or G-SYNC, monitor and display FPS, latency, consumption and other statistics or find a way to increase performance per watt without losing the Target FPS. On laptops you can also determine how much gaming time is left before the battery runs out. Additionally, you have the ability to describe what the adjustments do and what their expected impact on performance is.
If the name G-Assist sounds familiar to anyone, it’s because they have a good memory. In 2017, Nvidia presented, on the occasion of April Fool’s Day, a product called G-Assist who played alone.
Add Comment