For all those citing the supposed DR deficiencies of the Sony 24MP sensor, I would highly suggest taking a look at DXO's evaluation of the Sony A580, a camera known conclusively to have exactly the same sensor as the K-5, and is a traditional DSLR with no pellicle mirror.
DxOMark - Compare cameras side by side
DXO gives the A580 a DR rating of 13.3 EV, in contrast to the K-5 which has 14.1 EV, despite both having the same sensor and both being traditional DSLRs. On the other hand, the NEX-7 (also no pellicle mirror) is given a rating of 13.4 EV, slightly above the A580, so if anything there is even more DR to be potentially squeezed out of the 24MP sensor compared to the 16MP.
I would wager the difference is not inherent to the sensor, but to Sony's implementation as seen from their previous cameras. If Pentax can do whatever they did before (14 bit, ISO 80 or whatever) I'm guessing there is no reason why the 24MP's DR can't be as good as the K-5's.