Here's my two cents on the K-3 AF system rumor. BTW, I don't care about the K-3 being APS-C or FF, or being 16, 20 or 24 Mpix.
As long as the AF tracking can finally match that of a Nikon or Canon DSLR with similar price, I'll be fine. The possible, rumored new 36-point AF system seen on RiceHigh (yeah, I know, it's RH, so anything coming from there has to be taken with a grain of salt, but I'm sure Ricoh wants to improve this as much as we do) could give us just that: quicker AF-S and much more accurate AF-C with more options on where you want to focus, and with accurate, dynamic AFpoint-to-AFpoint tracking on top of that!
Add in an improved flash system with a much more accurate metering and a faster X-sync speed (1/180th can cause problem with freezing a moving subject when shooting in ambient light, so moving to 1/250th of a second would be meaningful improvement for me), and I'll be more than satisfied.
All this will be great news, provided it works as advertised, which has been Pentax's main issue in the past. I'm tired of the "workaround" solutions to Pentax's mirror slap issues on the K-7 (so many shots lost because of this), poor AF-C tracking and AF-S back focusing, and inaccurate flash metering when using the flash in a bounced position. They're just lame excuses to justify why the camera is not doing its job as it should.
I used the D300, D700, D800, D3X, D3, EOS 7D, EOS 5D Mk II and EOS 5D Mk III on various pro shootings (as a 2nd photographer, you have to use the material the 1st photog wants you to use, so the workflow stays the same). And with the exception of the lame AF on the 5D II when not using the center AF point, all these cameras worked as advertised, without the need for a "workaround solution". Nikon's recent troubles with the D600 (and pretty much its denial of the problem) has proven Pentax is not the only one to suffer from such problems. Nikon's "solution" to the D600 was to have the sensor cleaned, which doesn't solve the problem, as the dust and oil come from within the camera (the mirror box, actually), not outside.
So far, only Canon has always acknowledge such issues and worked to solve them. The only time they dropped the ball was with the 1D Mk III, and they learned their lesson from this (the hard way, for people switched to Nikon and its D3 mainly because of the AF issues with the 1D3).
Canon may not be very innovative (they're actually very conservative), but they understand how important recalling faulty products is for their business. Nikon used to think like that, but now, they will release some SB-910s instead of replacing or recalling faulty SB-900s. Seems the D600 will have a successor, the D610, which will "solve" the issue of dust on the sensor. You just need to buy a $2000 DSLR. Go tell that to the guys who paid the same amount of money for their D600 cameras, LoL. Yet they were able to recall the D800 DSLRs which had fault AF points to correct the problem. Hopefully, Ricoh will act more like Canon and less like Nikon (or the old Pentax) when such issues are discovered. I'm sure Ricoh wants to be taken seriously, and they will if they do their homeworks.
In the past, the only thing that kept me from moving to another brand were financial issues (I had to sell the D700 and 16-35mm F/4 I had bought to cover for major house repairs which wouldn't be covered by the insurance, as they still had some good value, something my old K-7 didn't).
Since then, I have been more into teaching photography then making shootings for a living, so the switch wasn't as important as it was before, especially since I bought an used K-5 for $500 (so much DR and so much improvement over the K-7 in low light).
But if the K-3 lives up to the expectations, I can frankly say I won't have any reason left to switch, and I'll stay with Pentax for good. If they don't come up with an AF system that can at least stand the comparison with Canon (fast, accurate) and Nikon (best tracking ever), I'll have to reconsider again, depending on how I'm making my living as a photographer (mainly teaching or shooting?)
But to be honest, I'd be more happy to see a fast (F/4 or F/2.8) DA* or WR ultra-wide lens announced.
Right now, we can only choose between the overpriced DA*16-50mm (overpriced regarding its optical quality, not its build quality), the good but slow (F/3.5-5.6) 18-55mm WR and the convenient but average and slow 18-135mm WR. None of these lenses are ultra-wides anyway, so it leaves me with few options when shooting wide in harsh weather: stick to the DA*50-135mm or use an ultra-wide lens with a plastic cover on the camera, which collects humidity that eventually can make it into the camera through the lens...
I want a DA12-24mm F/4 WR much more than I want a new K-3, I have to say.