Man do I love these roadmaps

I was talking to someone at the AVS Party at CEDIA about Media Center and the subject of performance came up. You may have heard that Microsoft will be introducing new Media Center Extenders with the ability to stream DivX/XviD content (ahem give us x264 support too) from your Media Center PC. Great right? The problem is that these Extenders often do a poor job of replicating the Media Center experience, they are not as responsive as your $2K MCE PC and at the end of the day it doesn't deliver a great user experience.

I was playing with the Xross Bar interface on some of Sony's receivers at the show and although it looked great, the animations were ridiculously choppy and slow. I want instant response when I select an item and I want smooth animations; we can run games like Gears of War and Bioshock at 2560 x 1600 but we can't get text menus on non-PCs to scroll smoothly?

Fundamentally it boils down to an issue of processing power and cost, with CE (consumer electronics) manufacturers working to keep costs down as much as possible. The problem is that the demands we're placing on these CE devices are increasing, but there are honestly no good microprocessor options today that can run the applications we're throwing at them. Sony can't simply choose to put a Core 2 in its receivers; most CE manufacturers even hate dealing with heatsinks on processors, much less something that has to be actively cooled.

Both AMD and Intel have super-simple, low cost x86 cores on their roadmap, designed specifically for these types of integrated applications. AMD's Fusion strategy addresses these issue specifically and although Intel hasn't come out and confirmed it, they've got visions of a similar strategy.

Here's hoping that in the next 2 - 3 years we'll have powerful enough mini-x86 cores to handle the more demanding CE and mobile applications. I'm fully expecting to see some impressive advancements in this area, and I've got a feeling we'll be hearing more about Intel's plans at IDF in a week.

The benefits of having x86 processors in CE devices are more than performance-oriented, there are also significant benefits on the software side. Maintaining multiple codebases across different architectures is a pain, just ask Apple. I firmly believe that the CPU Apple wanted to use in the iPhone didn't yet exist: a low-cost, low-power x86 chip, to enable a somewhat common codebase between all of its products. Even the Apple TV used an x86 chip that isn't widely available in order to keep a common software platform.

Don't underestimate the importance of being able to have everything run on the same platform. I want to be able to run Firefox on my fridge dammit.

We've got the ability to run far more connected/automated lives today, the problem is that most of the solutions for things like home automation, media sharing, integration, etc... are all very proprietary. They all run on different platforms, with different software stacks, most of which don't talk to one another without custom development or sloppy interfaces. The goal still is and has always been: all your content, at any time, on any device, anywhere.

We have the ability to achieve a crude version of that goal today, but what's holding us back the most is a lack of a common, mostly open, platform. Everyone wants you to buy in to their own closed platform for home automation, content distribution, computing, etc... but that's not doing anything to advance the industry. I'm hoping that a broad move to x86 in CE devices will at least pave the way for more applications that will run on everything from your PC to your HD-DVD/Blu-ray player, with a very similar experience regardless of what device you're running on.

I feel like I've delivered an overly complicated version of where I see things going, I'll give it another try when my head is clearer in a couple of days :)