I just saw an interesting post over on Twitter that talks about how Flash will take over the mobile gaming space. Paolo Munoz suggests that with our current position for gaming on the web, that Flash on devices should become a natural progression for the platform.
There are some interesting points, some of which are good like the Facebook applications/games strategy. Yet some points are assumptions and not well understood, and so that’s something that I’ll explain here. This post is not meant as a criticism in any way of Paolo, but more an exploration of the fact/fiction.
Flash on Smartphones
While the delivery of Flash Player 10 to (high end) smartphones will undoubtedly be a great boon for those working with AS3, but it’s not a panacea for development. You will absolutely have to purchase hardware for the last mile of development, you will have to deal with new interaction paradigms, and you will be working within performance limits.
Device support will initially be very limited, maybe two or three devices and running only in the browser. Flash Lite 3 will continue to be the mainstay on shipping devices for around 1 year from my own projections. The Open Screen Project however enables us to drive the adoption of the latest player, where possible, within a much shorter period of time.
We’ll see betas later in the year by leading OEMs, but it is unlikely that these will be for consumers, it’s a bit of an unknown.
Flash and GPUs
In this section Paolo picks out the fact that performance has always been a key concern for the community. That’s exactly why Flash Lite has a different memory management model and rendering capabilities, it’s a demonstration of just how different things are on mobile and devices.
In the past year we’ve seen devices with extremely high resolution screens, high end processors to drive them, huge amounts of memory and OpenGL ES/OpenVG hardware. They are very expensive and typically available in major markets in low/medium volume. It’s not a huge issue in reality, but to reach users you’ll need to get smart on your targets and markets.
Ok so the use of hardware acceleration for Flash Player is first and foremost to gain “acceptable performance” in line with Flash Lite’s capabilities on these devices. We’ll be doing our best to ensure that everything is available, but everyone must understand that some devices will just not be capable.
Filters and other high requirement features may not be supported in hardware, in fact the hardware may not be there
In these situations there will be a fallback, but you’ll need to get smart on what they are, how they work, and what to do about it just in case.
We’ll be here to help obviously but to bang the drum again, get working with Flash Lite now to understand the complexities.

#1 by TJ Downes on July 30, 2009 - 4:12 pm
Good post mark, and on the money.
Even when we look at standard software built for mobile devices, they are very stripped down applications that do not require a lot of system resources to use. There are a lot of things to overcome to get there. The power required to drive large amounts of memory or processor power, not to mention any sort of 3D acceleration, is just not possible with today’s mobile device.
Even once we are able to provide this sort of power to these devices we then have to overcome the issues of heat. The more power you provide to a device the more heat it will generate. If this heat is not kept in check it will degrade performance severely and even damage the device.
What Apple has done with gaming on their phone is amazing, but they do have control over the entire platform. Essentially, the games are optimized for a single device, whereas Flash and Flash games are designed to run on any mobile device that supports Flash.
I think once these issues are solved, and hopefully they will be soon, we will start to see real progress in mobile devices as a gaming platform, with Flash being a huge contender in this arena.
I think fuel cells will eventually power our mobile devices and we can work on solving the remaining issues!
#2 by Mark Doherty on July 30, 2009 - 5:25 pm
Thanks Tj,
I’m not a real believer in fuel cells, electricity will always be cheaper because it has to be.
Programming for battery efficiency is an art that’s frankly been lost today and replaced by convenience, a real shame.
Mark