5 things we learned at Unite Berlin 2018

• 1 Unity’s big Google Cloud deal shows a focus on online games

  Many games these days are connected online. As Unity CEO John Riccitiello noted on stage during the Unite Berlin 2018 keynote, ‘connected games’ as the company calls them make up 90 per cent of the top grossing apps on Google Play.

  To power this wave of connected games, Unity has signed an extensive partnership with Google that will see it move the entire Unity infrastructure to Google Cloud.

  It’s a move Unity claims will let developers “benefit from Google Cloud’s openness, security, scalability and innovation through data”.

  The road ahead

  As part of that deal, the two companies are also working on a number of initiatives. This includes a new open source multiplayer matchmaking system project set to go live in summer 2018.

  Further down the line other new initiatives include game server hosting, offering streamlined resources to develop and scale hosted multiplayer titles; a sample FPS of a real-time multiplayer game; and a new ECS Networking Layer.

  Time will tell whether this deal will be good for developers, but by partnering with a giant tech company like Google and letting devs utilise the Google Cloud for their connected games - it’s clear Unity sees online as the current and future of gaming.

  
  

• 2 VR/AR/MR still excites

  For years before the first consumer virtual reality headsets from Oculus, Valve, Sony and Samsung came out, the question was: at this early stage are developers more excited by this technology than consumers?

  While the companies involved, particularly Oculus, tried to play down expectations, the market for VR headsets has certainly failed to really capture the imagination of the public thus far.

  But developers still appear excited by the potential of the tech, as shown by a number of VR demos for services and games on the Unite Berlin expo floor.

  New realities

  While VR struggles in the consumer space, augmented and mixed reality are now the hot and trendy buzzwords. And this time, in large thanks to Pokemon Go (despite its arguably limited AR capabilities at launch), consumers appear interested in its possibilities for entertainment and practicality in the workplace.

  So it was no surprise to see companies showing off avatar creation for the AR space - based on your own appearance, or how AR and VR can be used in the automotive sector - such as getting a virtual look inside a car.

  Unity aims to power many of these experiences with the newly unveiled Project MARS - Mixed and Augmented Reality Studio. The toolset is Unity's attempt to help developers create AR experiences "with little to no custom coding".

  Mixed reality start-up Magic Leap meanwhile, a company that has raised in excess of $2 billion to date, was on hand to discuss tips for designing in MR and tell attendees why the technology’s future is so exciting.

  
  

• 3 Unity’s future lies in technology, not just games

  For a while now Unity has been making its big push into virtual and augmented reality, and it’s also been used in industries such as film.

  And as the company grows, it seems while games are likely to remain its core, it’s now officially expanding into other sectors outside of it.

  Driving into new industries

  Unite Berlin saw the company officially make moves for the automotive sector. The expo floor was full of games companies, but in the centre was a special section dedicated to automotive. Other booths dotted around were also promoting their services in the field.

  There was also a special Unite Berlin AutoTech Summit which held 18 sessions across two days. Again all dedicated to an industry that lies outside of games.

  It will be fascinating as Unity continues to grow and expand how it will bring together - or spread apart - these separate sectors.

  In five years from now will Unite be a place for developers from across numerous industries? Or will it need to tailor its events to make it more relevant for the very different kinds of businesses it’s trying to cater for right now?

  
  

• 4 Nested prefabs finally arrive

  A running joke in the Unity community, nested prefabs have finally arrived in preview, along with a new prefab workflow.

  It was announced at the end of the Unite Berlin keynote via a surprise appearance from former CEO and Unity co-founder David Helgason - a cameo appearance that was even a shock to current CEO John Riccitiello.

  As we noted in our story, the new prefab workflow enables developers to split up scenes and prefabs on a granular level.

  Rethinking the workflow

  “The long-term goal has been not only to implement support for nesting, but to rethink the core prefab workflows so different team members can simultaneously edit prefabs confidently and efficiently,” read a statement from Unity.

  You could tell by the cheers just how much developers want a decent solution for nested prefabs.

  And when asked whether the confetti explosion that ended the show was to mark its end or if it was to celebrate nested prefabs - Unity VP of engineering Lucas Meijer told PocketGamer.biz it was, of course, solely for those prefabs.

  
  

• 5 Unity is making machine learning accessible

  Machine learning is one of the big buzzwords in the games industry right now and its possibilities for development and marketing are numerous.

  One announcement made during Unite Berlin was that of animation blending system Kinematika.

  Developers can record motion capture footage for various stunts, movements and other special manoeuvres they want to use for their animations, then through applying machine learning can implement those scenes into a game.

  Seamless animation

  But what specifically makes Kinematika unique is that it combines tiny fragments from the library into a sequence that matches the controller input, environment content and gameplay requests.

  It’s an accessible way to make smooth animations between states.

  Kinematika isn’t the only use case Unity has for machine learning though. In September 2017 it introduced machine learning agents, which can teach ‘agents’ in the game how to interact with the environment and players, or in some cases imitating how users play the game.

  In theory this technology can then be used to dynamically adjust the game difficulty level and make AI act more human-like.