What makes me hesitant to forge an opinion about Tegra 2 right now has multiple reasons. Tegra took ages to get into products after the initial presentation. Other manufacturers have much more experience in the embedded market and I plain don't like it when companies that once had a big name rush into a different market and claim "that's mine". Nvidia's CEO is a really arrogant person. That chart about Tegra's theoretical market that completely ignores the competition shows the thinking behind their strategy. I think it can only fail. They might build some nice tablets and some entertainment systems for cars, but I doubt that they will be able to penetrate the market enough. I see Qualcomm as most promising company for getting marketshare right now.
Also I think that Cortex A9 is just Cortex A9, no matter what graphics is used. The chances are high that the CPU becomes the bottleneck. ION has basically been a desaster. The shared memory system cripples Atom's performance in applications and the Atom cripples the graphics performance. As long as there is no realworld benchmark available, I don't believe in shiny product presentations. So far I haven't seen Nvidia get it right in the past and personally I'm not interested in 1080p or something. Tegra 2 looks interesting, but I'll wait for some substance first