Here's my 10 minute exposure, as you can see it doesn't look too bad. The amp glow can easily be subtracted or cropped at this level. I'd be interested to see one of your longest exposures.
I'm working on a project to calibrate sensors to enable long term exposures, so for now it's a research interest, but when I'm finished, I can take some calibrated pics to show you.
I've also discovered that 'hot pixels' do work when it's colder. I think most analysts just assume they're dead, but they can work perfectly.
"Theoretically, the longer the exposure the fainter you can detect, however light pollution limits the total time." (quoting myself) - I wasn't clear, what I mean is that light pollution limits the total exposure time of each frame, for example where I live, 6 minutes creates a bright white sky at night.
As far as those articles, I know they are recommending stacks of 5 minutes, but their point is to optimize the efficiency of your imaging time. Things change subtly in my case, I'm thinking of photos where efficiency is not important, only sensitivity. I would have to read the article more carefully though. Clearly if you're at the point of waiting for photons to hit, you need a long exposure. I'm testing those limits.
As to the noise, I'm chilling my camera in the freezer and it works no problem, and the noise noticeably goes down.
Anyhow, my point was to not artificially limit programs, as you never know what your users will do. If extending my limit requires extra programming, I don't expect anyone to do that, and for my experiments I can just try my other camera. Do you know the exposure limits with the SX50 or SX60?