I, like many other, use routinely DSLR for deep sky imagining and I'd wager that most use a mix of Canons and Nikons, with few Sonys thrown in the mix. Some will be modified and some won't. Here is a like for like (if such thing exists) comparison between my modified (IR cut filter removal) Nikon D5100 and an unmodified Fujifilm XT-1 using my 80mm APO. The subject is M42 (quelle surprise!) both because it is easy and because there is plenty of emission lines (OIII and H-alpha in the main) and reflection going on (plus IR). No filter was used in either case. The XT-1 shot was taken with high waxing moon (91% illumination) nearby in a very humid night. Having left the heating strip attached to the other scope I had to fight all night to remove condensation from the front lens of the scope but some light drop-off must be accounted for in this case. Other than that they were processed in the same way and auto-stretched at the end of the pre-processing procedure. No denoise was applied either. The image show a similar area around the "Running Man" nebula to better highight the baseline noise properties of the two cameras. Note also the effect of unfiltered light in the D5100 image which is going to make both stars and nebulae brighter although both were exposed for a very similar amount of time at the same ISO (though both cameras are ISO-less). Pixel size is essentially the same for both cameras so potential resoluion is about the same but the Nokon employs a standard Bayer colour filter arrangement while the Fuji uses the X-Trans arrangement. The Fuji also does away with the anti-aliaising/de-moiree filter in front of the sensor, which might explain some of the better than expected performance in the deep red region (for an unmodified camera anyway).
