ambo, it's rather appalling, but many people apparently try to judge the noise performance of a preamp by turning the gain all the way up and listening for hiss with nothing connected to the inputs. That is approximately as intelligent as blowing into the microphone to see whether a P.A. system is on or not. The makers of the RNP acknowledge that under those conditions, its noise level will be a few dB higher than some other preamps. And I guess that a lot of other people who have never tested this for themselves are simply repeating that statement, perhaps without understanding the context in which it was most likely meant.
What FMR doesn't point out, but could very well, is that any preamp which is quietest under those absurd "test" conditions is almost certain to be noisier in actual use than a preamp with a design optimized for the actual range of source impedances and gain values that are used in the majority of practical applications. Trying to be "all things to all people" may please the marketing department, but whether that approach ends up putting quieter preamps in all our hands may be another matter, and that was something I wanted to find out. I also wanted to find out what the actual relationship of microphone noise was to preamp input noise, and to pay attention to the differing frequency spectra of noise, since a certain dB level of noise can still have a flat spectrum or have low- or high-frequency emphasis, etc., which can have a huge effect on audibility.
What I did was to capture (by making 30-second-long, 24-bit 44.1 kHz recordings) the noise of about a dozen different preamps--all of which were set to a gain level that I typically use when recording, all of which had their phantom power turned on, and all of which were being driven by the type of microphone I use most often, with a shorted test head (in effect, a capacitor of equivalent value with little acoustic sensitivity) substituted for its capsule. Thus the self-noise of the microphone amplifier was included in a realistic manner, the gain setting was in a realistic range, and these settings were the same among all the preamps.
Under those conditions, the unweighted long-term average noise levels of the preamps I tested were all within a handful of dB of one another. That outcome was interesting all by itself--any greater differences became evident only when some kind of weighting was applied to the noise samples which I'd recorded. I believe that this is partly because below a few hundred Hz, all the samples were dominated by the 1/f noise of the microphone itself. But there are also real, physical limits to how quiet a preamp can be.
At low sound pressure levels our ears are drastically less sensitive to lower frequencies, so weighting is really essential when trying to judge the practical effect of noise. With any reasonable weighting applied, the RNP emerged as one of the quietest among all the preamps I tested--holding its own rather nicely with my Grace Lunatec V3 and Millennia Media HV-3B, for example. So that's why the RNP's manufacturer says one thing (because it's perfectly true--under a set of circumstances that are irrelevant in practice for most people) while I say another, and yet we aren't really disagreeing.
--best regards