Now that we’ve had a chance to reflect on Meta Connect 2025, we’re looking closer at how the announcements will specifically impact Industrial VR Training. While the event covers broad applications, our focus is on what these developments mean for Virtual Reality used outside of gaming, particularly in corporate training and education.
The software announcements, in particular, grabbed our attention. Two, in particular, are likely to be very significant for the future of industrial VR training. There is also the possibility down the road that the new hardware Meta is promoting in the form of AR glasses will have implications for the types of training and education applications that we produce. While our current focus is on pioneering VR experiences, we’re equally prepared for an AR-driven future. The processes for developing augmented reality and virtual reality are very similar after all, and we have a wide experience in both, ensuring our clients are always at the forefront of immersive technology, no matter which direction the market takes.
But even though there is always naturally a big focus on hardware at Meta Connect, it was the software developments outlined by Meta that point to more immediate implications for us.
Firstly, there was the new game engine. Intended to be an alternative to Unity or Unreal, it is called the Horizon engine and comes with a number of advancements. It promises 4 times faster loading times, which will hopefully address concerns by some people that the Horizon app has been slow and laggy. The engine also supports large numbers of users – 100+ – in that same instance. The engine is not available to app developers at present, and is for use on the Horizon app only. But all research and development into improving VR is welcome as far as we are concerned.
The other big thing, which is also still in the realms of the hypothetical as far as developers outside of Meta getting their hands on it, is Meta’s foray into Gaussian Splatting. They didn’t invent the technology, but their results with it are looking very impressive.
Gaussian Splatting
Gaussian splatting is a state-of-the-art technique in computer vision for creating high-fidelity 3D scenes from a set of 2D images. It shares a common goal with NeRF (Neural Radiance Fields). That is turning photographs into 3D models. But differs radically in its implementation and performance.
The key innovation lies in its representation of the scene. Instead of using a neural network to query a continuous volume, Gaussian Splatting models the scene with a vast cloud of discrete elements called 3D Gaussians. Think of each Gaussian as a small, flexible ellipsoid that can be stretched and rotated. Each one possesses unique properties: a position in 3D space, a color (viewing-angle dependent), a size, and an opacity (alpha).
Naturally, this technology would appear to have uses in VR. While it doesn’t appear that you would be able to generate models that you could actually interact with with this technology, it could be useful for easily making beautiful and complex backgrounds for VR environments.
A few of us tried out several of Meta’s environments created using this method and it was a profound experience. The environments are so photorealistic that the sense of immersion is immediate and powerful. You are fully convinced you’re “there.” This level of realism is a game-changer. For training, it could create a practice environment that is indistinguishable from the real world. For sales and marketing, it will be an incredibly powerful tool that lets customers experience a product or place firsthand. Perhaps most importantly, this photorealism makes the technology instantly accessible. People of all ages, especially those who don’t play video games, are immediately convinced and captivated. There’s no learning curve, just the powerful, authentic feeling of presence.
The Future of Industrial VR Training is Built on Better Tools
The big takeaway from Meta Connect? The future of immersive tech will be won by those who can merge high performance with authentic realism. For us at Mersus Technologies, this direction is perfectly in sync with our own beliefs and practices, ensuring we deliver not just cutting-edge XR experiences but actually the most effective, accessible, and trustworthy training solutions on the market.
This synergy means that Meta’s powerful emphasis on performance and realism isn’t just an industry trend for us; it’s a tailwind that propels our entire business forward, directly fuelling our purpose to elevate skills, boost productivity, and democratise access through next-gen technology.