Cameras: Pixel Size
Pixel size determines how much space a particular camera covers with a particular telescope. In general, you want a pixel size which generates about 1-2 arcseconds per pixel when placed in your telescope. If it's under 1 arcsecond per pixel, you'll oversample the image. What that means is you're not really gaining a benefit in detail because the scope is being pushed too much for deep space objects.
If it's over 2 arcseconds per pixel, you'll tend to undersample it. If you undersample too much (the pixel is too large) your stars can start to look like blocks instead of round. It's more common to see it in older cameras where the pixel size was really large. It's not something to get too concerned about with video astronomy and the modern cameras out there as we're not really striving for perfection. Even if it's a little undersampled, you can still get great images.
Here's an image that was undersampled. The arcseconds per pixel is 3.6 which is above 2, but the image still looks acceptable.
Oversampling is better than undersampling as it can be corrected by binning. Binning combines pixels in software to form a larger pixels. For example, 2x2 binning takes a square 2 pixels wide by 2 pixels high and makes them into one bigger pixel. The downside is that it reduces the resolution of your image. For example, with my ASI294MC Pro, the sensor has 12mp. If I bin it 2x2, the number of pixels drops so that it looks like a 3mp sensor.
Here's an image of M17, the Swan Nebula, that's oversampled (0.6 arcsec/pixel). It was done with my 12" Skywatcher Dobsonian
In both cases, the image looks pretty good.
Here's another image (M8, the Lagoon nebula) where I binned the ASI294MC Pro for use with the dobsonian to get 1.26 arcsec/pixel.
So, bottom line, pick your camera so that you're close to the 1-2 arcseconds per pixel but don't be overly concerned about it.