To Plugin or not to Plugin? - user demographics and the Google Chrome NPAPI problem

[Brian Ellis]

If I were to make a game that depended on the advanced rendering options available only to the plugin, can anybody tell me roughly what percentage of users will turn away without installing the plugin? (and understandably, that is before the v42 update to Google Chrome)

For those who are unaware, Google is phasing out support for the NPAPI plugin architecture. And as of v42 released on the 15th, NPAPI support is disabled by default (and many users are unlikely to ever enable it). Is there a road-map to migrate the plugin to a newer API/architecture?... is there even an "agreed upon" API to replace NPAPI as an industry-wide solution, or is the world of browser extensions falling back into chaos?

These things will ultimately have an impact on user demographic, and before I direct my team towards one platform or another, I would very much like to know how much of an impact these things will have.

- Brian

[Ian Ballantyne]

I would suggest looking at a site such as http://webglstats.com/ that describes what percentage of visitors have the capabilities required to run without such a plugin. For Turbulenz at least the majority of features (aside from a few physics features) are available in the WebGL/Canvas version of the engine and don't require the plugin. Since the adoption of WebGL in Internet Explorer along side Firefox and Chrome, the canvas option has been the main focus of the technology.

As far as phasing out of NPAPI, since the WebGL/Canvas implementation is favoured in Chrome, the plugin is less relevant for Chrome users than it would be for older browsers without a WebGL option. I don't foresee a move to a new API at present however Turbulenz is still investing time in the native Turbulenz Engine on which the plugin is based as we use it our games on Steam: "The Marvellous Miss Take" and "Oort Online".

It seems that Google is happy that the functionality that plugins provide is currently available in the browser. As far as Turbulenz Engine is concerned there are certainly a handful of features that the plugin could provide that the browser doesn't but as far as making games with the engine goes this is more related to the limits of WebGL 1.0 capabilities.