Roku Developer Program

[EnTerr]

I think Roku platform has great potential for casual gaming as a microconsole. The competition - Amazon FireTV (box and stick), Nexus Player (+ the new NVIDIA Shield, Razer, + Android TV at-large), PlayStation TV, the indies (Ouya, Mojo, GameStick) - are already at it. Roku's potential as a micro-console is quite interesting one - both tempting and challenging, because:

1. Roku has a lower price point, about 1/2 of that of the dedicated microconsoles. Which leaves only FireTV and Nexus Player competing in the "complete, with a controller, for under $100" corner. (Yet a strong competition that is, since they are accomplished streamers too!)


2. Large existing deployments - even if i consider only the OpenGL-capable models (that's models 3xxx and 4xxx), there are at least 7 million of them in use.


3. Low technical specs (low CPU/GPU performance, RAM, flash), no doubt because of (1), need to lower cost to manufacture/BOM.


4. Limitations of the controller/remote: the IR remote does not support multiple simultaneous button presses. The radio remotes (BT and WiFiD) hardware supports it but that is not exposed to BrightScript. In addition the gyro/accelerometer API is not accessible from B/S even after multiple developer requests.

Mind you, #3 (low specs) does not mean "useless for gaming". It is more of (to borrow from Terminator Genisys) "old but not obsolete". In my tests, performance-wise the #42xx models (Roku 3, "2015 Roku 2") are on par with FireTV Stick, and there is lots of games coming for Amazon's Stick.

#4 (remote limitations) affects the choice of controls for the game. For example, half a year ago one Romans_I_XVI was disheartened by the limitations of the remote but after some encouragement + idea, he turned said weakness into a strength: using only the 4 arrow keys to control a space ship, what can be simpler than that? (The hidden ergonomics here being that to press another arrow on the D-pad, one have to release the previous key, thus avoiding the remote issue). The game he wrote - "Retaliate" - is now topping the chart with over 100K installs (guestimated), in only 4 months - so yes, there is hunger for games on Roku.

Both #3 and #4 limit the kind of games Roku is suitable for. Yet there are casual games that are a natural fit! For example, puzzles (Tetris, 2048, jigsaw, Bejeweled, MInesweeper, Lode Runner), side-scrollers (Flappy Bird) and fixed shooters (Space Invaders), board games (chess, Reversi), Pong...

There are things which the Co can do to help in developing more games. Relatively uncomplicated things like:

1. Expose gyro readings from the "game remote" to BrightScript apps


2. Via ECP, allow mobile apps to "inject" gyro readings into the player, thus turning smartphones into "game remotes". (I for one am looking forward for opportunity to integrate that in my remote app)


3. Add bluetooth back in the next hardware iteration, to support bluetooth HID profile. So that a wide variety of wireless gamepads can pair and be used with Roku. And of course RokuCo will sell one "the Roku gamepad", which development will take near 0 effort (in two steps: one - get a generic HID gamepad branded in purple; two - profit). Besides, being able to connect such controllers with "proper" joysticks will make game developers take you more seriously.


4. Fix the simultaneous buttons handling so it works in B/S, at least for the radio remotes

Your thoughts, Roku and non-Roku people?

[Komag]

One more
E. Allow multiple sounds to play at same time, not just one (or two, depending on model), but more like 5 or 6.

[EnTerr]

And there you have it - the just announced today, new AppleTV with game controller and apps. I for one am itching to see what the API* is like.

Run, Roku - Run!

"Komag" wrote:
One more
E. Allow multiple sounds to play at same time, not just one (or two, depending on model), but more like 5 or 6.

That's a question of hardware and my impression is RokuCo uses whatever is at hand but won't go out of their way to add fancy (and expensive) chips or do software emulations. Illustration - the very limited (limitations undocumented) audio format support. You can work around the limited channels with the smart idea by @BradC though.

PS. (*) if you are a registered Apple developer, apply in their free hardware kit raffle before noon 9/11, here https://developer.apple.com/tvos/developer-kit

[Komag]

Yup, that's what I've done, I have a full priority system based on previous sound length and importance, along with a volume threshold based on distance and whether it's behind closed doors or not, all taken into account.