Firstly, the physical size/ shape. Does the camera fit without binding on the body of the microscope, and is it small enough to be supported just by the C-mount without getting in the way, or damaging either camera or microscope?
Secondly, the control. With the camera being part of an integrated laboratory set up, incorporation into other software, rather than being controlled by a package all by itself can be an advantage. For instance, incorporating the camera into a LabVIEW environment?
Thirdly, light sensitivity. On a microscope you have no control over the aperture, and often the lighting is fixed/ difficult to change/ would heat the specimen too much. (have a look at IDT’s LED lighting)
Lastly, and more importantly, the pixel size. This may not seem important at first glance, but briefly:
The magnification with a given microscope/ objective will produce a certain size of image. If this image is magnified more by using a different objective, it can result in reduced depth of field, reduced light level etc. However, if a sensor with smaller pixels is used, a greater resolution image can be achieved without using a higher magnification objective, effectively this magnifies the image with no reduction in depth of field, as the same number of pixels on the sensor is concentrated on a smaller part of the image. So, the smaller the pixels the better, to maintain better depth of field at better magnifications.
The ‘normal’ (if there is one!) size of pixel for a high speed camera is 12 microns. Some have larger pixels, to increase the light sensitivity, and a few have smaller. Of those with smaller pixels is the NX5 (7 micron pixels) which utilises IDT’s proprietary pixel pipeline technology, which yields a better than expected light sensitivity from smaller pixels.