I shoot with a friend who has a K-70, which is his first camera (picked it up in April, 6 months ago) other than his smart phone. We shoot Milky Way landscapes, so essentially the same problem you are having for our landscape shots (without astrotracing).
I shoot with a K1 with a DA 15-30/f2.8. He is shooting with his K70 and my borrowed Sigma 18-35/f1.8. Essentially, we have found them very similar. The K1 has about 1 to 1.5 stop advantage over the crop sensor K70.
However, the K70 with the Acceleration Chip picks up 1.5 stops, starting at ISO 600. What the K1 give up with a slower aperture (f2.8) lens the K70 picks up with a faster aperture (f1.8). So, what we have found shooting together over the last 6 months, is that the kits are pretty equal in terms of overall results.
The K70, with its Sony sensor is also
ISO invariant. Which means that you can shoot, capture the light and then make your adjustments in post essentially loosing nothing. Actually, you can gain by shooting at lower ISOs which tends to limit the noise, while capturing better color (star color), and then boost in post. Now that does not mean that you will capture no sensor noise. Note - actually we have found that the K70 also has the same white dot problem as the K1 has.
Shooting off tripods, we crank up the ISO to whatever (51200) and shoot say a 2 to 5 second frame to check framing, levelness, etc. then lower it to ~800 for the rest of the shooting. For the landscape segments, a 1 to 2second exposure (sometimes 4 seconds) or several 2 second exposures stacked, yields some pretty nice results.
In post processing, you can then adjust your exposures to keep the sense of night, while bringing out some of the shadow detail (well in the dark of night, everything is in the "shadows"). But, the key is - with the sensor being ISO invariant, you have the latitude of adjusting the exposure in post to provide the overall sense that you are looking for.
Here is a shot at midnight (3 frames stitched). Actually this is a test shot, that I'm using to help me process a 27 frame stitch of exactly the same view. I'm trying to get the lighting right. This was shot at 15mm with the large stitch shot at 30mm which captures about 4x more light.
Here is shot - 18 frames stitched - that I want to reshoot again - trying to get the headlights to light up the canyon at night.
There is a lot of trial and error while you are shooting out in the field, but with some reasonable post processing you can get the overall look you desire. It's not like shooting during the day with a light meter, where you can meter the scene and expose for perfection. This is more hit and miss, getting a feeling for the location, and the overall darkness that you are dealing with.