On Tue, Feb 02, 2010 at 12:25:16AM +0100, Jörn Nettingsmeier wrote: > i'm no circuit designer, but i'd also argue it's not just the preamp > design, but also the fact that you only have a very low voltage to drive > it, and possibly rather cheap converters, which should all add to the > EIN value of such a device. One of characteristics of a good design is that the input stage dominates in the noise figure, and that the EIN stays more or less the same for say the upper 20 dB of the gain range. So to get the full picture you need EIN at different gain settings. A peculiar thing for instance is that the cheapo 8-ch Behringer AD/DA has quite a decent noise figure *at maximum gain*. But it degrades immediately as gain is decreased (and the first mechanical detent on the gain pot can be anything between -5 and -15 dB). > so the comparison to "where it's getting interesting" is a bit off imho, > given the price range of the units. you'd hardly find a single > "interesting" preamp for the price of one of those entire gadgets. > (or at least that's what i want to believe, having just paid an arm and > a leg for "interesting" stuff...) An EIN of -125 dBu, unweighted RMS, 20kHz from 150 ohm starts to be interesting because you're approaching the theoretical limit. Anyone claiming less than -131 dBu for this would be selling snake oil. OTOH, many active (condenser) mics have a source Z well below 150 ohms, so figures for these are interesing as well, and can be lower by a dB or two. I've never measured the 'interesting stuff', you recently acquired, but it has at least on paper one of the best noise figures you can spend your body parts on. Ciao, -- FA O tu, che porte, correndo si ? E guerra e morte ! _______________________________________________ Linux-audio-user mailing list Linux-audio-user@xxxxxxxxxxxxxxxxxxxx http://lists.linuxaudio.org/listinfo/linux-audio-user
