Blender cookie did a video on Blender's new Bmesh addon which will make it an ngon modeler.
Don't know if that's of interest to the game community or not, really, but a lot of low budget game developers use Blender.
I don't think it would directly affect game programlng, since it will be triangulated just like quads were before. Anyhow, it's great to see the project improving.
Yeah, it's more indirect. It makes modeling a little easier, but in the end, you need to have triangles for games.
Looks similar to the fgon feature already in Blender, but just auto runs the logic on each mesh change. Still, I'd be weary of using a 3rd party Blender builds. Last thing I need is to model in one version only to lose it later down the road. The other caveat is that Blender's tessellation routine never evenly distributes triangles.
Often when you want to optimize this mesh it will produce less than desirable results due to shared vertex structure. From my experience, if you're going to model for games you need to get down and dirty with triangles and optimize it from the get go. Or use curves and surfaces (nurbs) as they tessellate very well.
It's going to be incorporated into Blender 2.63 or something like that. That video I linked to shows some reasons why it actually ends up making a better mesh in the long run when you do eventually return to triangles or quads. I'm not going to give it a serious look until it is fully incorporated. I'm still not up on the 2.5 changes. I did an animated model for someone a little while ago and had a whale of a time getting everything to work right. I still haven't learned weight painting. I just choose vertices and groups for each bone by selection. When you are dealing with game models, you really don't have to get very fancy because of the lower polygon count. I just baked all my color textures onto one bitmap with that model, also, which was pretty cool. I'm not up on engine stuff, but the guy I did it for said it would render faster that way. I want to try a high poly model baked render for normals onto a low poly model in the future. It's really hard to keep up with all of it.
I can see how it's slightly more beneficial for the artist, but I would hold my reservations whether Blender can properly triangulate the mesh afterwards. I've had experiences as shown above that required effort on my part to correct. Perhaps I missed the important part where he describes the intelligent algorithm Blender uses. He tends to babble in the video.
Indeed combining textures into one will render fast. If lighting remains constant, it's the best way to go. If you're developing for mobile devices, it's the only way to go Baking normal maps is fun too. I'm not sure how to do it in Blender, but I wrote my own program that does it for me. If you check out the Stanford 3d scanning repository, you can download and test with these models.