I'm looking at the Deity recorder, not for concert recording but for interviews and so on. I've downloaded the manual (unfortunately way too large a file to attach here) and will read it through before deciding. I'd need to get a new pair of microphones to use solely with it, which I don't look forward to, or else work out some sort of adapter for a pair of Schoeps capsules on active cables. All in all, it's at most a maybe.
-- I don't want to be tedious and pedantic, but in this this thread many people have mentioned 32-bit recording as being an attraction of this type of recorder. I think they may misunderstand what that feature can and can't give you. Just to keep it brief, Deity's claim:
> 32-bit float means you’ll always have perfect gain staging
... is misleading and wrong, but an interesting choice of misstatement since it caters to the illusion that many people seem to have. I've extracted the specs from Deity's manual; they're attached below. Please note the s/n spec, which is about that of a typical 16-bit recorder. Note that even that spec is achieved only if the preamp gain is set to a rather unlikely value.
"32-bit float" is a data format--a way of storing the sample values that come out of an A/D converter. That converter will have a certain noise floor and maximum output level given the circuit it is in; the best available converters can give ~22 bits of range between that floor and that ceiling. The 32-bit encoder then takes this (less-than-24-bit) output from the converter and translates the linear PCM samples into a floating-point binary format for storage. But no encoding of those sample values, no matter how many bits it uses, can possibly have any greater dynamic range than the A/D converter that provided them. Thus the fact that the 32-bit float data format (in the abstract) has its own, much wider dynamic range is irrelevant.
For analog microphones, any digital recorder has to have an analog mike preamp followed by an A/D converter. A given pair of microphones in a given recording situation will produce a noise floor and a peak output voltage; both will vary according to the sensitivity and noise of the microphones, and the loudest and softest sounds that occur. There's a huge amount of variation among those things--no one preset gain for a mike preamp can possibly cover all situations optimally. If that gain is too low, the input noise of the preamp may exceed the noise contained in the signal from the microphones; if it's too high, overload can occur in the preamp and/or the A/D. The fact that the A/D's output is then translated into a 32-bit data format has NO EFFECT on that situation, since it comes into play only once the numeric sample values come out of the A/D converter.
What's required, ironically, is proper gain staging--the thing that the ads say is guaranteed to be "perfect" because of the 32-bit data format, but in fact isn't guaranteed at all.
[Edited later to add: The lack of a "pass-through" feature in the U.S. version of this product means that once you start recording, you can't hear anything through the headphone jack. It reminds me of two-head versus three-head analog tape recorders, except that this is like zero-head; the meters move, but you don't know what you're getting. I don't know whether that cutoff occurs in pause mode or not. If it does, then I'm no longer interested in this recorder at all, since to set up for recording you'd have to record a test, then play it back separately; there'd be no adjusting the microphone position while listening in real time.]