Hi,

Today we already have an optimized Flash Player for mobile devices running on over a billion devices, many other chipsets, it's called Flash Lite.

Alhough we are building Flash 10 for devices and yes it will be full featured, and of course with optimizations.

We are not looking for beta testing at this time, though I would suggest checking out the Distributable Player on Adobe Labs.

adobe.com/go/distribute

Regards

Mark

www.flashmobileblog.com

Hi,

Flash Lite exists on many very different devices, all with different hardware specifications.  Some developers have found it difficult to understand that targetting a mobile device requires an understanding of performance, both of the device and the player.

We design the player to scale up from a minimum specification and so if you target a low-end phone you have to scale your application appropriately.  In my experience most developers failed to recognise that all devices are not made equal.

Flash Lite 3.1 was designed to render the majority of Flash content on the web today, at shipping this was 92%.  So you can think of it as 92% compatible with all the APIs that developers were using.

You can see a detailed API comparison here:

http://www.adobe.com/products/flashlite/version/flashlite_feature_comp arison.pdf

As for creating a high quality game on XGA+LCD you'll need to ensure that the device has a very powerful CPU in the 500mhz ARM11 range.  Aiming anywhere below that will cause Flash to struggle because the device will be strained trying to draw to the screen.

You might be better off considering C++ for such an application, but do give FL3.1 a chance and see how you get along.

Mark

We do make use of HW acceleration with Flash Lite 3.1 but in a limited way so far.  We use the OpenVG1.1 and OpenGL ES 2.0 apis and require that those are hooked up by the OEM.  That effort is clearly going to be a long haul, but we're chipping away at it and have seen great results already.

My recommendation is to throw the Cortex A9 CPU in the device, as we are working with ARM to optimize for this architecture with our future runtimes.

So in the future you'll see a mix of different use cases, where we rely more on available hardware accelleration where possible.  This couldn't be more useful when working on very high resolution screens with a single threaded runtime.

Regards,

Mark