itch.io | mastodon
New VR Project
I received my Vive Focus developer kit and have been testing a new project on it. While standalone headsets are the priority I intend to release the project for the original Vive as well. The game is meant to be a slow-paced and relaxing experience focused on observation and puzzle solving. I want to keep the scope limited, with a world size larger than my previous projects but with an emphasis on several interactive features and items per room-scale platform. I'm having fun developing for a standalone headset with positional tracking. Being able to explore a virtual space without a tether is quite nice.
I created a new short horror game for Vive. The thought was to refine the low-light look of 'The Folded Castle' while reducing the space needed for navigation. Settling on a 2.5m x 2.5m play area seemed like the right compromise to let players move through the environment naturally while not forcing them to rearrange their living rooms. The wall meshes were positioned so the player can safely switch to Developer Chaperone style with low bounds opacity for better visual presentation.
Early tests of the dark rooms made me decide to add rear-facing glare on the player's lantern to hide the Vive's panel grain. That had the bonus effect of making the lantern easier to find if dropped by the player. The angle of the lantern light was also narrowed and the creature forms were kept very vague. The spot light cookie, dust, and bugs were added to give the scene parallax and make it more believable. The overall aim was to keep information somewhat limited, making the player listen carefully and use her imagination. Lastly, I wanted to create a longer experience than my earlier prototypes so I made a simple puzzle for the player to solve.
This was a good exercise in learning more about the Unity editor and C# scripting, I hope you enjoy it. Download Ruin's Hold on itch.io.
Here's my latest VR scene, this one was inspired by early 3D graphics. The goal was to make a VR experience that would run well on PCs that are far below the minimum specification for the Vive. This also allows newer GPUs to run the scene with increased supersampling for crisp visuals. I think the result looks pretty good and I'm excited to start targeting standalone headsets as they hit the market. Take a moment to explore and interact with the scene: launch
The Folded Castle
Here's a new short VR experience with a horror theme. The concept is to allow the user to explore multiple locations without teleportation controls or jarring transitions. There's also a new batch of meshes included to use in your own projects. Take a look at the .obj files and see if you get some ideas.
The Folded Castle
Polygonal Editing in VR
I just noticed that ArtStage has been updated with a bunch of mesh editing tools. I tried the tools briefly and they are a lot of fun to play with. If you own a Vive and have any interest in low-level mesh editing I definitely recommend giving it a try.
Here's a small VR scene. I wanted to try a different relocation system and play with a faceted look. It should run on PCs below the minimum specification for the Vive so give it a try even if you have an older setup.
I'm making the models I used for this project available for free. There are also a lot of meshes that didn't end up in the final version so take a look and see if there's something you can use. The models are .obj and scaled for direct import into VR scenes. You're welcome to use them for your own prototypes or full games.
Virtual Reality and Art Direction
updates - 2018/4/8, 2018/2/28, 2017/6/7, 2016/10/26
Three major virtual reality headsets have been released and I’ve been thinking about various approaches for art creation in this emerging (re-emerging?) medium. While there has been a push to market 'VR-ready' PCs to consumers I believe an interesting and ultimately more sustainable approach is for developers to creatively limit themselves and succeed through art direction. Here are a few of my thoughts along those lines.
VR development is still in an experimental phase but there is no doubt that high frame rate is critical for comfortable VR experiences. While the cost of high-end PCs needed for rendering realistic scenes at 90+ frames per second is coming down it will remain an adoption barrier for many users. However, low-cost, standalone VR headsets with positional tracking are becoming available. The trade-off is that such systems do not allow for extremely realistic visuals. Despite that, I believe such hardware will attract a larger number of users due to the lower price, ease of setup, and portability compared to desktop VR. Wireless video transmission solutions which allow for untethered play are making desktop-based VR more appealing but the cost and setup hassle remains. A standalone headset, executed properly, will allow a user to simply turn the device on and start experiencing VR.
I propose that developers alter their approach to creating VR applications by making use of stylized artwork to target these standalone headsets. Thoughtful and consistent art direction can provide the bulk of a game's visual appeal, making efforts towards photorealism unnecessary. It may help to look at older games to see how developers creatively overcame hardware limitations. Asset creators and programmers working on Gamecube and Playstation 2 consoles made the most of reduced triangle counts, low resolution textures, and a handful of visual effects to deliver compelling experiences, often at a rock-solid 60 frames per second.
Virtual scenes may be easier to read without extra details since bold and uncluttered shapes can help with navigation and interactive object identification. Polygonal edges, contrast, and shading gradients are often enough to draw a user’s gaze to important points within her view. Users are also likely to immediately understand the function of an iconic object with an abbreviated outline so items with additional ornamentation might not be needed. Sticking to stylized visuals throughout the development process can also boost the production speed of a project. This is especially important for VR since many prototypes will have to be discarded as teams learn what works best in an uncertain medium.
Another thing to consider is that the more visually realistic a world is the more the user will be distracted by unusual behavior (e.g. AI glitches, collision errors). Even if you take care to prevent such occurrences, you are better off not presenting your world as strictly realistic in the first place. VR is also a potential uncanny valley amplifier. Ultra-realistic yet slightly wrong looking human characters already make some users uncomfortable when viewed on traditional displays. Seeing similar characters at close range in three dimensions is likely to be disturbing. The creation of low-poly, stylized characters sidesteps this issue. VR users are now embracing simplified hand models as well, often preferring them over detailed and specific hands that jar with expectations of a virtual self.
Standalone headsets will have storage capabilities similar to tablets and other mobile devices so it's also worth thinking about appropriate download sizes. Lean applications may have a better chance of being downloaded in the first place and might be less likely to be removed when the user is approaching the storage limit of her device. Prioritizing standalone from the start of development also eases porting since any assets intended for a mobile GPU should be trivial to render with a full graphics card. The same can't be said if transitioning a project from PC to mobile and painful content cutting may be required.
Lastly, a user may actually enjoy looking at a virtual environment more if its world is kept visually vague. Painters often leave some areas of a composition loosely textured to let a viewer's gaze rest and make the features of foreground objects feel more prominent. Wonder can be inspired in an observer who is invited to lean forward and guess at the nature of background surfaces and spaces that are only suggested by an artist. In this way, dialing back from hyper-fidelity may both reduce hardware requirements and better engage your user's imagination.
Developers are still learning how to make intuitive and enjoyable virtual reality experiences. What is known is that high frame rate is critical for user comfort and stylized art can provide performance benefits. These benefits allow a broader range of hardware to be considered VR-capable, making VR appealing to users who may not be willing to purchase high-end gaming PCs. The use of abbreviated and efficiently renderable assets can also increase production speed of new applications which fosters needed innovation within a nascent medium. Over time the language of virtual reality will solidify with solutions to interaction and art direction tailored to the medium. These are exciting times, I can’t wait to see what you make!