Sunday, 8 February 2015

Astronautics Simulation - Continued

I spent last week testing the simulation on my Oculus Rift (DK2). I also had two other people try it out, one who has used an Oculus Rift (DK1) before and another who has never tried it. This was a happy coincidence really but allowed me to try and gauge motion sickness when using the system.

Please see my findings below:

Main Manu:

What works:

  • The menu panels work pretty well and look really cool! Due to me using (a type of) raycasting it is able to select menu elements just by looking at them and pressing the A button on a pad or the Return key.
  • The moon orbiting the planet (One of the many BegoƱa made) really pops out from everything else on the scene and every person who has played on the simulation wanted to walk underneath it and look up instinctively.
  • Luke's placeholder skybox holds up quite well on the rift, with no seams or artifacts, users of the system always end up looking around and behind them and feel like they are floating in space. I might leave this skybox in since it seems to be working well but I'll have to do further testing on this once the other skyboxes are fixed.

What doesn't work:

  • The controls, or shall I say a lack thereof. All users of the system wanted to be able to move around freely. This was a reoccurring theme for most of the play sessions. It appears that constricting movement (which is listed as a bad idea according in the best practises document) is not the way to go. All users including myself felt constricted, when wanting to get a closer few of the menu's elements.

MiniGame 1:

What works:

  • doesn't crash? Apart from that not much. The assets and skybox do well together and the Triton moon looks great. 
  • The game functions as well, and does what it sets out to do. Only a few Oculus specific bugs cropped up.

What doesn't work:
  • The gameplay however is not ideal. This is mostly a design issues as I designed the game without ever using an Oculus Rift. The freedom of the camera controls (even when I try and constrict them) interferes with the gameplay to a large extent. Constricting camera movement causes disorientation and potential motion sickness after one game has been played through. This playthrough was also done by the person with the strongest stomach. This is what I was afraid of during development I wanted to test for.
  • There are a few bugs that occur due to locking the camera at certain times. For example the user can get turned around and due to the camera lock get stuck on what it behind them. Even when not the case, the camera panning can make them feel as if they are looking behind them, when they are just off-center to the moon. This is a pretty good simulation of the disorientation, and I suppose one could argue the game works to well! But it is not fun and needs to change.
  • The UI is far to large and blurry, these UI elements were placeholders anyhow, and need to be changed for better ones.

I will have to redesign this game. The good news being that I have an idea in mind that would be similar, but lend much to the visual feedback that seems so pivotal when using the Oculus Rift.

MiniGame 2:

What works:

  • The good thing about this game is that less seems to be more. Everything works really well in this game. All the people who tried it, appear to be able to play it for a prolonged time. Well after the timelimit. This is probably due to the fact that the game is not very fast paced. The movement is also not constricted and the user is allowed to look around freely and is encouraged to do so.

  • The skyboxes also look really cool. The areas that have no seams are really immersive and draws the users attention which is exactly what is intended.

As a side-note debugging your game level by standing inside it is a surreal experience!

What doesn't work:

  • The seams! The damn seams in the skyboxes are a visual  eye-sore and needs to be fixed. I have a new skybox ready to test and I hope that this will resolve the issue.
  • The only other issue is the lack of content to look at in terms of gameobjects. More of these need to be added and this shouldn't take to long but relies on the skyboxes heavily so these need to be fixed first.
  • Again the UI here (even though placeholder) needs to be changed for better ones that are less blurry and off center. It is tricky to view a stereoscopic UI in a 2D environment during development.

At least now with some testing behind me and more to come in the near future I can make a start on fixing these issues.