Stephen b,
I understand what you mean. And I have some similar experience. The twilight factor means that a higher magnification wins over a lower brightness as long the brightness is adequate to reveal details. Once, I compared my Conquest HD 8x42 to my Pentax 65mm spotting scope at 30x. I aimed against a dimmed area, and when I barely could identify a bicycle with the 8x42 I could clearly do it with 30x65, and also see details. Yes, the image of 8x42 was brighter but the brightness with 30x65 was adequate to, with the help of the much higher magnification, see the bike much better.
However; IF the low light conditions had been much worse than that, it had reached that level when I had not been able to see anything at all with 30x65, while I still had the chance to see something with 8x42. Of course it's an extreme difference to compare 8x to 30x.
Another way I have expressed this in another thread: if you compare a 6x30 to a 10x50(same exit pupil) in low light in the wilderness: as long as you are able to see anything in the 6x30 you will see it better in the 10x50. But if it's too dark to see anything in the 6x30, you will not see anything in the 10x50 either. Magnified darkness=darkness.
Still there are more factors to take in consider here: in some cases, a higher magnification(because of the narrower TFOV) helps to avoid light sources when observing an object in a low light area. If so, 12x50 even more can be better than 7x50. But if we are observing an even illuminated object which covers the entire FOV(example: a wall or sky background), the brightness will be proportional to the area of the exit pupil if light transmission is of same level. And that undependent of magnification/aperture.
BUT: it's well known that we don't perceive brightness difference linearly, but in some way logarithmic. Twice brighter image usually doesn't look really twice as bright, and therefore we may not perceive a 7x50 as almost three times brighter than a 12x50, though that's the difference of the exit pupil area. This contributes to that we don't gain as much as we could expect by lowering the magnification in the intention to see better in low light.
Excuse my somewhat not perfect english...
Patric