There is no way any of my clients so far would know a picture was shot on FF or APSC. I am not sure who would know this from IQ alone without close inspection.
I could probably tell if comparing two images of the same scene taken at the same time (same light)... and I might be able to tell if conditions are less than ideal, but only with large display/close inspection.
Every factor plays into the others in some way.
Lets take your 85/1.4 for example. And lets say you want the DOF of f/2.8 on your APS camera for a headshot from ~10ft. And lets say the lighting determines that is an exposure of 1/125 at ISO200. But the lens doesn't reaches maximum sharpness until around f/5.6, so the maximum sharpness is not achieved.
This time I put the lens on a FF sensor... because the FOV is wider I have to get closer for the same framing of the subject, and because of the lower DOF I can stop down to ~ f/5.6 and have a very similar resulting DOF (actually less). So I get my subject separation (BG blur) at a sharper aperture, and there's a little less BG included in the image (because I've moved closer to it). But since I am using the entire lens's FOV and not just the "better" center, the edges may still be worse off. The lens does vignette quite a bit at large apertures, but it's minimal by f/5.6 on FF, and it's at about the same level at f/2.8 on APS. But since I've stopped down two stops for sharpness gain (central), DOF, and Vignette control, the exposure has to be either at 1/30th or ISO800 (or some compromise of both). The ISO advantage is gone and maybe even compromised beyond the APS.
That is how I can tell the images apart...
There are several "assumptions" made with this scenario/comparison that don't really hold up in real life. It assumes I must have the DOF of ~ f/2.8 on APS...that's not likely; and when there is very little (or a whole lot of) DOF changes in aperture/FL/distance don't make a huge difference (more/less of "little" is still "little"). It assumes I must keep the framing of the subject the same (not likely). It assumes I can eliminate parts of the BG included (quite possibly, but maybe not). It assumes constant/equal lighting levels outside of my control (why?). And it assumes I only have this one lens to use, which isn't the case (the comparison/results change if we look at 50mm/APS vs 85mm/FF).
The one thing that does hold true is that larger sensors always make better/larger prints/display because their physical size doesn't need "stretched" as much (that's assuming equal IQ in the recorded image). That is also discernible with close evaluation... but they're rather minor differences for smaller (typical) sizes.
The other factors (ISO performance, DR, Color, etc) are really more about the technology/generation/processing and pixel size/number. Newer is better. Larger pixels are better. And having more pixels is better. But "more" usually means "smaller"...
I wish I could say FF was some kind of magic bullet, but it's not.
APS lenses are cheaper... it's easier to design/make lenses to the same standard for a smaller sensor, and APS only uses the smaller/sharper center of less expensive FF lenses. *Because* I use FF I buy/use lenses that are sharper at wider apertures and have better overall IQ...i.e. they cost a lot more.
.
