I just watched the most recent episode of Wil Wheaton's YouTube show, Table Top, and it featured the Dragon Age RPG. It definitely looked like a neat little system even though my familiarity with the setting is nearly zero. It actually reminded me that many months ago I had downloaded a quickstart version of the rules, but after skimming through them never really got around to testing them out.
The quickstart rules look like an ideal entry point into the game as the rules are pared down pretty well (and I understand it's a pretty rules-light system to begin with), so it won't require much reading to get prepared. There's also an adventure (which I haven't read) and several pre-gens (which I did skim through).
The core mechanic is simple and quick, basically boiling down to "roll 3d6, one of them is the Dragon Die (a different color), then add modifiers." My growing preference for gridless combat is realized in this game, which defaults to Theater of the Mind style (personally I prefer the middle ground of using minis without a grid, and that's certainly easy enough to do as well). I thought the Elf with the "Smelling" focus was really cool (in the Table Top episode). The "armor as DR" is also a nice feature; I'm cool with AC as a target number in RPGs, but using damage reduction or protection tests (as in The One Ring) certainly make for a more realistic differentiation between light and heavy armors.
Finally, I think the stunt system is phenomenal. It calls to mind a blog post that I recently read on monsters in 13th Age. Namely, the idea of roll information density, which to put it simply means that a single roll generates more information than a simple binary "success" or "failure." The same hold true for the narrative dice in Fantasy Flight Games "Edge of the Empire," the new Star Wars RPG. In 13th Age, certain die results ("natural even hit," "natural 16+," etc.) carry additional effects as specified in the monster's stat block. This gives a single roll more meaning, but it also saves time by reducing decision-paralysis. Instead of deciding when you'll use a specific ability, that ability is triggered by the die roll. In Dragon Age, rolling doubles triggers a stunt, and you get a number of "stunt points" equal to the value of the Dragon Die. You can spend these stunt points however you want, with better stunts obviously costing more points, but you can also stack smaller effects. It provides meaningful tactical choices while still being quick to resolve at the table.
The biggest downside are that there are only 3 classes (Mage, Warrior, and Rogue), and characters seem to be a little "same-y" based on my quick scan of the pre-gens. Most differentiation seems to be narrative, and while I like narrative elements to have a big impact they don't seem to be terribly well-supported with mechanical consequences. Granted I haven't seen the character generation rules so I'm not sure how stats and focuses are allocated, but as far as I can tell what your good at seems to be determined in very broad strokes ("I have high Str, grrr! I can smash things and hit things and climb good!"), with a very limited number of Focuses. Furthermore, a Focus is pretty narrow and is tied to a certain Ability (of which there are 8). I think I would just prefer a greater number of Focuses, a shorter list of Focuses, and/or Focuses that were more broad. There's also disparity between focuses; for example, compare what is essentially weapon focus (axes, heavy blades, etc.) with Calligraphy, Etiquette, or Rowing. The result I predict is that with 3d6 a +2 modifier is going to have more of an impact than it would on a straight d20 roll, but you get that modifier in such a hyper-specialized area (or with every attack roll in the case of the no-brainer choices) that for most tasks every character feels more or less the same. Anyways, I hesitate to criticize this too much without seeing the full rules or even playing the game; it's just something that jumped out at me.