Okay, thanks for the data.
I think 2.3 arc seconds isn't that bad for luminance and it is very rare that I get FWHM measurements that are under 2 arc seconds in luminance or in a finished RGB image. H-alpha is a completely different story and I've gotten narrow-band images that are sometimes down near 1.3 arc seconds (in the finished master integration). These are all for my Tele Vue NP127is (5" refractor) when using either a Sony IMX178 or IMX183 sensor.
However, on a recent trial over three nights I was able to get a completed RGB image from my 9.25" EdgeHD that came out to 1.85 arc seconds (using a one-shot-color QHY5III-178C). However, for that image I used five second exposures and ended up rejecting almost half of my subs for either star size or shape (mostly because of the seeing conditions). In this case I had a few subs that were around 1.5 arc seconds and I really don't think it is possible to do much better than that in luminance regardless of the size of your telescope (unless you image from the top of a mountain with good seeing or somewhere else that has really superb conditions -- usually when near to the ocean). In any case, I had to reject over one thousand subs to get the best fourteen hundred for the final integrated master.
Interestingly enough, even the final 1.85 arc second FWHM didn't look that good but I think that is because I was using an image scale of 0.3 arc seconds per pixel and thus even these "small" stars were measuring at something around six pixels wide. Then contrast that with the two or three pixel-wide FWHM you'd get when using a shorter focus refractor at an image scale of 0.75 arc seconds per pixel but with a similar two arc second FWHM. This may be one of the reasons why a lot of users think that a refractor produces sharper looking results (it's not the optics, just the change in the image scale).
Of course, you can always reduce the final image scale (i.e. make the image smaller), but since I was using the IMX178 I only had six megapixels to begin with and I was also imaging a pretty small object (M57, the Ring Nebula). Thus, the larger aperture and larger image scale didn't really do me much good even given my better than average seeing conditions (at least, better than I usually get).
Many seem to wish simply for better weather (usually meaning clear skies), but I'd prefer to have one night with clear skies and good seeing over two nights (or more) with clear skies but poor seeing. In fact, in most cases I'd take better seeing over somewhat darker skies.