Originally posted by severalsnakes AFAIK, the sensors in the K-30 and K-50 are the same 16mp jobbers. I went from the K-30 to the K-S2 and the sensor upgrade immensely impacted lowlight performance and noise at high ISOs for the better. If you can swing the K-S1/S2 or K-70, you're going to have a much more capable machine for low light.
I'm surprised to hear that you noticed such a large difference going to the KS-2. I'm always interested in new cameras, and when I've checked out sample images from newer APS-C cameras over the years since I bought my K-30, I haven't seen much of an improvement in image quality or noise performance compared to my K-30.
I also like to look at DXoMark sensor scores since they have always tracked very closely with the relative performance I've seen from the numerous cameras I've owned over the years.
DXoMark hasn't tested/rated the KS-2, but they did test the KS-1 which uses the same 20MP sensor as the KS-2. And it actually scores just below the K-30 (78 vs 79), with the KS-1 scoring 1061 in low light performance vs 1129 for the K-30.
They also haven't tested the K-70 with its 24MP sensor, but the 24MP sensor in the K-30 barely nudges out the 16MP sensor of the K-30/50, with a score of 80 vs 79.
Long story short, I think the K-50 is still among the best choices in its price range for low light performance.