This is a comparison done by the same guy that @hank likes. In some ways, I like this guy better than the higher profile youtubers. At least he’s down to earth:
For me the most interesting part, because few others talk about it, was how much quieter the H2D was than the X1C. He doesn’t seem to know that the standard distance for measuring decibel is 1 meter away, not here, there, and anywhere. Nonetheless, he gave enough info that one can do a conversion to the standard distance, so I’m not holding that against him.
Noise was a big reason my wife wanted my 3D printers in the garage, but the H2D may be quiet enough it doesn’t face exile. Air filtration on the X1C was a bit of a joke, IMHO, but if the H2D delivers, then it may turn out to be quite good for indoor use. It would also be consistent with what one would expect from a more professional printer. So far, no youtubes have really addressed that in any quantitative way, at least none that I’ve yet seen.
In contrast, this other guy (alias “talking hands”) has some serious concerns about support layers using different materials:
He sounds competent, but one just never knows. Tao did indicate that they would be improving the firmware over the next few months, and maybe this is an area they will improve. Fingers crossed 
because I think it may be a big reason why at least some people would want to buy an H2D, and it was the highest on my list of reasons as well. I didn’t see the guy use the calibration plate, and maybe it would help? Who knows. I can’t imagine why else it might fail though. It’s gotta be either X, Y, or Z. No reason, at least not yet, to think it is fatally flawed in this regard if it is able to successfully shift between nozzles on a multi-color print, as has been demo’d to death. Only thing I can think of is that maybe the purge tower has trouble if the filaments don’t want to stick. After all, that’s what’s guiding the filament selection in such an application.
PS He did say that he relied 100% on the supplied bambu print profiles and didn’t attempt any tweaking. That leaves a lot of room for potentially finding a solution. In any case, I think a good test print to try would be this, as you want your test prints to find problems, if there are any to be found, not gloss over them. That way, after you clear up any problems that you may find when the stakes are low, you don’t run into any surprise problems in the course of printing whatever your ultimate print objectives are.