1. Price. It's just cheaper. About 40% the price of the latest Roku.
2. Development. Maybe this is a personal one for me. I keep wanting to work on a Roku Channel (app) but I really don't want to take the time to figure out their proprietary language.
Honestly, as long as you stick within their control schemes you're fine. Trying to break out of the roku app controls leads to a total nightmare so be weary of that trap ;)
It always boggled my mind why they didn't just take a JS implementation and port it, but I guess it's hard to justify throwing away IP which already exists because of a prior product.
The engineers at Roku and the community were always great though.
They really should have used lua. It would have been very simple to put in and it's robust enough and efficient for small memory devices.
While I appreciate the power/flexibility/efficiency of js it is no where near as efficient as lua. Lua keeps things as ints until they aren't needed etc. Very much build for embedded systems almost. I'm probably biased in that I've worked on a lua VM for mobile handsets for a few years.
* Stream Netflix/Hulu/Amazon/HBO Go to the TV.
* Stream videos off my NAS, such as DVD rips, home movies, etc. (For this I use Plex.)
* Occasionally, stream movies and pictures I took with my phone to the TV.
* Very occasionally, stream live events, such as the New Years ball drop, or the World Cup.
* Have an interface that is easy enough to use by non-technical people.
Chromecast, at least on paper, seems to do these things markedly worse. I appreciate it's simplicity and I'd probably get one to play around with if I did not have the Roku players already, but so far, it seems like the shabbier streamer. A Roku 2 or a refurb Roku 3 is about twice the price of Chromecast, yet seems to deliver much more value and much less headache. So my question is the same as above: why use Chromecast at all?