Adobe Flash And HTML 5: The Latter's Claimed Benefits Don't Necessarily Jive

Another day, another big ‘Flash-killer’ HTML v5 announcement. As I first mentioned two weeks ago, now-in-public-Platform-Preview Internet Explorer 9 will support the upcoming update of the web page standard. And as the latest poster child for GPGPU (general-purpose graphics processor function acceleration), IE9 will (as with Adobe Flash v10.1) leverage the GPU to accelerate HTML 5 processing when possible.

Listen to folks such as Apple’s Steve Jobs, and you’ll hear how resource-intensive and molasses-slow Adobe Flash is in comparison to HTML 5. But as Jan Ozer (who I also mentioned two weeks ago) recently learned, this common conception isn’t always correct. An important qualifier upfront; Jan wasn’t able to get Boot Camp working on his MacBook Pro, so ran Windows and OS X on two different systems:

I tested on the Mac using a MacBook Pro (3.06 GHz Core 2 Duo, 8 GB RAM, OS 10.6.2) while testing on Windows using an Hewlett Packard 8710w mobile workstation (2.2 GHz Core 2 Duo system running 64-bit Windows 7 with 2 GB of RAM).

 Nonetheless, he found that the HTML 5 vs Adobe Flash results varied widely, depending on what browser was in use and what operating system the browser ran on. John Gruber neatly summarized the results:

Bottom line: Flash plays H.264 video at least twice as efficiently on Windows as on the Mac; Safari’s native HTML5 video playback is very efficient.

Neither result is surprising to me. Apple’s a big advocate of HTML 5, so you’d expect that an Apple-developed web browser would be a particularly strong performer on HTML 5 code. And unlike Microsoft with Windows, Apple doesn’t provide APIs in OS X that enable third-party developers to access the hardware-acceleration capabilities of its systems’ GPUs (aside from the hardware ‘hooks’ in industry-standard APIs used by OS X such as OpenGL), so you’d expect that the Flash v10.1 beta would run more efficiently on Windows.

What we have here is a real-life situation of shades of gray, versus the black-versus-white world view that ‘fanboys’ on both sides of the Flash-versus-HTML 5 controversy would prefer to paint. I encourage you to read Ozer’s report in detail, expand on his work with your own testing if necessary (and if so, please report your results here for the benefit of both myself and your fellow Brian’s Brain readers), and then reach your own conclusions for your particular design situation.

Followup: Commentary from the Slashdot community