Open Research Challenge to GearVR Devs

A GEARVR NEWS CHALLENGE

by VRift720


GENERICBANNER_TITLE

INTRODUCTION:

      Hello Friends and VR Enthusiasts.  I have some important questions for you, especially if you are a developer for VR applications, games, or experiences.  I not only have some questions, but I have a challenge for you, if you’re game.  If you’re up for it.  It’s your call, but I think it could be good for everyone if someone will take this idea seriously. 

      This post is lengthy, but I feel it’s worth it if this idea will help all of VR out like I feel it would.

Here goes:

  1. Do you as Devs know what we (the Masses) want from VR?
  2. Do you know what we like about VR in general?
  3. Do you know, in great detail, what the Masses most enjoy while in VR?
  4. Do you know which Room-vs-Avatar “Scale Ratio” we mostly prefer?
  5. Do you know which depth level of stereoscopy we mostly prefer?
  6. Do you know exactly how motion sick the average person can get?
  7. Do you know precisely which actions in VR feel the best? ….the worst?
  8. Do you know our endurance and limitations as a whole?
  9. Do you know exactly what causes us the most grief? … or Pleasure?
  10. Do you have a VR Rating Scale for Each VR User Based Against Our Peers?
  11. Would you like to make games for specific subsets of people, knowing exactly what they like and require in your game’s VR settings to make that game a success?

What if there was a way to know?  A way to gather all of this data into one place and measure it against everyone else out there?  Wouldn’t you want access to that data?


EXPERTS:

      For the most part, there are a lot of industry experts solving these issues.  But their research is expensive, time consuming, and not readily shared if there is a way for them to use that knowledge to make more money by keeping it close to their breast.  The top companies like Oculus and FB are sharing their data, true, but they still don’t know everything yet, and the process is slow.  But what if there was a way to speed that process up and help everyone in the process?   

     The problem with all of these experts is that they are not actually asking the people who know best:  US, the users of VR.  And they can’t really just ask us, because there isn’t a tool available that can do this.  BUT THERE COULD/SHOULD BE!

      It would just take ONE Developer to set aside their current project and slam this out, not only for themselves, but for the good of the whole VR Industry.  I feel VR could gain immeasurably from what I’m about to propose…

GO TO THE SOURCE:

     I am calling for the development of a “PERSONALIZED-TESTING TOOL FOR VR” … whose job is to take one user through the entire lock, stock and barrel of VR issues, problems, bad practices, and good experiences, and let them rate each experience from 1 to 10 after experiencing it.  The premise is simple:  the user is sent through an experience without interruptions the first time (Trial Run A).  Then he goes through it again (Trial Run B) where it pauses just after each key position Developers want data from.  The user is then asked to rate that part of the experience in different ways. For example:

  1. Exactly how motion sick does this part make you?  1 being not at all, 10 being worst?
  2. Exactly how fun was this moment for you?  1 being no fun at all, 10 being most fun?

      The user would be subjected to different motion-sickness-causing events for a duration of time and asked to rate each effect. The data would aggregated and pooled against every other anonymous uploader and then that user would receive stats like:  Motion-Sickeness Rating, VR score, Presence Identifier, Scale Code, Fear Rating, and various other factors this test would help to reveal as contrasted against their peers.

      For example, for the Scale tests, they might experience the same beautifully-lit room at different scales and rate which scale feels the best against that room’s size.  That data is then compared against their height information to determine if they like vantage points similar to their real height or if they prefer new experiences with vantage points different than their own. Heck, the test could even change lighting and ask how much light the average user enjoys in their environments.

      Essentially, the app is asking “Does enjoyment come from fresh perspectives or does only discomfort come from those?”  Which one?  It would be important to know, and these tests would help reveal such data. 

      The industry would submit more test ideas beyond this article’s suggestions as needed and the Developer for this app would continue to increase its value to VR on the whole by implementing other Dev’s ideas for tests that would help them grow VR.

STARTING QUIZ:

      The app would start with a basic quiz about each user’s statistics, anonymously recorded and uploaded to a server for analysis, weighed against all other test takers.  Height, Weight, Blood Type, Race, Gender, your IPD (interpupillary distance, and the app would let us solve that with some kind of a test the app gives), and other basic statistics as needed or required to learn more about human VR preferences as a whole.

      These stats would help the test formula learn how majorities of people react differently to the same stimuli.  Maybe there are issues that some racial stimulus is responsible for that nobody ever knew about before, but which this data could revealonce assimilated? 

      So that if you tell GearVR your details in the GearVR Initial Setup Menu, certain games and apps could be configured … just for you.  What factors, in common, determine people’s reactions in VR?  Race?  Or Age?  Or Gender?  How do those traits effect one’s VR experience?  What if it’s your blood type, or what if it’s someone being rated a Type A or Type B personality type?  Some people love roller coasters, others are deadly afraid. 

      What might we learn by taking a comprehensive look at what things exactly cause people to enjoy Presence more than others?  Meaning that the Industry can begin to code for these various situations, leading to more comfort and enjoyment for every user who participates?

     Some Devs could take this data as a way to pitch a game for a certain segment of the populace who took the test.  For example, a Skydiving Game might appeal to 10,000 test takers who favor serious risk and want to feel the full force of falling in VR.  They can not only create the game for those participants, they can also advertise the game to the users whose Profile has that same rating. 

      For example, if our Motion-Sickness rating is “Low” and our Fear Factor is “Low” and our Action Level is “High” then we would get a message in Oculus Home Store saying “A SkyDiving Game is coming out in 3 more months and we’re letting you know because of the the settings you’ve entered in your GearVR Profile.  We’re sure you’re going to love this game!

GRAPHICAL TESTS:

      Next after the Basic Setup Questions, a series of graphical tests would begin. 

SCALE:  The Sense of Presence is very low first, set to 2 out of 10. The user is put in a 10-foot tall room at their natural height given during the basic questionnaire which goes first.    The Scale test then iterates through your natural height, then goes down in 6-inch increments until half your height.  And then it goes back to default and works upwards to 14 feet tall.  The user rates each step as to how much they enjoy each height level.  

       And then again, set from 2 to 4.  The stereoscopy iteration is to ask us if our sense of scale is enjoyed more or less as the 3D Depth level begins to increase in the scene?  And the system can tell us what actual height scale the majority of people prefer. 

      Next the test iterates through again like that until it reaches Level 10 (max stereoscopy).  Once completed, that data is aggregated, saved, and uploaded to this app’s servers.  And we, the users, proceed on to the next set of tests:

Square Spacer

STEREOSCOPY TEST:  We are put in a nice room with tons of butterflies flying in really close, right in our faces.  The test is explained.  The Butterflies will freeze at different depths around us.  We are to look around and find the butterfly that looks the best spatially, the most perfectly 3D-Located butterfly and tap to remove that one first.  Then choose the next best one.  We click each butterfly until we reach the butterfly we feel is the least interesting depth wise.  The app knows the exact depth of every butterfly and can pool those results.

      There are 30 butterflies, starting at 6 inches from our face, and going out to 15 feet away in increments of 6 inches or so.  This tells us if users prefer really close objects, far objects, or middle-spaced objects, and what the best sweet spot for depth really is on the current GearVR unit.  (The test can be updated for each generation as they improve technically.)

Square Spacer

STEREOSCOPY TEST 2:  This one measures our ability to judge 3D Depth.  The Depth information is not given to us.  We must guess.  We are placed in a room with a grid-like pattern on the floor that gets smaller as it goes out.  We are not told what the grid-size is, but standing in the middle of one, we might infer its 3×3 or 2×2 depending on our elevation (size).  We start out at our Natural height.  An object appears in front of us and we are told to choose it’s depth.  We get 4 choices to choose from. 

      The object moves 5, 10, 15, or 20 meters into the distance with each jump, randomly, so we can’t ever really know how far it jumped forward.  We can only guess.  The app compares our guesses against the true depth and that data tells game designers how bad people are at reading depth.  Or how good, let’s hope!  

      It also tells game designers how much 3D Depth people actually see in each device as they come out with the eventual technology improvements.  One part of this test can also just ask “Click 3 for 3D or 2 if you feel the object now looks 2D because it’s too far away to see any 3D effects…”  Then objects pop in and out at different depths from the user’s face.  They score it accordingly.  This would determine the most-agreed upon 3D-Depth Cutoff point for users of GearVR CE1.

     The App lets each user know after the tests are done what their 3D-Depth Rating is (compared with others) and how far they can still accurately see any 3D Effect.  Devs can learn all the secrets of how to build their worlds to better stay within the limits of each device’s capabilities and at which depth to cater the best effects for maximum coolness.

Square Spacer

MOTION TESTS:  These tests seek to understand our comfort level when doing a number of different bad/good/interesting movement-related things in VR.   The user is asked to sit down first (for real) and the app also matches their height when sitting down in the coaster.  They’re asked to look at the virtual keypad floating to the right of their faces after each test has been performed for when we enter the Paused state.  There is a roller-coaster track and a number of different tests are done on it, as follows:

  • We take off, going from 0 to 60 instantly.  Score?
  • We take off, easing in in speed up to 60.  Score?
  • We’re already going, and ease to a stop.  Score?
  • We’re already going & the demo freezes after 20 seconds.  Score?
  • We go up a hill along a curve.  Score?
  • We go up a hill with a jerky transition.  Score?
  • We go up a hill and slow down a bit.  Score?
  • We go downhill and slow to a stop. Score?
  • We come off the track 10 feet over a hill & land back on it. Score?
  • We go through one inverted loop. Score?
  • We do a 180-degree twist.  Score?
  • We turn sideways 90 degrees while moving forward.  Score?

      In the interest of brevity, you get the idea here.  Every physical motion that can be done is done here, and scored.  Various lengths of time are also a part of this process, from 30 seconds to over 5 minutes of this, to gauge how many people can go for how long before reaching their limit?

      A number of tests are grouped into a small unit and called “Level 1” so as the user progresses, the tests don’t feel too long, but are broken up and spaced out to give them some space.  The music changes occasionally to keep things fresh.

     If you reach motion sickness you are supposed to look down and tap the track pad simultaneously.  That portion of the demo will instantly end to keep it from making you hurl.  Any concluded test portion can be saved and resumed from later, if necessary to regain your physical composure and well being.

      In addition to these crazy tests, more motions tests are carried out next:

  • Standing up inside the game.
  • Turning left.  Turning right.  Turning around.
  • Leaning forward.  Back.  Left …. Right.
  • Dive rolling with accurate POV (for FPS games).
  • Driving a car.
  • Skydiving.
  • Falling huge distances.  How far until the tingles wear off?
  • Watching a conveyor belt move below you, you last how long?
  • Jumping (self controlled).
  • Camera shake.  Iterations in 0.5-second increments ending at 10 seconds.
  • Blink-Turns.  Blink Forward, Blink Back.
  • Being knocked down, back, the ceiling coming into view.
  • Swimming on the surface of water, head just above the line.
  • Climbing up a cliff.
  • Riding a bike.

      There would be way more tests, but you get the idea.  The user can take a break any time and return to complete new stages later, as desired.  It’s not any rush, the test should be comfortable overall, with deeply uncomfortable moments because we’re here to test the limits of human endurance as a Group and determine what the safe and enjoyable values are across most VR situations for most people.

FINALIZING THE TESTS; SCORING:

      It might take an hour or two to finish all of these tests but the users can be told that they can save their progress after any test is completed and come back later to finish up and receive all of their ratings and codes.  To see how they really compare against other VR users. 

      The app should look pretty and be fun to sit through as much as possible, with great graphics that make the user feel like it’s a real space.  After they complete their tests, the app uploads the final results where they are profiled against every other uploader and various statistics are given to help the user know their entire situation.  This ranges from comfort level, discomfort level, Presence Rating, 3D-Depth Rating, Motion-Sickness Score, and more …  It tells us how we stack up against the rest of VR users out there.

      This test could really teach us users (plus the Developers) so much about ourselves in VR.  It could lead to learning how to better cater their games to the kinds of fans they most wish to reach. 

      That data is to be shared with the whole industry, even the users, at any level.  But in order to remain anonymous, the app does not ask for a login.  It will prepare the test’s information in the Cloud for about 1-2 minutes at most.  And tell us the percentage of being done while we wait. 

      Once complete, it sends our Finalized Survey Report back to the user (still in GearVR), who can request to “Save this Data to My Phone” externally as a data file.  Or they could click “Map this Data to My GearVR Profile” which would of course be something that would have to be created by Oculus first in order to work. 

      A Profile Engine in GearVR would be made to work with this data.  The scores and the titles could then all be registered in our profile after receiving the Report from the company at Test’s End.  The anonymous data used to create the Ratings would not be recorded anywhere, as it is private.  Once the app maps the data to your Profile and saves your Profile successfully (including a backup file), it erases all of the data from the Test App and resets the App.  Each account can only do the test once per device, or once per year to keep the online system data from fluctuating too much.

ONLINE TEST-DEMOGRAPHIC REPORTING TOOL:

      One feature I am requesting if this app is made is that Devs will have the ability to load up and study an online-video version of this demo and then see groups of statistics or demographics for all users at each second of progress in the demo. 

      They will be able to scrub back and forth through any of the tests and know exactly at which point the most people started getting sick, or had the most fun, and draw their own conclusions as to the reasons from what is occurring in the scene at that position

      It is vital that Devs be able to research how we all reacted as a whole to every test as it was carried out.  So the tests all need to have real-time data gathering by making the users report their reactions at each point in the test as it is proceeding along.  The points could be hard-coded and a pause engineered to let the user score the previous segment where 1 is comfortable and 10 is extremely uncomfortable. 

      This is essentially scientific research and therefore must have a scientific method embedded throughout to offer the best analysis for each test given.

GEARVR TEMPLATES

     GearVR could then begin to have a system where Devs could apply Comfort Templates to each user based on their ratings, scores, fear factors, height, weight, motion sickness, and many other factors. 

      If the user didn’t like the results of their template, they could tell their profile to adjust the template toward more danger and discomfort if that’s what this user feels is missing, or away from more danger and discomfort if that’s what’s bothering them. 

      The data gleaned from this testing app could become invaluable to everyone in the Industry, from gamers to developers, to Samsung and Oculus themselves.

CONCLUSION:

      I really hope the Industry likes this idea, hears about this concept, feels I am correct in its value, and that ultimately someone out there will respond in favor of creating this idea for us.  So that our VR experience can improve in every way once Devs really begin to know what people really need, like, dislike, and don’t want in VR through essentially … asking us directly … but in a sort of fun way. 

      One last idea, a motivator:  the people who take the time to finish this extensive test to help VR along can also get $40 of in-store credit (in the Oculus Home Store) for taking the time to finish the survey and send the data for collection.  That would be nice. 

Your thoughts on this idea?

GENERICBANNER_TITLE


BACK TO MAIN PAGE

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s