As you might have noticed, Corey & I have submitted Cardinal Quest, an html5/js game that got quite a lot of attention on reddit, hacker news & twitter, to 8BitFunding in order to raise some money so that we can work on it full-time and release the flash version in a month instead of a year.
Wait, Flash version?
You ask? Yep, that's correct!
But...Cardinal Quest is an html5 game! It's the game of the future! Open standards! Mobile platforms! Canvas!
That was indeed the plan, which is why Cardinal Quest's prototype is written in js and html5.
Reality, however, has once again reared its ugly head. There are 3 main reasons why the final version will be written in Flash:
Browser support of html5 features is spotty at best
Each browser has its own quirks- some have buggy canvas implementations, some support only mp3 encoding in audio tags, others only ogg.
One of them doesn't yet support any html5 tags at all (workarounds exist, but are suboptimal to say the least).
Performance is not up to par
Even tho great strides have been made with regards to js execution in recent years, rendering graphics in canvas is still considerably slower than rendering graphics in flash.
Additionally, there is a huge difference in js execution speed between browsers - it is not uncommon for me to get 4x the FPS in Chrome as in Firefox.
There is no FGL for html5 games
Flash games have been around for years- they can be packed in a single SWF file, portals know how to host them, they are easy to embed and easy to distribute.
All those add up to the fact that it is significantly easier as a game developer to sell flash games than html5 ones.
The market for html5 games simply does not exist yet - at least not at nearly the same scale as the market for flash games.
Is html5 the future? I sure hope so!
Unfortunately, it isn't the present.